Archos 101 XS review, Ultrabook-killing machine

Posted by Charbax – August 25, 2012

In this video, I show web browsing speed on this OMAP4470 ARM Powered Laptop/Tablet convertible. This ARM Powered Laptop loads web pages faster than the Ultrabook, both over the same WiFi home network! I load web pages like Engadget.com, Gizmodo.com and TheVerge.com faster on the $400 Archos 101 XS than on the $1000 Toshiba z830 Ultrabook! And the firmware isn't even Jelly Bean yet! It'll get the Jelly Bean firmware by the time Archos starts selling this device next month in Europe or the month after that in the USA! Everyone knows Jelly Bean and Linaro speeds up Android even more!

This Archos 101 XS likely provides one of the industry's fastest performance for productivity on Android, it's the first OMAP4470 device announced and demonstrated that I know of. Productivity is in Chrome on Android and a few other productivity apps that can be the Office suite (included for free for word/excell/powerpoint stuff) and that can also be Remote Desktop for enterprise professionals who want productivity that way using Teradici and Citrix on Android stuff. I would like Archos to integrate Ubuntu on Android also, I hope they call Canonical to get that included with the Jelly Bean firmware. Thus you'd click on the Ubuntu icon to switch to Ubuntu in a second, do whatever you want in Ubuntu including run any Ubuntu application, and then have the same icon on Ubuntu to switch back to full Android in a second too.

While we're waiting and looking forward to even faster ARM Cortex-A15 with Mali-T604, that likely doesn't reach consumer devices until next year though. Right now, the latest and best class of ARM Cortex-A9 processors, with OMAP4470, with 32nm Exynos 4412, with 28nm Qualcomm S4 Pro Quad-core, with HiSilicon K3V2, i.MX6 Quad, we're getting some excellent memory bandwidth performance on ARM allowing for fast enough full 720p/800p even 1080p web browsing speeds on Tablet, Laptop screens and on any external monitors as a Desktop/Set-top-box with full fast enough performance for productivity!

The time we've been waiting for is here! The ARM Powered Laptop is faster than an Intel Atom Netbook! It's even faster than an Intel Core i5 Ultrabook!!!

While at $299 and an unlimited amount of cash (think: French/EU Francois Hollande national IT investment project) for mass production, I think Archos can single handedly be able to sell more Android Laptops like this one than all the Intel/Microsoft Ultrabooks/Netbooks and new Windows 8 convertibles put together. At $399 introductory Manufacturers Suggested Retail Price, Archos can still get in there and sell as many as Archos can afford to manufacture, selling easily at 50% cheaper than the iPad3 with keyboard ($598), at 63% cheaper than the Asus Transformer Prime/Infinity with keyboard ($648), this is the thinnest, lightest 10.1" ARM Powered Laptop/Tablet convertible yet. And it has features other tablet makers don't have such as full hardware accelerated video and audio codecs support, MicroSD/HDMI and USB Host, Bluetooth 4.0, WiFi Direct, 1080p@60fps/3D@1080p@30fps and multimedia streaming features such as Samba/Upnp and DLNA.

I think it's great value and I'm looking forward to try to use this as my main 10.1" tablet/laptop instead of my ultrabook for the next few weeks and months. Let me know in the comments what specific features you'd like me to test on this device in my next video of it.

  • Dick

    Very interesting comparison and review. The ARM devices can replace the i386 when it got the desktop applications running on it. Now the speed of ARM has improved and give same or close to performance as i386.

  • Anjali Maiji

    Your test is fucked up from the start – you have some web sites in the cache – we can even see you loading gizmodo first in the tablet and then being surprised its faster… Come on, do a video with real fairness. Empty cache, reset cookies and remove flash form your PC… Because what happens when there is a flash component on the web page you hit is that the PC launches flash player, downloads the extra code from the web (and we know flash can be huge) then has to wait to display that flash provides geometry and then this runs…. So you are not testing the same thing on both computers sorry… But we can feel your excitement – so keep up the fun life you live being a big archos fan… That’s fun to watch even when it’s obvious your are biased :-)

    Enjoy!

  • pick a name

    yeah Charbax, maybe invest a bit more time to get basic tech details right!
    Don’t get me wrong, this blog is nice and has very good coverage

    Anyway, I completely agree that ARM is finally fast enough for good desktop experience :-)

  • Archosfan

    he got cashes on both devices, so it’s not a cheat.

  • Archosfan

    he used synchronized Chrome, so he has both devices with cache

  • Anjali Maiji

    oh come on, you know better don’t you? the cache is not sync-ed between devices… only bookmarks, passwords and preferences…

  • David Vincent

    it was an impromptu test, so chill.

  • Maventwo

    Charbax, you are so enthusiastic so you sounds like Archos Salesperson!

    But now it is very interesting what all followers of Archos will be able to make faster tablets.

  • http://www.facebook.com/people/Emanuel-Burgos/100000376144589 Emanuel Burgos

    I would still rather have a netbook. I bought a Gateway 10″ Netbook a few months ago for $250 (it is less now…. I have even seen it as low as $199 at Besbuy). The only thing that I upgraded was the RAM (from 1GB to 2GB). I can run Microsoft Office, play flash videos, run XBMC, run some emulators and other basic games, and get real work done (with my other software applications). It also has a 9-10 hours battery life. Consequently, if all you are really doing is surfing then a tablet might be good enough. However, if you seriously want to get real work done, a netbook is still the way to go. Personally, I don’t see what the big deal of these tablets are all about since a netbook can do everything a tablet can do for much cheaper, better battery life, built-in keyboard, and with more usefulness! $400 vs $230, it is a no brainer!

  • http://ARMdevices.net/ Charbax

    I believe the ARM Powered netbook can be made for half the price of an Intel Atom netbook. This $400 package from Archos has more performance than a netbook and includes a very thin tablet form factor, HDMI and all the things that come with that. Archos could sell it $299 and still make a profit, but Archos needs larger profit margins at the moment as they aren’t huge on cash reserve and Archos is the first company with this form factor anywhere near that price anyway so they can sell as many as they can make at $399 so they have to sell it at that price for now. If you have the manufacturers build a basic ARM Powered netbook, using for example the dual-core RK3066 processor with Mali-400, thus better performance than a netbook, no touch screen and running OS adapted for no touch-screen (Android thus may not be totally suitable), thus for example running Chrome OS and Ubuntu on ARM, here are the BOM savings vs the Intel Atom netbook:

    – $30 Processor and SoC savings vs Intel Atom system
    – $30 Flash storage instead of 250GB hard drive
    – $10 cheaper/smaller battery and lower power casing components
    – $10 other motherboard savings
    – $30 Linux instead of Windows 7/8 for netbook

    Thus the ARM Powered betbook costs $110 cheaper to make than the Intel Atom netbook. ARM Powered Laptop is thus $100 to make while the Intel Atom netbook costs $210 to make.

    Thus MSRP for the ARM Powered Laptop can be anywhere between $99 subsidized to $149 unsubsidized. Equivalent Intel Atom netbook has to be sold at $249 to $299.

  • Anjali Maiji

    At this stage They could probably NOT sell it for $300. They have roughly a 40% margin on the product and need 20% for their disti and at the very least 16% (if they sell a lot ) to keep the company running… So unless they can sell direct worldwide at very low operational cost and with huge volume (they don’t have this capability today) they can’t really trim the price by 25%.

    It was also interesting to see that you finally admit the company is in trouble and low in cash… Not what you were saying in recent discussions on other article on their S1 results

  • http://ARMdevices.net/ Charbax

    This has nothing with being in trouble! It’s just that Archos cannot afford to manufacture 10 million tablets overnight, so as demand for Archos 101 XS is gigantic, they may as well milk the cow as long as possible. I believe Archos can manufacture it closer to $200 than $300, that’s including all considerations of the new OMAP4470 SoC, thinner more metallic casing, new MVA 800p screen, and the new plastics/nice coverboard keyboard. The only reason it may still be around or above $200 to make and not $150 is because for now these are the first batches of manufacturing, those require a lot of tuning, a lot of adjustments, yields have to be optimized in the first batches, but by Christmas, as yields are excellent, as mass manufacturing is going fully smoothly at maximum speed that Archos can afford, you can be sure that price per device is $50 lower. With $150 manufacturing price per device and large volumes, Archos can easily aim at a new $299 MSRP with keyboard, $60 (20%) for resellers and upwards $90 per device margin (30%) for Archos to keep for themselves. Even if Archos presses quality assurance at spending $180 per device for manufacturing that still leaves Archos with a $60 20% profit margin per device.

    Regardless, more cash always can speed things up and move planning along at allowing for more aggressive pricing to be aimed for earlier, with more volume sales etc it’s always nice. But Archos does not necessarily need it. Archos can manufacture as many as they can afford to manufacture, and that is plenty enough to make decent revenues and grow if needed. While the Archos 101 XS at $399 MSRP can bring Archos with a very good upwards $120 (30%) per device profit margin, this is a mix and Archos is very aggresively pushing for a whole range of cheaper Arnova and Elements devices, for example a $199 RK3066 dual-core 9.7″ IPS device can hardly bring Archos a $60 (30%) margin per device from a $100 per device manufacturing cost, more likely less than that.

  • Archos is dead

    Sorry to Rhin your party – Archos is bleeding money and will be bankrupt in 7 months – unless they can find some money from shareholders – see you in march and we will see…

  • Charbax President!!!

    The wonderful magical world of Charbax !

    Give him a new archos, a billion euro and he solves France debt problem within a year selling cheap tablets to the world ! He could even allow frenchies to get 18 weeks of vacation a year, retire at 45 and stop paying tax. May be he can also fix Greece and the euro.

    If only I could vote Charbax for president !!! :-)

    Keep the enthusiasm mate – that’s refreshing to see so much candor and naiveness. Wish you luck!

  • http://ARMdevices.net/ Charbax

    Yup, we Humans can solve any problem, we can send Humans to the Moon, now we can send Humans to Mars, we can use technology to solve all the worlds problems, it’s just best when open source free software is used so we can use all our talents and all our resources to improve everything without having to ask American corporations for the permission first.

  • Anjali Maiji

    remind me what is this Google android agreement again that Archos had to sign… :-)

  • prn

    “It’s even faster than an Intel Core i5 Ultrabook!!!” – yeah right! My Core i5 blows this thing right out of the water. Hell, even my Acer One D270 does the sunspire test faster. I’m all for your reviews, but please keep your performance statement at least on the same playing field.

  • http://ARMdevices.net/ Charbax

    So I guess you didn’t watch my video where engadget, gizmodo and theverge load faster on my Archos 101 XS than on the Toshiba z830 Intel Core i5 ultrabook.

  • prn

    yes. i did see your video. I just said that broad sweeping statement on the Core i5 maybe isn’t entirely accurate. I did the Browsermark test at a 300% better result on a Core i5 and the Sunspider at at 50% better result on the N2600. The 101XS is a great machine, but its not the fastes in therms of performance. Yes, it does beat out the N2600 (91.532) on the Browsermark test. Personally, it think the 8″ model is more interesting, since the 10XS is in the netbook territory in physical dimensions. Hopefully there isn’t any glaring missteps on that one, such as missing HDMI, 2.5mm audio jack, no SD slot.

  • http://ARMdevices.net/ Charbax

    The benchmarks probably don’t mean much, I showed here that in actual real use case scenarios web browsing is faster on the $400 OMAP4470 Archos 101 XS than on the $1000 Core i5 Toshiba z830 Ultrabook.

    Benchmarks don’t take into account real use case scenarios, and don’t take into account software optimizations, GPU optimizations, CGPU optimizations, multi-core optimizations, etc.

  • CyberGusa

    All you showed was that a few sites loaded faster, but ignored why! The Archos wasn’t loading or running Flash, it wasn’t running any WebGL, it has no plug-ins or extensions to slow it down.

    Even then your own site took longer to load on the Archos. So even with the advantage of not having to load as much it didn’t always have a lead.

    Never mind the performance difference needed to run a desktop OS versus a mobile OS. A desktop OS can also have things running in the background that can effect performance.

    While you didn’t even verify if the sites being loaded were specifically the same version. Besides the difference between Mobile and Desktop optimizes web sites there’s also version differences between Flash and HTML5 sites.

    Some sites also resize their graphics and there is also the question of whether the browser is set to only load the lowest resolution images from a site, which is one of the things even on a HTML5 site could effect page loading speed.

    There’s also the question of whether you have better wifi on the Archos than on your laptop. Not all ultrabooks have great wifi, and your model is one that has had reported problems.

    Besides, even if everything was perfect, past a certain point system performance becomes irrelevant for web browsing because you can never load a site faster than your connection allows and the fastest connections available is still slow by system memory bandwidth standards.

    So you could be comparing to a super computer and still show comparable browser performance on a much less powerful device. Also the browser is not going to take up more resources than it needs to run and would be basically the same no matter how powerful the system.

    What matters most is graphic rendering and most modern browsers are starting to use hardware acceleration. So system performance is much less a factor, other than how much it might get in the way if system resources are low or compromised.

    Basically meaning your test doesn’t mean much other than to show the Archos 101XS doesn’t suck and may be a pretty good product but tablets aren’t used just for web browsing and you should show other usages as well.

  • http://ARMdevices.net/ Charbax

    ARM Laptops/Desktops/Tablets/Smartphones/Set-top-boxes can now run the web browser FASTER than on the x86 Intel overpriced ultrabook.

    My test demonstrates Intel is dead.

    The web browser is more than 95% of the usage people do on a $1000 Ultrabook. Ergo, the ARM Powered Laptop can now take over.

    Without a touch screen, with mass production (such as Google sponsoring the manufacturing of 10 million ARM Powered Nexus chromebooks), the ARM Powered Laptop can be sold with a 13.3″ screen for $199 running Chrome OS on ARM and/or Ubuntu on ARM, that’s 5x cheaper than the Intel Ultrabook. That’s a whole ARM Powered computer with software sold for cheaper than just the Intel Ultrabook processor alone.

    While the ARM industry was able to release ARM Powered laptops over the past 3 years, the ARM Processors then were only really optimized for smooth smartphone sizes touch screen user interfaces, it’s only since OMAP4460 and especially now with OMAP4470, Exynos4412, Qualcomm S4 Pro Quad, HiSilicon K3V2, i.MX6 Quad and a bunch more of the newest ARM Cortex-A9 all the way down to 32nm and 28nm, it’s now with those that the memory bandwidth architecture allows for full speed 720p, 800p, 1080p and 1200p web browsing, you can even run Ubuntu apps and Desktop-optimized Android apps with full smooth performance now. It’ll get even better next year with ARM Cortex-A15 and newer GPU like Mali-T604, but what we’ve got now is enough for a full Chrome on Android that performs faster than on an Ultrabook, it’s enough for a full Ubuntu fully smooth, it’s enough for a full Windows 8/RT OS, it’s likely even enough for full “desktop apps” like Adobe Photoshop/Premiere though the best way to implement high performance apps is now going to be through using the cloud, even for HD video, you upload the original camcorder files to the cloud and use 1000 grid servers to render/encode your video editing in real-time, this is how professionals want it.

  • CyberGusa

    “My test demonstrates Intel is dead”

    It’s statements like that which erodes your credibility. It’ll take a lot more than web browser performance before a ARM tablet would ever be considered a replacement for a Ultrabook.

    “The web browser is more than 95% of the usage people do on a $1000 Ultrabook.”

    According to who? No one uses a computer just for web browsing, least of all 95% of the time! You’re making stuff up again!

    If people were only going to browse then they would still be getting netbooks! People get Ultrabooks because they want to do more than just web browse!

    While there’s also Pentium and Celeron low cost laptops if they want something between a netbook and a more powerful laptop/ultrabook.

    You’re basically confusing what most people use mobile devices for with what most people use Ultrabooks for!!!

    “the ARM Powered Laptop can be sold with a 13.3″ screen for $199 running Chrome OS on ARM and/or Ubuntu on ARM”

    Another example of you trying to present your dreams as reality. Fact is that Archos tablet is going for $400 and it’s smaller than a 13.3″, which would cost them even more to make!!!

    Really, get a reality check and look at what these products are actually getting priced at!!! You can dream up all the possible price scenarios but real life costs are never ideal and even if they were you have to factor cost of doing business, running the company, distribution, etc. and not just the BOM.

    Besides, you should be reviewing the product for what it is and not what you think it represents!

    It’s no doubt things are getting better for ARM but they still got a long way to go and they need the software to back them up otherwise it won’t matter how good the performance gets.

    Windows RT, Ubuntu on Android, Chrome, and a few other options are going to definitely help ARM but it’s only the beginning and the market doesn’t change overnight.

    Even the cloud isn’t going to replace much anytime soon. It’s still going to take years to set up the kind of services you’re thinking of and years longer before they’re trusted enough to really make people want to switch.

  • http://ARMdevices.net/ Charbax

    People use the web browser and nothing else. The Macbook is a $2000 Facebook machine. That’s a simple fact. You can lookup any number of PC/Laptop usability studies, people do NOTHING other than use the web browser. When people do use a few other apps it’s media playback which works perfectly fine on ARM and it’s basic stuff like BitTorrent downloads, basic photo editing, basic video editing, basic text editing, all of which work perfectly fine on ARM, there is NO need for x86/Intel anymore. That’s just scientific fact.

    Even for powering the cloud, for powering all the web’s servers, Intel is dead.

    Archos 101 XS is one product from Archos, first on market with OMAP4470, competing with over-priced iPad3 with keyboard $598 and competing with over-priced Asus Transformer Infinity with keyboard $648.

    The ARM Powered Laptop market, when pushed by giant corporations like Google, can thus take off touch from Nexus7 and use that price saving for a larger screen and a plastics keyboard, thus the ARM Powered Ultrabook can be made and sold for $199 without a loss, as Google makes upwards $100 per year on heavy web users, the $199 ARM Powered Chromebook Ultrabook can easily be subsidized, but even without subsidy is not sold at loss for $199.
    The ARM Powered Chromebook Ultrabook runs 2x longer on a 2x smaller battery, it’s 2x thinner, 2x lighter, and thus costs 5x less than the Intel/x86 Ultrabook.

    There is no market for Intel and Microsoft anymore. Those companies are definitely companies of the past. Supporters of those companies definitely live in the past.

    Hundreds of millions of people have already moved all their emailing, document editing, chatting, video-chatting, even photo and video editing to the cloud. The number of users using cloud exclusively is increasing by hundreds of thousands every single day. Every week about 5000 new companies move all their enterprise apps to the cloud. Pretty much over 95% of all startups use the cloud exclusively, just because it saves them a lot of money.

  • CyberGusa

    “People use the web browser and nothing else.”

    No, most people don’t use the browser and nothing else. Otherwise Chromebooks would have done better than they did!

    People play games (not just casual), watch videos (not just online), they get creative, they work… They don’t just browse the web all the time!

    Really, if they did then no one would be using desktops anymore. They wouldn’t even need laptops anymore.

    Browsing is just what they mostly do on mobile devices because aside from the apps there’s not much else you can use them for!!!

    You’re a blogger, you live most of your life online but you don’t represent everyone! So don’t confuse your usage with how everyone uses their devices.

    Really, there’s no point in getting a Ultrabook if all you need to do is web browse. People get more powerful devices because they do more than just browsing and other things that don’t need high performance to run.

    If you actually look at the studies, the only ones showing high web usage include devices like the iPad in the study!

    While it’s productivity that’s why desktops and laptops are still used!

    “Even for powering the cloud, for powering all the web’s servers, Intel is dead.”

    No, because they still need high performance solutions and ARM can’t yet provide that. Don’t confuse competition with obsolescence!

    “Archos 101 XS is one product from Archos, first on market with OMAP4470,
    competing with over-priced iPad3 with keyboard $598 and competing with
    over-priced Asus Transformer Infinity with keyboard $648.”

    And both will likely still outsell the Archos because people don’t buy on just performance. Besides, Asus has a better keyboard dock solution and Apple got’s it loyal consumers and both have better distribution and public awareness/marketing than Archos.

    “The ARM Powered Laptop market, when pushed by giant corporations like
    Google, can thus take off touch from Nexus7 and use that price saving
    for a larger screen and a plastics keyboard, thus the ARM Powered
    Ultrabook can be made and sold for $199”

    Google Nexus 7 base model is $200 and they’re only about breaking even with that model. There’s no way in hell they’re going to sell an even more costly model for that price!

    We’ve debated this before but you keep on trying to push the idea like it has a leg to stand on. Stop confusing what you want with what they’re going to do!

    “The ARM Powered Chromebook Ultrabook runs 2x longer on a 2x smaller
    battery, it’s 2x thinner, 2x lighter, and thus costs 5x less than the
    Intel/x86 Ultrabook.”

    Considering they have yet to even make a ARM Chromebook means you’re just making up numbers. While ATOM bases systems won’t cost 5x more, maybe 2x but they’re already starting to get competitive.

    Then there’s the low cost Sandy Bridge based Celeron & Pentium chips I mentioned before.

    While the Core i-Series chips may be 5x and more but they also can do things that ARM based systems aren’t capable of doing yet and so still represent a separate market!

    Ultrabooks with discrete graphics especially can do thing like play most of the newer games, run full desktop Photoshop, run CAD, and lots of other things that a ARM device wouldn’t be able to run even if the software was ported over.

    You may want to argue that all you need is a web browser but that’s not why people have laptops and desktops. Especially those who need them for work!

    “There is no market for Intel and Microsoft anymore. Those companies are
    definitely companies of the past. Supporters of those companies
    definitely live in the past.”

    No, they’re the present. What the future will be remains to be discovered but it’s very unlikely things will change that much anytime soon.

    It’s fine to be enthusiastic and root for what products you want but let’s not pretend it’s more than that right now.

    “Hundreds of millions of people have already moved all their emailing,
    document editing, chatting, video-chatting, even photo and video editing
    to the cloud. ”

    And a lot of people promptly went off the cloud because it’s not yet reliable, not everyone has the bandwidth or data plans that would allow heavy usage yet, and not everyone goes for having to pay for something they could just own and run on their system instead of basically renting it.

    Cloud services are growing but the infrastructure still needs to grow. Setting up the kind of world wide resources like youtube has with their servers is not something most companies have yet.

    Companies like Apple limit Siri for example, not just because of marketing, but because their services aren’t ready to support every single iOS product on the planet. They even had outages when it first launched because of server overload and working out the bugs as they slowly scaled up.

    Such infrastructure has already started to be built, but it’ll take years before it gets to the level that it needs to and that’s not even counting how long it’ll take to convince most people to trust it.

    For now cloud services provide a alternative but it’s not yet for everyone.

    Anyway, if you really want to promote this Archos product then finish reviewing it and show every possible usage. Everything else you’re doing right now is just a distraction.

  • anjali Maiji

    fully agree. the test is just plain wrong

  • prn

    “don’t mean much” – well, the frontpage of armdevices.net is plastered with performance numbers of the 101XS. Now, we all love a little sensational blogging and eye catching headlines, but as a raw performer, this product is no “ultrabook killing machine”. Also the price point. Resently Dixons did a campaing for the Toshiba NB Netbook at just 250USD. I would say that is a better deal by far, even as a dumb webbrowser terminal. Any news on the 8″ device? If that is a good companion on PDF reading it could replace my Sony PRS600 ebook reader.

  • http://ARMdevices.net/ Charbax

    This is a brand new ultra-thin and ultra-light 1280×800 tablet/laptop convertible from the French small specialized tablet maker featuring first-in-industry ARM Processor and setups and your price-dumped because not selling well several years old technology toshiba netbook is not.

  • Archofan

    OMG, Char, u r so Archosed … “1, 2, 3, BOOM, BOOM, BOOM! It was faster here, IT WAS FASTER HERE ~lifting the tab up~, Yea, yea, it was much faster!!! We have the full website … The full website loaded FASTER!!!! ” and ” Check, it was faster, the, the Archos is the FASTEST ONE!!!… but, of course there’s no flash here …” and last one ” WHO IS FASTEST? – This is FASTER, yeah…” Oughh, don’t be so crazy with Archos …

  • SparkyRih

    My Padfone (phone/tablet/netbook) kills this archos hybrid πŸ˜›

  • http://ARMdevices.net/ Charbax

    Perhaps you forgot that your Asus Padfone cost you something like $1000 and includes an average phone, a weird average system for tablet and an even bulkier than Asus Transformer Keyboard Dock. And that your Asus Padfone doesn’t have a better processor than OMAP4470, yours doesn’t even use the latest yet unreleased Qualcomm S4 Pro processor platform.

  • Archosfan

    Char, don’t be so Archosed … Archos xs does have strongest SoC processor, I agree, but, the performance of the tablet is not the best, WHY? Not because the processor is bad, or Android is not good, just because Archos, Archos’ processoring program is not well-optimized, so it could lose some other devices, don’t base only on the processor. You see, the iPad, has only dual-core 1Ghz, but you saw the performance? Archos’ 1Ghz could be like only 1/2 of the iPad, again, why? Because Apple optimized better, than Archos,

  • Archosfan

    Char, can I ask you to compare it with ASUS Transformer Infinity (TF700T), please, test gaming performance, web-browsing speed, wifi connect speed, screen rotation speed, home/back speed, smoothness when sliding pages.

  • Jesse Lee

    I agree with Charbax however Intel won’t die that easily. I think x86 is dying slow because it doesn’t have M$ to back it up anymore. Not that Windows 8 is any good because it isn’t. Ti has decent X11 drivers from what I have heard too bad ARM Mali-400MP almost useless to me because of no X11 driver support. I am sure they will have X11 drivers in the next few months for the hardkernel odriod-x. I saw that you have one as well Charbax can you tell me your experiences with the odriod-x?

  • max

    Charbax, It’s indeed a step forward from Archos’s previous tablets, But still they need to improve on battery life.

    And personally I don’t find this type of keyboard very useful, it lacks touchpad / mousepad and it could have come with a built in battery. That way they could have sold this tablet for $299 and the keyboard dock for $99.

    Lenovo S2110 comes with DualCore S4 and got same transformer like built in battery dock, selling for $399. Today I saw at Sears Transformer TF300 with keyboard dock for $399. Point is when this X101 will reach to real retail stores in October then Asus TF300 and Lenovo S2110 will give it competition with their full keyboard mousepad built in battery dock.

    I’m not saying this is a bad device, But I’m saying Archos could’ve played smarter and realeased the device now with a full fledged mousepad built in battery type keyboard dock sold separately for $99. Then they could have easily played the competition by being early in game. Just my perspective.

    *Note for people who are saying that Charbax sounds like Archos Salesperson Archos Fanboy etc.

    I’m sure those people must have seen /read the Bloggers who are from main stream websites and do review of an Apple product, they sound like Apple is their own company and praise the Apple product to an shameless extent.
    Internet is full of those Apple fanboy bloggers, even from main stream websites. Compare to them Charbax is very fair.
    And this is his style, He always sound excited when he review a new gadget.

    Charbax You Rock. :)

  • Robert Jakiel

    Charbax I don’t know what your fascination is with Archos but this is no Ultrabook killing anything. At $400 you’re getting a dual core A9 tablet with a keyboard. For $500 if you do some shopping w/ online codes you can get an Asus TF300T w/ kybd. dock. So that extra $100 gives you a quad core Tegra3 that can easily be OC’d to 1.6Ghz and running Jellybean right from the factory. In order for Archos to be competitive they are going to have to drop the price from $400 w/ kybd to $249 – $299 w/ kybd. Otherwise they will price themselves out of a market filled with superior competition.

  • Joakim

    Why would you want to kill the ultrabook? It hasn’t sold that well, estimates were 1 million by the end of the year, well off the 90 million that Intel PR was claiming they would sell this year. I have to say though that I bought a Zenbook Prime ultrabook recently and I love it. Only thing better would be if I could install FreeBSD instead of Windows 7 and have all the hardware work, but you can’t have everything. πŸ˜‰ That said, Windows 7 hasn’t been bad: it’s the best desktop OS MS ever put out, by a long shot, which is why it’s the first Microsoft OS I ever bought or used much. I seriously considered buying a Transformer Prime instead, but the inability to run virtualization software like VMware for development work ended up killing that option for me, admittedly not something that will affect most.

    Getting back to the point, I think the fundamental disagreement here stems from the fact that there are two distinct computing segments: mobile devices, ie smartphones and tablets though not counting the more powerful Intel laptops, and the more traditional Wintel desktop, which I was surprised to read a couple months ago sells more in laptop form than desktop now. CyberGusa is wrong that Wintel is “the present,” the present belongs to mobile devices, as there were 50% more mobile devices sold last year than desktop/laptop. Since 99% of those mobile devices run ARM chips, the present belongs to ARM. :)

    However, Microsoft is betting everything on Windows (Phone) 8 and Intel is along for the ride, as most of the Windows 8 devices announced at IFA today run Intel chips, mostly Atom Clover Trail. So it will be interesting to see if Wintel can strike back in the next cycle: I think Microsoft is doomed to at least get shut out of the much more lucrative mobile device market, who knows what Intel has up their sleeves. And with Cortex A15 coming up, ARM can put a lot more pressure on those Intel servers, laptops, and even desktops. The biggest issue is that there isn’t a good desktop OS made for ARM, as Windows 8 will likely bomb and there’s nothing else out there, but there certainly won’t be a problem getting server software.

    We live in interesting tech times again- I love it- looking forward to seeing how the tech market evolves over the coming years.

  • CyberGusa

    “CyberGusa is wrong that Wintel is “the present,” the present belongs to mobile devices, as there were 50% more mobile devices sold last year than desktop/laptop. Since 99% of those mobile devices run ARM chips, the present belongs to ARM. :)”

    Nope, first fast growing market doesn’t mean any other markets are being displaced. Second, you’re ignoring the differences in the market and trying to draw direct comparisons from very different business models.

    Most people don’t buy new computers every few months or every year or so but they do for mobile devices. So much of those figures are just people replacing their previous devices and most were never intended to replace a PC. Also people with PC’s tend to still use their old one in some capacity while mobile devices get tossed or recycled. Never mind you ignore that people use both mobile and non-mobile devices and it’s not either/or for most people!

    “The biggest issue is that there isn’t a good desktop OS made for ARM, as
    Windows 8 will likely bomb and there’s nothing else out there, but
    there certainly won’t be a problem getting server software.”

    You better hope W8 succeeds because general consumers won’t care whether ARM succeeds in the server market as that won’t directly effect them! Just like most people don’t care that Linux dominates the server and embedded market as it doesn’t effect what most people use on their desktops.

    While W8 isn’t the only option, it’s just the one with the best chances right now.

    You are right though that it’s software that will be the big factor and if ARM can get the software people want to use then it will succeed. We’re a long way from seeing whether that will happen though and also a long way from knowing whether Intel/AMD won’t effectively counter them in the coming years.

    You already covered what ARM is coming out with, so I’ll just point out what you’re missing.

    For example, a recent youtube video compared the boot time of the Samsung Aviv Tab (Qualcomm Snapdragon S4 1.5GHz) with Windows RT vs the Samsung Smart PC (Clover Trail) with Windows 8. The RT took 1:48 to boot and load a web page and the W8 device took just 45 seconds.

    Mind RT doesn’t support legacy and thus it’s even lighter and easier to run than Windows 8. So the results had more to do with the hardware differences. The transition of ARM from a mobile only solution to a general PC usage solution isn’t going to be a perfectly smooth transition as the platform has to evolve and adapt to the different needs and requirements of general PC usage.

    While Clover Trail is one of the first ATOMs that offer closer to ARM costs and run time. Windows RT costing more than Windows 8 also helps blur the difference.

    Mind also Google’s support of Intel means you can run everything on x86, even Android, but have much fewer choices for ARM. Features like lock down are also being emphasized by content providers who don’t trust ARM security. This is largely perception but gives x86 the advantage of tending to be more flexible for end users.

    While next gen ATOM coming out next year will bring a architectural update (Out Of Order Processing, etc), an additional FAB shrink to 22nm, adoption of a scaled down version of the GMA based on the Intel HD4000, support for up to 8GB of RAM, all ATOMs will become SoCs, among other changes.

    Intel will then advance the ATOM every year at faster than Moore’s Law. So the above is for 2013 22nm Silvermond, then 14nm for 2014 Airmont, 10mm for 2015, then 7mm and then 5mm is planned for before then end of the decade.

    So while ARM is evolving rapidly and are going to get better, it’s too early to say Intel is out of the game yet. They just had a very slow start but that doesn’t rule them out if they manage to catch up.

    Similarly, AMD is starting to advance though they may go a hybrid route to help catch up with Intel. Incorporating ARM Trust Zone technology into their APU’s for example. They’re still going x86 but that doesn’t exclude using other technology as well.

    While as already pointed out the software side of things is still the big question. So nothing can be for certain right now but we can agree that it’s interesting times and it will be interesting to see how it all evolves over the coming years.

  • Joakim

    CyberGusa, I didn’t say that the “other markets are being displaced,” but that is what is happening and will only increase. Whatever the reason for why people buy more mobile devices, all that matters is that there’s a lot more revenue in mobile devices, which means more money for the device makers to reinvest back into product development. Apple now has $117 billion in cash, driven by mobile devices such as the iPhone and iPad, double the $60 billion that Microsoft has. If the two get in a spending war, we know who will win. If the upcoming iPad mini is priced at $200-250 as suspected, that’s an example of a pricing war that only Apple can win now.
    You’re right that many use both a desktop/laptop and a mobile device right now, ie they have different uses, but there is also a significant minority that uses only a mobile device, as that’s all they need. That minority is only going to grow with time, as you only need a mobile device for most of what most people do, ie web browsing and email. Obviously everyone isn’t just throwing out their PCs, but the fact that the mobile device market is currently much bigger certainly marks it as the present regardless.

    ARM in the server market will affect consumers through lower prices, just as linux does the same in the server and embedded markets. I agree that Win8 is the only hope for ARM on laptops/desktops, at least right now. I did note that Wintel might strike back in the next cycle, so it’s not like I said they’re out. I see no reason why the Windows RT version of the Ativ Tab can’t be optimized to be faster than the Intel version, so I’m not sure that particular early benchmark matters. You say that “the platform has to evolve and adapt” for PCs, but that’s only a software issue: I don’t see any hardware problems holding the transition back.
    Why do you say Windows RT will cost more than regular Windows 8? It’s generally accepted that Windows RT devices will cost $200-300 less than the Windows 8 Intel versions. Also, Clover Trail probably doesn’t have the same “run time,” it seems to use about 2W while ARM chips usually run at a fraction of that in normal use. Actually, Google doesn’t support Intel that well and Android releases for x86 usually lag behind ARM releases, for good reason as almost nobody runs Android on x86. :) All the coming Atom updates you list won’t matter if they can’t get power consumption down, which they haven’t been able to do for many years now. ARM doesn’t need all those changes as long as it’s “good enough” and wins the power battle, as it has been doing for years. I suspect Intel just cannot get x86 to ever match ARM in power consumption, let’s see. I hope AMD can come back but I see nothing to indicate they will, at least on mobile devices.
    The front line right now is laptops and the other “hybrid” laptop devices that were shown off at IFA this week. Wintel has lost the smartphone battle for the foreseeable future: I rate the chances of Windows Phone 8 and Medfield getting any market share as practically nil. It will be interesting to see how much market share WinRT gets on laptops. They could do really well because of better power usage, but Microsoft not allowing Windows desktop apps to be simply recompiled for WinRT could be a big blow. If those desktop apps don’t end up mattering and most laptops go ARM, it will be game over for Intel on the consumer side, as they will be down from their current 40% share of consumer computing devices (everything but feature phones) to the 10-15% of devices that are desktops.
    And once Cortex A15 comes out, it could be game over for Intel, as ARM will then assault the last two strongholds of desktops and servers. I’m not saying this is what will happen for sure- like I said, I don’t know what Intel might have up their sleeve that hasn’t been revealed yet- but it looks increasingly like the most likely outcome. Microsoft can hang around for a bit longer because they don’t have any real competition on the desktop, unless Apple decides to go after the mass market on the desktop and starts licensing Mac OS X, unlikely but possible.

  • CyberGusa

    “I didn’t say that the “other markets are being displaced,” but that is what is happening and will only increase.”

    No, that’s is what’s meant when stating one type of product is replacing another but as I already pointed out that’s a erroneous assumption on your part.

    People aren’t stopping buying PC’s and it doesn’t really matter if they’re buying more mobile devices if most are just being replaced constantly and not being used for all the same things.

    While Apple never really had a large share of the PC market. So of course they make more with their mobile products that do dominate the markets, but it took a majority share of the mobile market for them to make more than they’re making with on average less than 10% of the PC market… No real surprise which they make more with!

    “If the upcoming iPad mini is priced at $200-250 as suspected, that’s an example of a pricing war that only Apple can win now.”

    Yes, Apple has more money but even they have to play it smart as it’s who can last long term that matters and money can be used up quickly if wasted. Basically, the biggest spenders don’t always win and Apple knows this! It’s why they go to so much effort to ensure they can get the lowest prices for making their products.

    Besides, you would have to ignore the differences in products. The mini iPad isn’t likely to be pushing any of the limits of its intended product range and Apple is always mindful of their other products. So can’t short change their iTouch and other iPads.

    While the alternatives offer things Apple doesn’t, and that also means the alternatives don’t necessarily directly compete.

    ” That minority is only going to grow with time, as you only need a
    mobile device for most of what most people do, ie web browsing and
    email.”

    This belies that people need to work and mobile devices are a compromise. They’re fine for consumption but when it comes to productivity they will be always less than ideal.

    What will far more likely happen is mobile devices will supplement and expand what can be done with our other devices.

    ” I see no reason why the Windows RT version of the Ativ Tab can’t be optimized to be faster than the Intel version.”

    Architecturally, ARM still has some things it needs to get better at like memory bandwidth… ARM was designed from the start to serve optimally for mobile devices but the things that are strengths for mobile design are sometimes weaknesses when trying to go into full PC usage.

    For now ARM is still a 32bit processor, while Intel has been 64bit for years now. So Intel has more experience and already know how to make powerful processors.

    It is ARM’s advantage that Intel has to learn the opposite and get better and making low cost and lower powered solutions but it’s arguably easier to scale down than it is to scale up.

    ARM is rapidly developing 64bit, at least one product already has 64bit memory bus and they’re demonstrating some 64bit for the server market but it takes years to transition from 32bit to 64bit. Even if ARM manages to do it faster than it happened for x86, the nature of the product ecosystem means it’ll still be years before we see 64bit outside of the server market and in the consumer market.

    Mind that present Intel ATOMs are still using the basically same architecture as when the ATOM was first introduced 5 years ago. The Clover Trail is just a tweaked version that utilizes many of the same tweaks and modifications they employed for Medfield. So it’s not like it has something that would give it an unfair advantage that wasn’t there for years.

    While as I already pointed out Windows RT is smaller and easier to run than Windows 8. So the results show that even with an advantage of having less to boot that the RT device still took a minute longer to boot.

    Mind the difference can be even larger for many ARM devices because most don’t use SSDs but rather more basic Flash Memory like used in USB drives and Memory Cards. Though, if the RT devices was using a inferior drive then that may have partly explained the boot time difference but otherwise it means there’s a significant difference in efficiency.

    “Why do you say Windows RT will cost more than regular Windows 8?”

    I’m talking about the software, RT will cost more to system makers because they’re including a version of MS Office Home and Student 2013 RT. Normally about a $99 value if you had to get it separately. OEM’s get the OS for a lot less than we pay retail. So we’re talking about less than $100, rumor is RT is between $80-90 but Windows 8 is rumored to be $40-50.

    Meaning a $30-50 difference, which is about the difference in ARM and ATOM costs. At least for the SoC part, and Intel is only going to get more aggressive next year when they go 22nm.

    So that one reason why they’ve started offering digitizer pens with ATOM systems but not ARM. People expect ATOM systems to cost more than ARM. So the margins allows them to throw a little something extra and still reach an acceptable price.

    While they’re pressured to keep the ARM device costs as low as possible…

    System makers are also used to dealing with small margins for ATOM devices but have yet to really push margins down on ARM as it has always been easier to just reduce costs first.

    “Also, Clover Trail probably doesn’t have the same “run time,” it seems
    to use about 2W while ARM chips usually run at a fraction of that in
    normal use.”

    No, as ARM increases to push for higher end performance they’re rising in power usage. The iPad 3 especially uses more than the average ARM device and is why they had to increase the battery capacity about 70% from the iPad 2 to get about the same run time. It even uses so much power while being used that’s it’s hard to charge unless you stop using it or shut it down and that’s with a 2W USB charger.

    Mind, besides the SoC most components like the LCD use the same power regardless of what’s running the system. While Clover Trail is a optimized SoC like Medfield, which employs many of the same power saving tricks that ARM uses.

    Besides, for both ARM and ATOM the 2W is a max indicator and TDP refers specifically to energy needed to cool the system. So most of the time they will be using a lot less power, which is the key difference between these new ATOMs and previous ATOMs is Intel has finally managed to get the idling power states really low.

    Though the newer ARM chips, based on Cortex A15 or similar still do stand to do a little noticeably better on run time but the new ATOMs should be close enough for the difference to be negligible.

    The improved power efficiency is for example why they’re finally switching to Out Of Order Processing next year with the 22nm update. They’ve been using In Order Processing to avoid the power usage increase Out Of Order imposes but now they’re getting efficient enough to allow it and still keep power usage low.

    “Actually, Google doesn’t support Intel that well and Android releases
    for x86 usually lag behind ARM releases, for good reason as almost
    nobody runs Android on x86.”

    No, this is recent and separate from the previous Android x86 project. Since Intel has only started offering products like the Medfield that could actually be used in similar mobile devices as ARM. Though you are right that support still lags a bit behind but it’s in the 90% range and will likely improve as the number of Intel based products grows.

    While services like Bluestack offer alternative ways to get Android apps working on x86.

    “Wintel has lost the smartphone battle for the foreseeable future: I rate
    the chances of Windows Phone 8 and Medfield getting any market share as
    practically nil. ”

    Chances are they’re start small and it will likely take years to get any significant chunk of the market.

    Also it’ll be at least two more years before Intel can start competing with the high end ARM phones, but Intel is advancing pretty quick. Compared say to how Nvidia did when they first entered the mobile market and it wasn’t until their 3rd gen product that they were considered starting to getting ahead.

    “It will be interesting to see how much market share WinRT gets on laptops.”

    This is a big question as it not only pits ARM vs x86 but also a more basic and locked down version of Windows vs the more flexible and capable version on x86.

    So should put to the test the claim people no longer really need much anymore.

    “And once Cortex A15 comes out, it could be game over for Intel, as ARM
    will then assault the last two strongholds of desktops and servers.”

    No, Cortex A15 will only increase ARM’s competitiveness but they need to go 64bit before they can seriously threaten x86 in the server market.

    For now ARM will be good for application that don’t need high performance and/or can get away with parallel computing. Everything else will still use x86…

    “I don’t know what Intel might have up their sleeve that hasn’t been
    revealed yet- but it looks increasingly like the most likely outcome.”

    Only if you don’t look at what ARM still has to overcome on their side. It’s not going to be easy for either.

    “unless Apple decides to go after the mass market on the desktop and starts licensing Mac OS X, unlikely but possible.”

    I think they will but they’ll wait till ARM gets a bit more powerful and after they evolve OSX to be more usable on mobile devices. So eventually they’ll replace iOS on devices like the iPad. Though will likely keep iOS for lower end devices.

  • Joakim

    Actually, people are “stopping buying PCs,” let’s look at the numbers. Since smartphone sales broke 100 million in 2007, PC sales have only gone up 33%, with several years of flat sales. Of course, there was a worldwide recession during the same timeframe, but if we compare 2007-2011 to the comparable period of 1999-2003, PC sales were up 50% during the last recession, albeit from a lower base. What’s the difference, ie why are sales weaker this time? Likely people buying smartphones and ditching the PC, a trend which will only accelerate. I don’t know why you think it’s a negative that “mobile devices… are just being replaced constantly.” As I pointed out, that’s a benefit for mobile device makers like Apple and Samsung because that gets them a lot more profits that they can then reinvest in product development, making their new products much better than PCs. But you’re right, they’re “not being used for all the same things” yet, but some are and it’s only a matter of time till more do. The biggest bottleneck for ARM on desktops right now is that there’s no OS that works well as a touch interface for mobile use and with point-and-click for desktop use. It could have been Windows 8, if only MS hadn’t decided to force that damned Metro launcher on desktop users.

    “While Apple never really had a large share of the PC market. So of course they make more with their mobile products that do dominate the markets, but it took a majority share of the mobile market for them to make more than they’re making with on average less than 10% of the PC market… No real surprise which they make more with!”
    Um, they make much more money with 30% smartphone share and 70% tablet share than Microsoft makes with 90% desktop share, so much cash that Apple has twice as much cash on hand as Microsoft! That’s why mobile devices matter much more now, there’s just so much money there. I agree that Apple doesn’t seem ready to start a spending war anytime soon- although the iPad mini could be the beginning, let’s see how they price it- but they have so much cash and Google probably loses money on Android, so Android is basically dead in the medium-term. Microsoft could suffer the same fate if Apple ever decides to go mass market on desktops. Apple could even go after google in search, by spending billions on making Siri almost as good at mobile search as Google is with desktop search. All that cash gives them lots of options.

    The big question is the one you note: will Apple be willing to potentially cannibalize their own high-end products to go mass market? I think they could differentiate the products enough to do it, say offer a budget laptop running Lion and a premium one with Mountain Lion, just like they now sell the iPad 2 at a pretty compelling price after the iPad 3 came out.

    “This belies that people need to work and mobile devices are a compromise. They’re fine for consumption but when it comes to productivity they will be always less than ideal.
    What will far more likely happen is mobile devices will supplement and expand what can be done with our other devices.”
    The supplementing part has already happened, which is why there are so many more mobile devices sold. But there are many people for whom a desktop was overkill anyway and a mobile device suits them perfectly as their only computing device. Most people don’t “need to work” on their computer, that’s only a small minority who were overemphasized over the last couple decades, for obvious reasons. There will always be a niche for a more powerful computer at home, but I think that computer will increasingly be a home media server attached to their big-screen TV, not a desktop PC on a desk, ie the power of the desktop PC will be put into the set-top box to become the sole powerful computer in the home. Unfortunately, nobody is pushing this trend yet, so it’s almost nonexistent right now and is not driving PC sales at all.

    “Architecturally, ARM still has some things it needs to get better at like memory bandwidth… ARM was designed from the start to serve optimally for mobile devices but the things that are strengths for mobile design are sometimes weaknesses when trying to go into full PC usage.
    For now ARM is still a 32bit processor, while Intel has been 64bit for years now. So Intel has more experience and already know how to make powerful processors.
    It is ARM’s advantage that Intel has to learn the opposite and get better and making low cost and lower powered solutions but it’s arguably easier to scale down than it is to scale up.”
    None of these are reasons for slow boot-up time. The most likely reason is that the software is not optimized enough on ARM and that the ARM processors aren’t as powerful, though I’m not sure that’s a bottleneck for boot time. 64-bit is irrelevant, as almost nobody needs more than 4 GBs of RAM. On servers, products like this will do fine for most uses, while driving power costs down. ARM is perfect for microservers, which are the next big server segment. If it were “easier to scale down,” Intel would have some market share on smartphones/tablets right now, instead of practically 0%! πŸ˜‰ Why do you think “Intel ATOMs are still using the basically same architecture as when the ATOM was first introduced 5 years ago?” Didn’t Intel see the huge volumes ARM pushed in smartphones during that timeframe? You make it sound like Intel just didn’t care, when it’s more likely they just weren’t able to get x86 down in power consumption.

    “While as I already pointed out Windows RT is smaller and easier to run than Windows 8. So the results show that even with an advantage of having less to boot that the RT device still took a minute longer to boot.”
    Both Windows 8 and Windows RT don’t load the desktop environment by default, so they are booting the same amount of software and Windows RT does not have “less to boot.” As for price, the numbers I see show Windows RT as $10-30 cheaper than Windows 8. In any case, if Windows RT devices are going to be $200-300 cheaper, as most seem to think, it’s not just the SoC or OS price driving that, it’s got to be all the other parts also. For example, the ARM devices will have smaller, cheaper batteries for similar runtime, because they draw less power. :) Not sure it’s a good thing for Intel if people “expect ATOM systems to cost more than ARM,” that could just mean a lot more sales of ARM devices, as has happened so far. It’s not like you can’t make huge margins on ARM devices also, Apple has already proven that with their much larger margins than even Microsoft. πŸ˜‰

    The iPad 3 uses a lot more power because its display has four times the pixels of the iPad 2, it has little to do with the A5X ARM chip versus the A5. Clover Trail has a 3-4 W TDP, 2W was the average load power usage. You claim that power usage on “the new ATOMs should be close enough for the difference to be negligible,” I suspect not. Let’s see what the upcoming power usage benchmarks of Windows 8 vs Windows RT devices show. :) I wouldn’t say it took Nvidia till 3rd gen to do well, Tegra 2 did very well also and showed up in many devices. But ARM had already done well by then so their job was easier, while Intel still has no market share on mobile devices.

    “‘It will be interesting to see how much market share WinRT gets on laptops.’
    This is a big question as it not only pits ARM vs x86 but also a more basic and locked down version of Windows vs the more flexible and capable version on x86.
    So should put to the test the claim people no longer really need much anymore.”
    I completely agree, this is the acid test for ARM’s dominance, with the caveat that MS is intentionally hamstringing Windows RT by not allowing desktop apps. We’ll see how big a deal that turns out to be.

    “‘unless Apple decides to go after the mass market on the desktop and starts licensing Mac OS X, unlikely but possible.’
    I think they will but they’ll wait till ARM gets a bit more powerful and after they evolve OSX to be more usable on mobile devices. So eventually they’ll replace iOS on devices like the iPad. Though will likely keep iOS for lower end devices.”

    I was referring to Apple starting to license Mac OS X to third parties like Acer or Asus to put on desktops/laptops, whereas you seem to be talking about putting full Mac OS X on mobile devices eventually, two different issues. I don’t think Apple is smart enough to go this route, but they certainly have the cash to try it.

  • CyberGusa

    “Actually, people are “stopping buying PCs,” let’s look at the numbers. Since smartphone sales broke 100 million in 2007, PC sales have only gone up 33%, with several years of flat sales.”

    You’re contradicting yourself, if the sales have gone up. No matter how slowly, then no one is stopping buying PC’s.

    Really, one market growing more rapidly than another is all but irrelevant to each other unless they actually overlap and that’s only starting to happen. So everything before doesn’t apply to what’s happening now!

    So the difference is as I already stated, there’s little to no innovation or advancement for the last couple of years. While the PC is already a mature market, there was already declining growth before the mobile market really took off. Such is the nature of mature markets!

    Just like what happened to netbooks, which during the same period also grew rapidly but just like other laptops the lack of innovation and advancement ultimately made them reach market saturation.

    Now Netbooks are still being sold, despite reports to the contrary, but are no longer a rapidly growing market.

    Market runs on cycles, whether the cycle goes up or down depends on a variety of factors but rates of growth are only rapid when a market is being established and establishing new territory. The moment that ends then growth slows considerably.

    This, however, is not the same as dying. Laptop sales have been mature for a very long time but still show continued small growth. While even when netbooks were popular they didn’t really effect laptop sales.

    Without any new innovation or advancement then of course a product line will then die out but that’s yet to happen to the PC market because they’re still trying! Only if what they’re doing now fails would it be valid to say they are on the way out but not till then!

    Really, people have been hyping the end of PC’s for ever 20 years. The one thing that holds true is technology evolves over time. What we called a desktop in the 80’s is not what we call a desktop now for example. It’s strange but people tend to think of technology statically as if it never changes and products can only be of one particular form and never change when that has never been the case.

    PC’s are always evolving, mobile market is influencing those changes right now but not in the way you’re thinking as it’s heading towards more convergence with a emphasis on symbiosis rather than survival of the fittest.

    The only possible exception is mobile devices do stand to possible replace netbooks or netbooks will evolve into mobile devices.

    “The biggest bottleneck for ARM on desktops right now is that there’s no OS that works well as a touch interface for mobile use and
    with point-and-click for desktop use. It could have been Windows 8, if
    only MS hadn’t decided to force that damned Metro launcher on desktop
    users.”

    This is another matter of perception, desktop users have actually been steadily lowering their usage of the Start Menu ever since Vista was released. Lots of people already use their desktop like a start screen or create their own custom menus in the taskbar.

    The Windows 7 Start Menu only allowed up to 10 pinned apps!

    The growing number of 3rd party options like Stardocks, virtual desktops, etc. are all indications of this trend.

    While whether Windows 8 succeeds or fails remains to be seen. Contrary to the negative rants you may have seen, there’s just as many people claiming they like it!

    So this is much like when MS introduced ribbons, people who don’t like it may refuse to understand but there are those who have the opposite opinion.

    I don’t know if you remember but people made pretty much the same complaints when GUI and the mouse were first introduced, yet what do we use today?

    Mind, people are stubborn and resist change. Sometimes it takes a bit of force to get them to even try change.

    Besides, the complaints are mostly nonsense as Windows 8 (like all previous versions) will be getting custom themes and 3rd party utilities so people can make it work anyway they want.

    While many of the complaint ignore the rest of the OS and focus to the exclusion of everything else on just the UI changes, but a OS is more than just its UI.

    Really, there are people who make Windows 7 look like XP and vice versa, along with looking like OSX, etc. There will always be those who don’t like the default but aside from choosing whether to use modern UI apps they will eventually have plenty of choices.

    Even some of the people who go out of their way to complain know this but do so anyway because they like to vent and bluster but eventually everyone adapts.

    While a good OS for ARM is something it needs, but it’s not the only thing it needs as already discussed about ARM hardware. I’ll just also add marketing because people tend to continue to buy what they’re used to and that’s not easy to change. So market momentum is another factor both sides have to deal with going forward.

    “Um, they make much more money with 30% smartphone share and 70% tablet share than Microsoft makes with 90% desktop share”

    MS primarily only makes profit from the software and so you’re ignoring what their partners make in hardware sales. While Apple does both themselves!

    While you’re also ignoring the examples I gave of companies that did not do well with the mobile market. In fact, aside from phones, Apple is the only one really doing well in the mobile market and it’s a good possibility that if anyone else was to do really well that it would take away market from Apple.

    Never mind the point on emerging markets that pretty much exclude traditional PC’s and thus adds the distortion of what people can get from what they really want to the numbers as well.

    “Both Windows 8 and Windows RT don’t load the desktop environment by
    default, so they are booting the same amount of software and Windows RT
    does not have “less to boot.””

    While the difference may not be that large it doesn’t matter that they don’t boot into the desktop environment. They’re still booting the OS and that includes everything needed to run the desktop. For which RT only needs the few elements needed to run the few MS apps they include with RT and not the full win32.exe, etc.

    Even without booting directly into the desktop mode, the service control manager starts services, and any Group Policy scripts are run on boot regardless.

    Also essential drivers required to start the Windows kernel are loaded, as well as loading into memory the system registry hive (which won’t have to refer to all the stuff that would be listed for full W8 but not RT)
    and additional drivers that are marked as BOOT_START, and of course the Windows Kernel has to run when the OS boots regardless of what environment you’re booting into!

    Booting to the Start Screen isn’t like booting to the log-in menu. Everything still loads in the background. Clicking the Desktop only runs the final graphics and runs explore.exe and is why you can get into the desktop fast instead of waiting for it to continue to boot like you would have after logging into Windows.

    Mind that Windows 8 not only supports Modern UI and legacy Desktop but also Modern/Desktop (at least for browsers). While RT only mainly supports Modern with a few included desktop that basically only counts as a partial desktop and has no overlap.

    So RT simply requires less to run than full Windows 8! Not because of the desktop per say but what the OS needs to load to support what runs afterward.

    MS basically created three sizes for the new OS. WP8 for the most basic that’ll be used on only phones. Then RT for ARM tablets as a in between option and finally full Windows 8 for all the rest.

    So the only question is whether the difference would have a big effect or not. It’s likely not too big of a effect but a full minute boot difference suggest more than just a lack of optimization.

    Even the 45 seconds for the ATOM system is kinda slow for Windows 8. Since with a fast SSD Windows 8 can boot in as little as 7 seconds. So it’s not like the ATOM system offering optimal performance either. So it remains to account for what the boot time is a full minute different.

    Though it also depends on whether it remains consistent for other systems. It’s just if it is the case then it could be indicative of some issues ARM has to overcome before continuing to push into the PC market.

    “For example, the ARM devices will have smaller, cheaper batteries for similar runtime, because they draw less power.”

    Again, that only applies to the high end systems. Newer ATOM based devices will be competing with ARM on its own cost and power usage turf.

    The upcoming Clover Trail have all the same enhancements they put into Medfield, which is already going into Smart Phones. While ARM is still more efficient, they still need to use more cores to match or exceed ATOM processors.

    So if a single core Medfield can hold its own against dual core cortex A9’s then the dual core Clover Trail stands to do even better.

    Medfield itself is getting a dual core Z2580 early next year and its GMA will reportedly be about a match for the iPad 3’s and that’s before Intel pushes it’s next gen 22nm architecture that will be out by the second half of next year. So Intel is already ramping up to compete with ARM’s upcoming next gen offerings.

    “Not sure it’s a good thing for Intel if people “expect ATOM systems to
    cost more than ARM,” that could just mean a lot more sales of ARM
    devices, as has happened so far.”

    Depends on whether the premium features, legacy support, etc can more than make up the difference in the eyes of the consumers. Especially if the price difference isn’t huge.

    Mind people like as much freedom as they can get. It’s why Windows is used more than OSX and is why Android is so popular despite it’s faults. So that’s a factor in Intel’s favor.

    Since with x86 you can run any OS, any software, etc. but on ARM the choices are more limited, you’re more likely to be locked down, have support that only lasts less than 2 years on average, etc.

    So it’s not just a matter of pricing…

    “The iPad 3 uses a lot more power because its display has four times the
    pixels of the iPad 2, it has little to do with the A5X ARM chip versus
    the A5.”

    Sorry but I specifically gave you the power usage increased caused by the screen itself, which was only 30% and doesn’t account for the rest. The iPad 3 power usage has already been analyzed!

    Fact is they doubled the number of GPU’s to support the screen and pushed graphics to game console level. They even had to increase the RAM a bit to support it all.

    So the increased power usage is not just the increased in pixels, never mind the main reason for the 30% increase was because of the back light needed to be more intense to get through those denser pixels. The actual pixels themselves don’t consume much more power.

    Really, try doubling the number of GPU’s in your system and not see a big spike in power usage. While increasing resolution means more work for the system. So of course it’s going to be using more power!

    ” Clover Trail has a 3-4 W TDP, 2W was the average load power usage.”

    No, Clover Trail TDP hasn’t been officially released. So wherever you read that isn’t official but also very unlikely as they’re replacing Oak Trail with Clover Trail and it’s specifically meant for tablets. So unlikely they will go over 2W, as any more would require a bulkier casing design to deal with the heat, though figures thrown around sometimes include the additional TDP for the rest of the system, which are usually already figured into system designs.

    Like the N2600 Cedar Trail has a low max TDP of 3.5W but counting the whole system it goes to 5W TDP.

    Well, at least we agree on the “acid test for ARM’s dominance” part…

  • http://ARMdevices.net/ Charbax

    The Wintel PC is dead. Samsung makes no money on the Wintel segment anymore, Samsung makes 95% of their profits on the ARM Powered devices, mostly on high-margin smartphones. The Wintel market is going to be even more dead as ARM Powered devices can fully replace Wintel machines for productive desktop/laptop use cases, which is what is happening right now with the latest high performance high memory bandwidth ARM Cortex-A9 like OMAP4470, Qualcomm S4 Pro, Exynos4412, HiSilicon K3V2, i.MX6 Quad, those have memory bandwidth and performance that exceed Intel Atom and thus offer fully satisfactory performance for productivity using MHL, using the new Keyboard Docks on Tablets, using HDMI Sticks and set-top-boxes and more Desktop/Laptop form factors to imminently come. The PC market isn’t going away but it’s obvious that Intel and Microsoft are dead.

  • CyberGusa

    Samsung isn’t indicative of the whole PC market and never were!

    “The Wintel market is going to be even more dead as ARM Powered devices
    can fully replace Wintel machines for productive desktop/laptop use
    cases”

    Again, this remains to be seen. Your enthusiasm aside a lot of people still need to be convinced of this and without a good desktop OS it’s going to be very hard to make that a convincing argument.

    So you better hope Windows RT does amazing well as it’ll be awhile before anything else could be as good a catalyst for ARM for the general PC consumer market.

  • juliusaugustus

    Wintel is not dead both microsoft and Intel will be around for the next century hate to break that for you. Millions of people buy pc’s every years and millions more will continue to. A ARM processor with the performance of a CORE I series processor has yet to come. But if there was profit in PC making then people would stop doing it.

  • Joakim

    “You’re contradicting yourself, if the sales have gone up. No matter how slowly, then no one is stopping buying PC’s.”

    Nope, I was very precise in my point, that PC sales growth is down in this recession compared to the last recession and that’s likely because some market segment has stopped buying PCs and is only buying mobile devices. I see, so the reason growth is now down is because “the PC is already a mature market” which “was already declining growth before the mobile market really took off?” Where do you see “declining growth” in the unit sales I linked to? From 2003-2007, right before smartphone sales took off, PC sales grew 60%. In the previous recession from 1999-2003, it grew 50%. This time it grew 33%, with two years of flat sales, 2008-2009 and 2010-2011. It could be because it’s a “mature market,” as you say, and there could only ever be demand for 300-400 million PCs worldwide this year even if smartphones and tablets didn’t exist, but given the explosive growth and sales of smartphones and now tablets, it’s more likely that these buyers are people who would have bought a PC and now don’t want one or are ditching their PCs. Netbooks were overhyped to begin with, an attempt to push a new form factor with a ton of marketing, just like we’re seeing with ultrabooks now.
    “Market runs on cycles, whether the cycle goes up or down depends on a variety of factors but rates of growth are only rapid when a market is being established and establishing new territory. The moment that ends then growth slows considerably.”
    Given that this cycle is worse than the last cycle and people are still buying tons of smartphones and tablets, one of those “factors” is that people are ditching PCs for smartphones and tablets. The PC market was growing pretty fast from 2003-2007, so it wasn’t slowing down until smartphones hit the scene.
    “This, however, is not the same as dying. Laptop sales have been mature for a very long time but still show continued small growth. While even when netbooks were popular they didn’t really effect laptop sales.”
    Who said it was “dying?” I said Wintel is dying and that the desktop PC would eventually be replaced by a home media server, nobody said desktop/laptop sales were “dying.”
    “Really, people have been hyping the end of PC’s for ever 20 years. The one thing that holds true is technology evolves over time. What we called a desktop in the 80’s is not what we call a desktop now for example. It’s strange but people tend to think of technology statically as if it never changes and products can only be of one particular form and never change when that has never been the case.”
    You’re just playing with words at this point. What changed from “a desktop in the 80’s” to “what we call a desktop now?” That’s just a form factor, one that has basically stayed the same since then.
    “PC’s are always evolving, mobile market is influencing those changes right now but not in the way you’re thinking as it’s heading towards more convergence with a emphasis on symbiosis rather than survival of the fittest.”
    More meaningless words, convergence how? What “symbiosis rather than survival of the fittest?” The points I made were that laptops were likely to go all ARM soon and in the medium-term the desktop PC is likely to die out, which has already happened to some extent with how much laptops dominate Wintel sales. In 2000, laptops were 25% of PC sales, now desktops are 30% of PC sales. Just as the mobile laptop replaced the desktop, new mobile devices like the tablet and smartphone are about to do the same.
    “This is another matter of perception, desktop users have actually been steadily lowering their usage of the Start Menu ever since Vista was released. Lots of people already use their desktop like a start screen or create their own custom menus in the taskbar.”
    Those people were putting shortcuts on their desktop since Windows 95, Vista didn’t change anything. You may be right that the Metro launcher simply replaces the desktop littered with shortcuts, for the people who do that. The fundamental problem is that the Metro environment forces a touch environment on desktop users, who mostly don’t have a touch screen. That is jarring and will not end well.
    “The Windows 7 Start Menu only allowed up to 10 pinned apps!”
    I’m using it right now, 10 is plenty for me. :p But then the problems won’t come from techie users like me or even likely those who put tons of shortcuts on their desktop, but from the majority who just got used to the Start Menu and don’t want to change to something else.
    “The growing number of 3rd party options like Stardocks, virtual desktops, etc. are all indications of this trend.”
    Penetration of these is practically nil. It’s not the more tech-adept, who used all those options, who will hate the Metro launcher, it’s all the people who never customized anything and just got used to the Start Menu after 17 years.
    “While whether Windows 8 succeeds or fails remains to be seen. Contrary to the negative rants you may have seen, there’s just as many people claiming they like it!”
    Windows 8 will be a giant failure on the desktop, worse than Vista, which after all didn’t make such a drastic UI change. On touch devices like tablets and hybrid laptops, it will do better, since that’s what it was designed for, but will not even beat Android in market share. Please submit your own predictions.
    “I don’t know if you remember but people made pretty much the same complaints when GUI and the mouse were first introduced, yet what do we use today?”
    GUI and the mouse brought a much bigger crowd, who didn’t want to work on a text terminal. Touch will not do the same, especially for the work-oriented apps that MS has traditionally focused on. Touch-focused mobile devices are about information and media consumption and Windows is already a gigantic failure in that regard, as nobody uses Windows for mobile and MS has no credible equivalent for iTunes.
    “Mind, people are stubborn and resist change. Sometimes it takes a bit of force to get them to even try change.”
    Or they reject your product, just like they rejected Vista. πŸ˜€
    “Besides, the complaints are mostly nonsense as Windows 8 (like all previous versions) will be getting custom themes and 3rd party utilities so people can make it work anyway they want.”
    Wrong, you cannot boot directly to the desktop and you cannot bring the Start Menu back, Microsoft is not allowing either. Samsung is trying to create a fake Start Menu “widget” of their own, it remains to be seen if that will be allowed by MS.
    “While many of the complaint ignore the rest of the OS and focus to the exclusion of everything else on just the UI changes, but a OS is more than just its UI.”
    The rest of Win 8 is actually pretty good, faster than Windows 7 and I like the other desktop UI changes, like the progress meter and the more cartoonish windows. But you cannot make such a drastic UI change like the Metro launcher and expect most users not to care, that’s just a fact.
    “While a good OS for ARM is something it needs, but it’s not the only thing it needs as already discussed about ARM hardware. I’ll just also add marketing because people tend to continue to buy what they’re used to and that’s not easy to change. So market momentum is another factor both sides have to deal with going forward.”
    As I said, none of your ARM hardware complaints are substantial, all that matters is software at this point. If people were still buying “what they’re used to,” Windows wouldn’t have gotten shut out of the explosive smartphone and tablet markets. After all, Windows Mobile was around long before the iPhone. πŸ˜‰
    “MS primarily only makes profit from the software and so you’re ignoring what their partners make in hardware sales. While Apple does both themselves!”
    Doesn’t matter, most of the profit is in software, not hardware, which is why Microsoft’s margins are much better than any of the hardware OEMs.
    “While you’re also ignoring the examples I gave of companies that did not do well with the mobile market. In fact, aside from phones, Apple is the only one really doing well in the mobile market and it’s a good possibility that if anyone else was to do really well that it would take away market from Apple.
    Never mind the point on emerging markets that pretty much exclude traditional PC’s and thus adds the distortion of what people can get from what they really want to the numbers as well.”
    At this point, you are just rambling. None of this has much to do with what we’re talking about. Samsung is also doing very well off Android, which is why Apple’s suing them! Anyway, the point is that Microsoft has completely failed in the mobile market, whether anyone does well other than Apple is besides the point. Emerging markets don’t “exclude traditional PCs,” they use a lot of them too: they’re just much better served by mobile devices these days.
    You then ramble some more about the possible reasons that Windows 8 booted faster than Windows RT on that Samsung tablet. Given how fast Android boots on comparable ARM hardware, I’m pretty sure the problem is not the hardware.
    “Again, that only applies to the high end systems. Newer ATOM based devices will be competing with ARM on its own cost and power usage turf.
    The upcoming Clover Trail have all the same enhancements they put into Medfield, which is already going into Smart Phones. While ARM is still more efficient, they still need to use more cores to match or exceed ATOM processors.”
    Medfield is in maybe two smartphones, not exactly a success story. Clover Trail has now been shown on more tablets and hybrid laptops than Medfield was ever shown in any devices. Yes, ARM is not as powerful but you don’t need that much CPU power for most computing these days. That is why ARM dominates on mobile, because the number-crunching is good enough and the battery life is much longer, and why it can do really well on laptops too. Once it takes over that 90% of computing that is now mobile, since everything but desktops are now mobile, it won’t be long before most desktops go ARM too.
    “Mind people like as much freedom as they can get. It’s why Windows is used more than OSX and is why Android is so popular despite it’s faults. So that’s a factor in Intel’s favor.
    Since with x86 you can run any OS, any software, etc. but on ARM the choices are more limited, you’re more likely to be locked down, have support that only lasts less than 2 years on average, etc.
    So it’s not just a matter of pricing…”
    You’re forgetting one big freedom that a lot of people care a lot about: battery life. You say Atom is now comparable, but that’s yet to be shown. Also, pricing is big, which is why the cheaper Android devices and Windows still get a lot more sales than Apple on the software side and why ARM has a big advantage for hardware. And the only reason ARM will be limited on Windows RT tablets is because Microsoft is locking people in with UEFI boot, not an issue for other ARM devices with unlocked bootloaders.
    “Sorry but I specifically gave you the power usage increased caused by the screen itself, which was only 30% and doesn’t account for the rest. The iPad 3 power usage has already been analyzed!”
    Yes, and that analysis found that the backlight on the new iPad used more than twice the power of the last one.
    “Fact is they doubled the number of GPU’s to support the screen and pushed graphics to game console level. They even had to increase the RAM a bit to support it all.
    So the increased power usage is not just the increased in pixels, never mind the main reason for the 30% increase was because of the back light needed to be more intense to get through those denser pixels. The actual pixels themselves don’t consume much more power.
    Really, try doubling the number of GPU’s in your system and not see a big spike in power usage. While increasing resolution means more work for the system. So of course it’s going to be using more power!”
    A long way of saying I was right and the increased resolution of the display, along with all the other hardware necessary to support a much higher resolution screen, is why the new iPad draws much more power. Since none of that has anything to do with the new A5X ARM chip in the new iPad, I’m not sure why you brought up the new iPad to begin with.
    “No, Clover Trail TDP hasn’t been officially released. So wherever you read that isn’t official but also very unlikely as they’re replacing Oak Trail with Clover Trail and it’s specifically meant for tablets. So unlikely they will go over 2W, as any more would require a bulkier casing design to deal with the heat, though figures thrown around sometimes include the additional TDP for the rest of the system, which are usually already figured into system designs.”
    Those were just the numbers I found on some googling, but as you say, official numbers are not yet posted. However, these are numbers from people who have been talking to the various suppliers, so they’re probably accurate, ie Atom still has a ways to go to match ARM for power usage.

  • CyberGusa

    “Nope, I was very precise in my point, that PC sales growth is down in this recession compared to the last recession”

    And you’re still ignoring what I’m pointing out to you. The lack of innovation and advancement is new. Like factors that there was no major OS to upgrade to, lots of PC sales are driven by new software. But in the present situation Vista failed, and Windows 7 only started to take over this year after MS made it clear XP would not longer be supported after 2014.

    So you are basically looking at just one part of the picture and assuming the forest from the trees. There’s a lot you’re not accounting for and that’s leading you to the wrong conclusion.

    “Where do you see “declining growth” in the unit sales I linked to? From
    2003-2007, right before smartphone sales took off, PC sales grew 60%. ”

    No, according to your link they only grew 11.8% to 15.3%! Which is smaller than the jump from 2002 to 2003 that was 2.7% to 10.9%!

    Besides, that was a recovery period, the market tends to snap back during those periods.

    While you seem to forget the end of 2006 is when Vista came out and that killed a lot of the reason for people to get new systems when it was badly received.

    People were stuck with XP and that meant many did not need to upgrade and it took nearly two years to come out with Windows 7 and not till recently has that started to really sell.

    Mind the lack of laptop/desktop sales didn’t stop netbooks from being a hit, capitalizing on the lack of need to use newer hardware and the slowing economy need for more affordable solutions. So the demand was there, just not being directed to the rest of the PC market.

    After 5 years of pretty much no change though is why the netbooks are no longer a rapidly growing market. Most people already have netbooks and without a good reason to upgrade they’ll only purchase to replace and not to upgrade as there’s no real upgrade right now.

    However, that’s as pointed out already going to change. The industry is finally starting to innovate, there’s potentially a new OS that will help drive sales (barring failure of course), next gen hardware is coming out that will offer features and capabilities that weren’t possible before, and hopefully the economy will start recovering over the next few years.

    ” I said Wintel is dying and that the desktop PC would eventually be
    replaced by a home media server, nobody said desktop/laptop sales were”

    No, people aren’t going to be using technology just for entertainment. People need to work, be creative, etc. Only if Wintel doesn’t evolve will it die and that doesn’t seem to be happening!

    “You’re just playing with words at this point. What changed from “a
    desktop in the 80’s” to “what we call a desktop now?” That’s just a
    form factor, one that has basically stayed the same since then.”

    No, lots has changed in the last 30 years and not just form factor. Command Prompt to GUI, keyboard only to inclusion of the mouse, working with programs to whole OS, having to program your own system to just purchasing the software we need, the whole growth of the Internet to the point we’re starting to work in the cloud, the whole range and number of ways we apply computer technology has all changed.

    A lot of the changes were driven by changes in software, we had no use for a mouse before GUI, scroll wheel wasn’t needed before scrolling menus/pages, Windows Start key wasn’t needed until they added the Start Menu to Windows… And those were just the relatively small changes which along with many other changes all added up to a decidedly different environment from where we were just a few decades ago.

    So no, those weren’t just words but an observation of the way things really are!

    “it’s all the people who never customized anything and just got used to the Start Menu after 17 years.”

    Those are arguably a minority, people who don’t want to customize either don’t care or they’d get something that helps them not worry about customizing like a Apple, which as already pointed out isn’t being used by the majority in the PC market.

    The lack of customization also defines the struggle between Android and iOS. So clearly the desire for customization is not just for the minority.

    The tech oriented are just more extreme about it, which is one of the reasons why there’s over 600 different distros of Linux.

    While the point of the declining Start Menu usage is from the vast majority. MS has actual usage data to back that one up and it doesn’t take a computer expert to use 3rd party utilities or simply clutter their desktop with shortcuts.

    “More meaningless words, convergence how? What “symbiosis rather than survival of the fittest?””

    Smart Phone themselves are a example of convergence. While for PC’s we’re seeing Hybrids, Convertible, and interchangeable form factors. These aren’t either/or choices but combining of multiple choices.

    Windows 8 is specifically designed so it can be used on a variety of systems. Windows RT and Windows Phone 8 are designed to work with it. They’re providing the same NT Kernel to all three! So app developers and service providers can cater to all three!

    Linux is similarly working on such capability with Unity, and Apple is increasing added support to OSX with increasing number of iOS like features like notifications, along with sync and their own growing number of online services.

    Google itself created Chrome OS, and is working on creating a desktop friendly UI mode for Android, along with working with Canonical to try and get Ubuntu to work with Android for a literal desktop mode.

    All this is not only convergence but symbiosis because they’re making all the different usages work together.

    “At this point, you are just rambling. None of this has much to do with what we’re talking about.”

    No, it does, I’m pointing out the numbers you’re looking at aren’t showing you the whole picture! You’re equating the growth of market of one type of product with the lack of growth in another, but what I’m pointing out is all the different factors that show they’re not directly linked!

    Emerging markets is one of the reasons for the mobile market’s rapid growth, which doesn’t effect the PC market.

    While the examples of PC companies that did not do well in the mobile market help show how Apple is the exception and not the rule, especially when Apple represents the entirety of it’s share of the market while the competition is split among many but those many combined are still bigger than Apple.

    All goes to show not any one figure or trend truly shows what’s going on.

    Perhaps I’m giving you too many different factors to keep track of but that’s part of the point, there’s a lot involved and not including all of them is what leads to the wrong conclusions!

    ” Yes, ARM is not as powerful but you don’t need that much CPU power for
    most computing these days. That is why ARM dominates on mobile, because
    the number-crunching is good enough and the battery life is much
    longer, and why it can do really well on laptops too.”

    Nothing is ever good enough for long. PC’s started far less powerful than they are today and that didn’t happen because they were good enough, they were. Needs just changed and increased over the years.

    The same is happening to ARM, the devices we use now are many times more powerful than ARM devices just a couple of years ago. This trend shows no sign of slowing down as we’re heading towards more and more quad cores, RAM is going up to 2GB, ARM devices are starting to use SSDs, to offer more port options.

    Even basic web browsing has increased in performance demand over the years. The needs for graphical performance, new features, running web apps, etc. are all increasing the demand for performance.

    While the Cloud may help offset that, the cloud itself has limits and still has a lot of growing to do. So in the meantime device performance still matters. Even Chromebooks show this with the bumping up of specification in the newer models and allowing features like the native client so apps can take advantage of hardware performance.

    Gaming especially is helping drive the demand for performance. Sure, services like OnLive can let people game on virtually any device but that requires a reliable connection and high bandwidth, which most people still don’t have and besides, the image quality is a little lacking.

    Popularity of casual gaming helps offset that demand a bit but people ultimately want the best gaming experience they can get.

    Now while ARM is starting to reach game console level graphics with their latest offerings, that’s still a level that the PC industry past years ago. Existing consoles are about 7 years old now and weren’t cutting edge even when they were new.

    Ports of games like original Max Payne are example of games PC’s had just over a decade ago.

    So ARM still has a very long way to go and what’s okay is both relative and ever changing.

    “Yes, and that analysis found that the backlight on the new iPad used more than twice the power of the last one.”

    No, that analysis studied system power usage and that includes the graphics. You seem to miss the part that stated “Also 4
    times as much memory and processing power is needed for the images.” Also they were estimating for some of those figures as they didn’t know details like the exact number of LEDs.

    Really, if the light was 2.5x the original then it would get hot pretty quick. Even LEDs generate heat, and they are actually sensitive to heat which is why you see LED bulbs with heat sinks to keep them from over heating.

    While the point of the doubling of GPU’s already proved the point. Each GPU wasn’t consuming less power than before and double the number of GPU’s means a definite increase in power consumption.

    Also the point that all ARM device makers are starting to push for similar graphical performance shows that regardless of them all using higher resolution screens that the devices will still increase power usage as they push for more and more performance.

    “A long way of saying I was right and the increased resolution of the
    display, along with all the other hardware necessary to support a much
    higher resolution screen, is why the new iPad draws much more power.”

    No, your point was just the screen. My point was it’s the whole system and the fact that they’re pushing for higher performance would have increased the power usage regardless of the screen.

    “However, these are numbers from people who have been talking to the
    various suppliers, so they’re probably accurate, ie Atom still has a
    ways to go to match ARM for power usage.”

    Yes and no, like I stated a single core ATOM may use about as much power as a dual core ARM but that’s enough to give around the same performance and run time.

    Both are improving, and Clover Trail is specifically meant to replace Oak Trail. The series itself may be made for a range of application, like Servers don’t need to be limited to 2W, but tablets do. Especially if they are to be charged over USB, which present chargers max out at about 2W.

    The newer ARM based on Cortex A15 or similar, like the Krait, are going to give ARM a overall edge but like I pointed out also Intel is getting ready with its next gen offerings as well and they’re pushing their FAB advantage which can potentially let them close that remaining gap as long as they can stay ahead.

    They’re already confident enough to plan for up to quad cores with their next gen chips and still plan to go fan-less, which means a continued drop in power usage.

    So we’ll see how that goes, I’m not saying ARM will lose or won’t eventually win but it’s definitely too early to underestimate Intel.

    Many ARM manufacturers had a lot of problems with the move to 28nm, and if that continues then it could give Intel the edge even if ARM still keeps the edge on design efficiency.

    While ARM needs to make the jump to 64bit to compete with Intel on the higher end, which even if you don’t think they’ll need in consumer products they’ll definitely need in the server market and other high end usages.

  • Joakim

    “And you’re still ignoring what I’m pointing out to you. The lack of innovation and advancement is new. Like factors that there was no major OS to upgrade to, lots of PC sales are driven by new software. But in the present situation Vista failed, and Windows 7 only started to take over this year after MS made it clear XP would not longer be supported after 2014.”

    The facts are not with you here. Vista came out in late 2006, during the 2003-7 PC boom I mentioned. Windows 7 came out in 2009, when PC growth was leveling off, yet sold twice as many licenses as Vista. So despite Windows 7 doing much better, PC sales slowed down when it came out, while PC sales grew a lot faster during the bust that was Vista, both of which are the opposite of your claim.

    “‘Where do you see “declining growth” in the unit sales I linked to? From
    2003-2007, right before smartphone sales took off, PC sales grew 60%. ‘
    No, according to your link they only grew 11.8% to 15.3%! Which is smaller than the jump from 2002 to 2003 that was 2.7% to 10.9%!”

    I have aggregated the sales growth by 4-year periods, 1999-2003, 2003-2007, and 2007-2011, in order to compare similar periods, not sure what’s hard to understand about that.

    “Besides, that was a recovery period, the market tends to snap back during those periods.”

    Yes, but your claim was that the market was already maturing, but sales growth was actually increasing before the smartphone hit the scene!

    “Mind the lack of laptop/desktop sales didn’t stop netbooks from being a hit, capitalizing on the lack of need to use newer hardware and the slowing economy need for more affordable solutions. So the demand was there, just not being directed to the rest of the PC market.”

    Netbook sales of 30 million last year, 8% of the PC market, is a “hit?” What does that make ARM then, with 550 million smartphones and tablets sold last year, a blockbuster that is killing the PC? I think so. :)

    “‘ I said Wintel is dying and that the desktop PC would eventually be
    replaced by a home media server, nobody said desktop/laptop sales were’
    No, people aren’t going to be using technology just for entertainment. People need to work, be creative, etc. Only if Wintel doesn’t evolve will it die and that doesn’t seem to be happening!”

    Your assumption is that you need Wintel or a desktop to do work or be creative. Plenty of people get creative work done on a Mac, arguably more so than Windows now. Once Ubuntu on Android is mature, it’s possible people will just have one smartphone that they dock at a monitor and keyboard/mouse on their desk, ie no Wintel desktop necessary, just as Charbax always dreams about. πŸ˜‰ I don’t think Ubuntu will be it, but someone new might come along and deliver it. But the fact remains that the vast majority of people are interested only in consumption and entertainment most of the time, so the work/creative portion is pretty small.

    “No, lots has changed in the last 30 years and not just form factor. Command Prompt to GUI, keyboard only to inclusion of the mouse, working with programs to whole OS, having to program your own system to just purchasing the software we need, the whole growth of the Internet to the point we’re starting to work in the cloud, the whole range and number of ways we apply computer technology has all changed.
    A lot of the changes were driven by changes in software, we had no use for a mouse before GUI, scroll wheel wasn’t needed before scrolling menus/pages, Windows Start key wasn’t needed until they added the Start Menu to Windows… And those were just the relatively small changes which along with many other changes all added up to a decidedly different environment from where we were just a few decades ago.
    So no, those weren’t just words but an observation of the way things really are!”

    Yes, but the fact that the guts of a desktop computer have changed a lot over the years is irrelevant to the fact that that form factor is now obsolete, which is why out of 900 million computing devices sold last year, only 10% were desktop PCs. The rest were all mobile, laptops and smartphones and tablets, which is why the desktop PC is about to die out altogether.

    “‘it’s all the people who never customized anything and just got used to the Start Menu after 17 years.’
    Those are arguably a minority, people who don’t want to customize either don’t care or they’d get something that helps them not worry about customizing like a Apple, which as already pointed out isn’t being used by the majority in the PC market.
    The lack of customization also defines the struggle between Android and iOS. So clearly the desire for customization is not just for the minority.
    The tech oriented are just more extreme about it, which is one of the reasons why there’s over 600 different distros of Linux.
    While the point of the declining Start Menu usage is from the vast majority. MS has actual usage data to back that one up and it doesn’t take a computer expert to use 3rd party utilities or simply clutter their desktop with shortcuts.”

    This is perhaps where we fundamentally disagree. You seem to think that people who litter their desktop with shortcuts or use “3rd party options like Stardocks, virtual desktops, etc.” are the majority, ie those changing the UI of their OS. I’m fairly certain those people constitute a very small minority. The vast majority mainly only “customizes” in the sense that most people “customize” Android: they want access to a large variety of apps that let them do more with their computing device. That’s not the UI customizations we were talking about.
    Most people don’t change the UI, which is why they will get so thrown by the massive changes Win 8 makes to their basic UI. As for the MS claim that they have usage data behind it, I suspect their sample is not diverse enough.

    Joakim – “Windows 8 will be a giant failure on the desktop, worse than Vista, which after all didn’t make such a drastic UI change. On touch devices like tablets and hybrid laptops, it will do better, since that’s what it was designed for, but will not even beat Android in market share. Please submit your own predictions.”

    You did not submit your own predictions. You claim the Wintel PC will be around: please back that up with some actual predictions for how well Windows 8 will do, as I have.

    ‘”More meaningless words, convergence how? What “symbiosis rather than survival of the fittest?”‘
    Smart Phone themselves are a example of convergence. While for PC’s we’re seeing Hybrids, Convertible, and interchangeable form factors. These aren’t either/or choices but combining of multiple choices.
    Windows 8 is specifically designed so it can be used on a variety of systems. Windows RT and Windows Phone 8 are designed to work with it. They’re providing the same NT Kernel to all three! So app developers and service providers can cater to all three!
    Linux is similarly working on such capability with Unity, and Apple is increasing added support to OSX with increasing number of iOS like features like notifications, along with sync and their own growing number of online services.
    Google itself created Chrome OS, and is working on creating a desktop friendly UI mode for Android, along with working with Canonical to try and get Ubuntu to work with Android for a literal desktop mode.
    All this is not only convergence but symbiosis because they’re making all the different usages work together.’

    It is not “convergence” for Wintel to slap on some mobile features like a touch interface before it disappears, that’s obsolescence. πŸ˜€ If everybody is using Android/iOS ARM smartphones/tablets with touch interfaces and a desktop mode, that’s not PC convergence, that’s the death of the Wintel PC that Charbax trumpets! :) The fact that mainframes still exist is not “symbiosis,” that’s a small niche that still exists yet is basically dead for most users. If there are only so many Wintel desktop PCs as there are mainframes today, that’s the death of the PC. The point is that you are just playing with words to obscure the fact that the Wintel PC is indeed dying off.

    “You’re equating the growth of market of one type of product with the lack of growth in another, but what I’m pointing out is all the different factors that show they’re not directly linked!”

    Of course they’re linked, it would be silly to suggest that the great rise in laptop sales over the years has not affected the sales of desktops. The same is happening with mobile devices crowding out PC sales.

    “Emerging markets is one of the reasons for the mobile market’s rapid growth, which doesn’t effect the PC market.”

    The fact remains that they prefer mobile devices to PCs and if these new mobile devices didn’t exist, many’d be buying a lot more PCs, so yet again there is a direct link. I take your point that a majority might choose not to buy any computer if there were no mobile device, but that’s true in the first world also.

    “While the examples of PC companies that did not do well in the mobile market help show how Apple is the exception and not the rule, especially when Apple represents the entirety of it’s share of the market while the competition is split among many but those many combined are still bigger than Apple.”

    What does this have to do with anything? Whether it’s PC companies who do well in mobile or new companies is completely irrelevant to this discussion. As for other companies being bigger than Apple, Apple has more profit than all of them combined, so no, Apple is still much more important. Apple had almost as much profit last year as Google had revenues! That’s how much Apple is killing Android and other companies.

    “All goes to show not any one figure or trend truly shows what’s going on.
    Perhaps I’m giving you too many different factors to keep track of but that’s part of the point, there’s a lot involved and not including all of them is what leads to the wrong conclusions!”

    Yes, it’s complex but your other “factors” are largely extraneous, ie you’re mostly just throwing out things that don’t matter. For example, you talk about 64-bit like it’s a big deal, when most ARM devices have less than 1 GB of memory. The 4 GB limit of 32-bit ARM chips is completely irrelevant to these devices, yet you keep banging on about it.

    “Nothing is ever good enough for long. PC’s started far less powerful than they are today and that didn’t happen because they were good enough, they were. Needs just changed and increased over the years.
    The same is happening to ARM, the devices we use now are many times more powerful than ARM devices just a couple of years ago. This trend shows no sign of slowing down as we’re heading towards more and more quad cores, RAM is going up to 2GB, ARM devices are starting to use SSDs, to offer more port options.
    Even basic web browsing has increased in performance demand over the years. The needs for graphical performance, new features, running web apps, etc. are all increasing the demand for performance.
    While the Cloud may help offset that, the cloud itself has limits and still has a lot of growing to do. So in the meantime device performance still matters. Even Chromebooks show this with the bumping up of specification in the newer models and allowing features like the native client so apps can take advantage of hardware performance.”

    Not sure what your point is here, since you yourself admit that ARM devices are getting “many times more powerful” all the time. My point is that ARM devices have now hit the mass middle, where they’re powerful enough for most people’s uses. Yes, if you’re running some computational fluidics simulation, you’re not going to run it on an ARM device, but then you’re probably not going to run that on a Wintel PC either, more likely an IBM POWER mainframe. πŸ˜‰ And as the mass middle moves up, ARM will move with it. As for web browsing getting more demanding, most of that is useless and almost nobody uses all those new “features,” like SVG or HTML5 video. Of course device performance still matters, but the point is that ARM now suffices and is also getting faster as necessary.

    “Gaming especially is helping drive the demand for performance. Sure, services like OnLive can let people game on virtually any device but that requires a reliable connection and high bandwidth, which most people still don’t have and besides, the image quality is a little lacking.
    Popularity of casual gaming helps offset that demand a bit but people ultimately want the best gaming experience they can get.
    Now while ARM is starting to reach game console level graphics with their latest offerings, that’s still a level that the PC industry past years ago. Existing consoles are about 7 years old now and weren’t cutting edge even when they were new.
    Ports of games like original Max Payne are example of games PC’s had just over a decade ago.
    So ARM still has a very long way to go and what’s okay is both relative and ever changing.”

    And how many people play 3D PC games that need lots of number-crunching? They’re a small minority of computing users, a vocal minority, but small nonetheless. OnLive just basically went bankrupt, so obviously that didn’t work. No, there are many times more people “casual gaming” on ARM devices than there are those who “ultimately want the best gaming experience they can get:” the vast majority of people are not interested in first-person shooters like Quake or Doom 3.

    “‘Yes, and that analysis found that the backlight on the new iPad used more than twice the power of the last one.’
    No, that analysis studied system power usage and that includes the graphics. You seem to miss the part that stated “Also 4
    times as much memory and processing power is needed for the images.” Also they were estimating for some of those figures as they didn’t know details like the exact number of LEDs.”

    Of course it includes graphics, that’s my whole point. The new iPad increased the number of display pixels by four, so various graphical subsystems like the GPU or backlight used a lot more power. However, your initial claim was the ARM CPU was somehow to blame for the increased power draw in the new iPad, which is not the case.

    “Really, if the light was 2.5x the original then it would get hot pretty quick. Even LEDs generate heat, and they are actually sensitive to heat which is why you see LED bulbs with heat sinks to keep them from over heating.”

    News flash, the new iPad runs very hot!

    “While the point of the doubling of GPU’s already proved the point. Each GPU wasn’t consuming less power than before and double the number of GPU’s means a definite increase in power consumption.
    Also the point that all ARM device makers are starting to push for similar graphical performance shows that regardless of them all using higher resolution screens that the devices will still increase power usage as they push for more and more performance.”

    None of this has anything to do with ARM CPUs using more power, as you are talking about the GPUs and a completely separate factor of a display arms race. You initially mentioned the new iPad using more power as an example of ARM chips using more power, but now admit that it is the display and GPU that are the real issue. However, if all these devices, whether ARM or Intel, have much bigger displays, the fact remains that ARM will still have an advantage because it uses less power than Intel chips. :) It is true that higher-resolution displays will also stress the CPU a bit more, but that’s negligible.

    “‘A long way of saying I was right and the increased resolution of the
    display, along with all the other hardware necessary to support a much
    higher resolution screen, is why the new iPad draws much more power.’
    No, your point was just the screen. My point was it’s the whole system and the fact that they’re pushing for higher performance would have increased the power usage regardless of the screen.”

    My point was that the benchmarks show that it was only the screen and associated display hardware that drove the new iPad’s higher power usage and that the ARM CPU basically did not contribute. If they had not decided to up the resolution to such a ridiculous extreme and had only upgraded the ARM CPU, power usage would probably have gone down, not up.

    “The newer ARM based on Cortex A15 or similar, like the Krait, are going to give ARM a overall edge but like I pointed out also Intel is getting ready with its next gen offerings as well and they’re pushing their FAB advantage which can potentially let them close that remaining gap as long as they can stay ahead.
    They’re already confident enough to plan for up to quad cores with their next gen chips and still plan to go fan-less, which means a continued drop in power usage.”

    ARM is not standing still either, with their new Cortex A5, which is almost as good as the A9 but uses less power and is cheaper. It is going out new phones now, as Charbax just saw. I’ll believe Intel’s power usage claims when I see it. Considering no smartphone runs on Intel chips now, obviously nobody believes them yet. πŸ˜‰

    “So we’ll see how that goes, I’m not saying ARM will lose or won’t eventually win but it’s definitely too early to underestimate Intel.”

    I agree, but it’s certainly not looking good for Intel right now.

    “Many ARM manufacturers had a lot of problems with the move to 28nm, and if that continues then it could give Intel the edge even if ARM still keeps the edge on design efficiency.”

    ARM is not winning because of technology: it’s winning because it’s good enough and has a whole new business model, one built around charging much less money. That will be very difficult for Intel to beat. They did beat AMD at it once, but the much lower power usage of ARM is a problem they haven’t been able to solve.

  • Joakim

    “And you’re still ignoring what I’m pointing out to you. The lack of innovation and advancement is new. Like factors that there was no major OS to upgrade to, lots of PC sales are driven by new software. But in the present situation Vista failed, and Windows 7 only started to take over this year after MS made it clear XP would not longer be supported after 2014.”

    The facts are not with you here. Vista came out in late 2006, during the 2003-7 PC boom I mentioned. Windows 7 came out in 2009, when PC growth was leveling off, yet sold twice as many licenses as Vista. So despite Windows 7 doing much better, PC sales slowed down when it came out, while PC sales grew a lot faster during the bust that was Vista, both of which are the opposite of your claim.

    “‘Where do you see “declining growth” in the unit sales I linked to? From
    2003-2007, right before smartphone sales took off, PC sales grew 60%. ‘
    No, according to your link they only grew 11.8% to 15.3%! Which is smaller than the jump from 2002 to 2003 that was 2.7% to 10.9%!”

    I have aggregated the sales growth by 4-year periods, 1999-2003, 2003-2007, and 2007-2011, in order to compare similar periods, not sure what’s hard to understand about that.

    “Besides, that was a recovery period, the market tends to snap back during those periods.”

    Yes, but your claim was that the market was already maturing, but sales growth was actually increasing before the smartphone hit the scene!

    “Mind the lack of laptop/desktop sales didn’t stop netbooks from being a hit, capitalizing on the lack of need to use newer hardware and the slowing economy need for more affordable solutions. So the demand was there, just not being directed to the rest of the PC market.”

    Netbook sales of 30 million last year, 8% of the PC market, is a “hit?” What does that make ARM then, with 550 million smartphones and tablets sold last year, a blockbuster that is killing the PC? I think so. :)

    “‘ I said Wintel is dying and that the desktop PC would eventually be
    replaced by a home media server, nobody said desktop/laptop sales were’
    No, people aren’t going to be using technology just for entertainment. People need to work, be creative, etc. Only if Wintel doesn’t evolve will it die and that doesn’t seem to be happening!”

    Your assumption is that you need Wintel or a desktop to do work or be creative. Plenty of people get creative work done on a Mac, arguably more so than Windows now. Once Ubuntu on Android is mature, it’s possible people will just have one smartphone that they dock at a monitor and keyboard/mouse on their desk, ie no Wintel desktop necessary, just as Charbax always dreams about. πŸ˜‰ I don’t think Ubuntu will be it, but someone new might come along and deliver it. But the fact remains that the vast majority of people are interested only in consumption and entertainment most of the time, so the work/creative portion is pretty small.

    “No, lots has changed in the last 30 years and not just form factor. Command Prompt to GUI, keyboard only to inclusion of the mouse, working with programs to whole OS, having to program your own system to just purchasing the software we need, the whole growth of the Internet to the point we’re starting to work in the cloud, the whole range and number of ways we apply computer technology has all changed.
    A lot of the changes were driven by changes in software, we had no use for a mouse before GUI, scroll wheel wasn’t needed before scrolling menus/pages, Windows Start key wasn’t needed until they added the Start Menu to Windows… And those were just the relatively small changes which along with many other changes all added up to a decidedly different environment from where we were just a few decades ago.
    So no, those weren’t just words but an observation of the way things really are!”

    Yes, but the fact that the guts of a desktop computer have changed a lot over the years is irrelevant to the fact that that form factor is now obsolete, which is why out of 900 million computing devices sold last year, only 10% were desktop PCs. The rest were all mobile, laptops and smartphones and tablets, which is why the desktop PC is about to die out altogether.

    “‘it’s all the people who never customized anything and just got used to the Start Menu after 17 years.’
    Those are arguably a minority, people who don’t want to customize either don’t care or they’d get something that helps them not worry about customizing like a Apple, which as already pointed out isn’t being used by the majority in the PC market.
    The lack of customization also defines the struggle between Android and iOS. So clearly the desire for customization is not just for the minority.
    The tech oriented are just more extreme about it, which is one of the reasons why there’s over 600 different distros of Linux.
    While the point of the declining Start Menu usage is from the vast majority. MS has actual usage data to back that one up and it doesn’t take a computer expert to use 3rd party utilities or simply clutter their desktop with shortcuts.”

    This is perhaps where we fundamentally disagree. You seem to think that people who litter their desktop with shortcuts or use “3rd party options like Stardocks, virtual desktops, etc.” are the majority, ie those changing the UI of their OS. I’m fairly certain those people constitute a very small minority. The vast majority mainly only “customizes” in the sense that most people “customize” Android: they want access to a large variety of apps that let them do more with their computing device. That’s not the UI customizations we were talking about.
    Most people don’t change the UI, which is why they will get so thrown by the massive changes Win 8 makes to their basic UI. As for the MS claim that they have usage data behind it, I suspect their sample is not diverse enough.

    Joakim – “Windows 8 will be a giant failure on the desktop, worse than Vista, which after all didn’t make such a drastic UI change. On touch devices like tablets and hybrid laptops, it will do better, since that’s what it was designed for, but will not even beat Android in market share. Please submit your own predictions.”

    You did not submit your own predictions. You claim the Wintel PC will be around: please back that up with some actual predictions for how well Windows 8 will do, as I have.

    ‘”More meaningless words, convergence how? What “symbiosis rather than survival of the fittest?”‘
    Smart Phone themselves are a example of convergence. While for PC’s we’re seeing Hybrids, Convertible, and interchangeable form factors. These aren’t either/or choices but combining of multiple choices.
    Windows 8 is specifically designed so it can be used on a variety of systems. Windows RT and Windows Phone 8 are designed to work with it. They’re providing the same NT Kernel to all three! So app developers and service providers can cater to all three!
    Linux is similarly working on such capability with Unity, and Apple is increasing added support to OSX with increasing number of iOS like features like notifications, along with sync and their own growing number of online services.
    Google itself created Chrome OS, and is working on creating a desktop friendly UI mode for Android, along with working with Canonical to try and get Ubuntu to work with Android for a literal desktop mode.
    All this is not only convergence but symbiosis because they’re making all the different usages work together.’

    It is not “convergence” for Wintel to slap on some mobile features like a touch interface before it disappears, that’s obsolescence. πŸ˜€ If everybody is using Android/iOS ARM smartphones/tablets with touch interfaces and a desktop mode, that’s not PC convergence, that’s the death of the Wintel PC that Charbax trumpets! :) The fact that mainframes still exist is not “symbiosis,” that’s a small niche that still exists yet is basically dead for most users. If there are only so many Wintel desktop PCs as there are mainframes today, that’s the death of the PC. The point is that you are just playing with words to obscure the fact that the Wintel PC is indeed dying off.

    “You’re equating the growth of market of one type of product with the lack of growth in another, but what I’m pointing out is all the different factors that show they’re not directly linked!”

    Of course they’re linked, it would be silly to suggest that the great rise in laptop sales over the years has not affected the sales of desktops. The same is happening with mobile devices crowding out PC sales.

    “Emerging markets is one of the reasons for the mobile market’s rapid growth, which doesn’t effect the PC market.”

    The fact remains that they prefer mobile devices to PCs and if these new mobile devices didn’t exist, many’d be buying a lot more PCs, so yet again there is a direct link. I take your point that a majority might choose not to buy any computer if there were no mobile device, but that’s true in the first world also.

    “While the examples of PC companies that did not do well in the mobile market help show how Apple is the exception and not the rule, especially when Apple represents the entirety of it’s share of the market while the competition is split among many but those many combined are still bigger than Apple.”

    What does this have to do with anything? Whether it’s PC companies who do well in mobile or new companies is completely irrelevant to this discussion. As for other companies being bigger than Apple, Apple has more profit than all of them combined, so no, Apple is still much more important. Apple had almost as much profit last year as Google had revenues! That’s how much Apple is killing Android and other companies.

    “All goes to show not any one figure or trend truly shows what’s going on.
    Perhaps I’m giving you too many different factors to keep track of but that’s part of the point, there’s a lot involved and not including all of them is what leads to the wrong conclusions!”

    Yes, it’s complex but your other “factors” are largely extraneous, ie you’re mostly just throwing out things that don’t matter. For example, you talk about 64-bit like it’s a big deal, when most ARM devices have less than 1 GB of memory. The 4 GB limit of 32-bit ARM chips is completely irrelevant to these devices, yet you keep banging on about it.

    “Nothing is ever good enough for long. PC’s started far less powerful than they are today and that didn’t happen because they were good enough, they were. Needs just changed and increased over the years.
    The same is happening to ARM, the devices we use now are many times more powerful than ARM devices just a couple of years ago. This trend shows no sign of slowing down as we’re heading towards more and more quad cores, RAM is going up to 2GB, ARM devices are starting to use SSDs, to offer more port options.
    Even basic web browsing has increased in performance demand over the years. The needs for graphical performance, new features, running web apps, etc. are all increasing the demand for performance.
    While the Cloud may help offset that, the cloud itself has limits and still has a lot of growing to do. So in the meantime device performance still matters. Even Chromebooks show this with the bumping up of specification in the newer models and allowing features like the native client so apps can take advantage of hardware performance.”

    Not sure what your point is here, since you yourself admit that ARM devices are getting “many times more powerful” all the time. My point is that ARM devices have now hit the mass middle, where they’re powerful enough for most people’s uses. Yes, if you’re running some computational fluidics simulation, you’re not going to run it on an ARM device, but then you’re probably not going to run that on a Wintel PC either, more likely an IBM POWER mainframe. πŸ˜‰ And as the mass middle moves up, ARM will move with it. As for web browsing getting more demanding, most of that is useless and almost nobody uses all those new “features,” like SVG or HTML5 video. Of course device performance still matters, but the point is that ARM now suffices and is also getting faster as necessary.

    “Gaming especially is helping drive the demand for performance. Sure, services like OnLive can let people game on virtually any device but that requires a reliable connection and high bandwidth, which most people still don’t have and besides, the image quality is a little lacking.
    Popularity of casual gaming helps offset that demand a bit but people ultimately want the best gaming experience they can get.
    Now while ARM is starting to reach game console level graphics with their latest offerings, that’s still a level that the PC industry past years ago. Existing consoles are about 7 years old now and weren’t cutting edge even when they were new.
    Ports of games like original Max Payne are example of games PC’s had just over a decade ago.
    So ARM still has a very long way to go and what’s okay is both relative and ever changing.”

    And how many people play 3D PC games that need lots of number-crunching? They’re a small minority of computing users, a vocal minority, but small nonetheless. OnLive just basically went bankrupt, so obviously that didn’t work. No, there are many times more people “casual gaming” on ARM devices than there are those who “ultimately want the best gaming experience they can get:” the vast majority of people are not interested in first-person shooters like Quake or Doom 3.

    “‘Yes, and that analysis found that the backlight on the new iPad used more than twice the power of the last one.’
    No, that analysis studied system power usage and that includes the graphics. You seem to miss the part that stated “Also 4
    times as much memory and processing power is needed for the images.” Also they were estimating for some of those figures as they didn’t know details like the exact number of LEDs.”

    Of course it includes graphics, that’s my whole point. The new iPad increased the number of display pixels by four, so various graphical subsystems like the GPU or backlight used a lot more power. However, your initial claim was the ARM CPU was somehow to blame for the increased power draw in the new iPad, which is not the case.

    “Really, if the light was 2.5x the original then it would get hot pretty quick. Even LEDs generate heat, and they are actually sensitive to heat which is why you see LED bulbs with heat sinks to keep them from over heating.”

    News flash, the new iPad runs very hot!

    “While the point of the doubling of GPU’s already proved the point. Each GPU wasn’t consuming less power than before and double the number of GPU’s means a definite increase in power consumption.
    Also the point that all ARM device makers are starting to push for similar graphical performance shows that regardless of them all using higher resolution screens that the devices will still increase power usage as they push for more and more performance.”

    None of this has anything to do with ARM CPUs using more power, as you are talking about the GPUs and a completely separate factor of a display arms race. You initially mentioned the new iPad using more power as an example of ARM chips using more power, but now admit that it is the display and GPU that are the real issue. However, if all these devices, whether ARM or Intel, have much bigger displays, the fact remains that ARM will still have an advantage because it uses less power than Intel chips. :) It is true that higher-resolution displays will also stress the CPU a bit more, but that’s negligible.

    “‘A long way of saying I was right and the increased resolution of the
    display, along with all the other hardware necessary to support a much
    higher resolution screen, is why the new iPad draws much more power.’
    No, your point was just the screen. My point was it’s the whole system and the fact that they’re pushing for higher performance would have increased the power usage regardless of the screen.”

    My point was that the benchmarks show that it was only the screen and associated display hardware that drove the new iPad’s higher power usage and that the ARM CPU basically did not contribute. If they had not decided to up the resolution to such a ridiculous extreme and had only upgraded the ARM CPU, power usage would probably have gone down, not up.

    “The newer ARM based on Cortex A15 or similar, like the Krait, are going to give ARM a overall edge but like I pointed out also Intel is getting ready with its next gen offerings as well and they’re pushing their FAB advantage which can potentially let them close that remaining gap as long as they can stay ahead.
    They’re already confident enough to plan for up to quad cores with their next gen chips and still plan to go fan-less, which means a continued drop in power usage.”

    ARM is not standing still either, with their new Cortex A5, which is almost as good as the A9 but uses less power and is cheaper. It is going out new phones now, as Charbax just saw. I’ll believe Intel’s power usage claims when I see it. Considering no smartphone runs on Intel chips now, obviously nobody believes them yet. πŸ˜‰

    “So we’ll see how that goes, I’m not saying ARM will lose or won’t eventually win but it’s definitely too early to underestimate Intel.”

    I agree, but it’s certainly not looking good for Intel right now.

    “Many ARM manufacturers had a lot of problems with the move to 28nm, and if that continues then it could give Intel the edge even if ARM still keeps the edge on design efficiency.”

    ARM is not winning because of technology: it’s winning because it’s good enough and has a whole new business model, one built around charging much less money. That will be very difficult for Intel to beat. They did beat AMD at it once, but the much lower power usage of ARM is a problem they haven’t been able to solve.

  • CyberGusa

    “The facts are not with you here. Vista came out in late 2006, during the 2003-7 PC boom I mentioned.”

    Sorry but the facts are with me, Vista flopped! Which lead to the decline! Contrary to whatever you may be thinking declines don’t just happen overnight. There’s always things/events that lead to it!

    Really, people weren’t expecting Vista to flop. So it’s not like they exactly saw it coming but the effects are undeniable once it did!

    XP at that point wasn’t going to help push new sales. So all the way till Windows 7 release in 2009 they had nothing new to offer! Even then people had grown so used to XP that it wasn’t until MS announced that they were finally going to drop support of XP in 2014 that Windows 7 finally started replacing it and it wasn’t until this year that they broke the 50% mark!

    Also, there’s this thing called market momentum you might want to look up. The effects of changes in markets are like waves. They don’t just turn direction on a dime but take time.

    Vista was still riding the wave of the good times but it put the brakes on from then on and Windows 7 was basically just the fixed version of Vista but by then the economy had already changed.

    While doing better than a failed version is pretty easy to do as long as you don’t release another failed version but Windows 7 didn’t really started taking off until recently as I pointed out.

    And we’ve yet to see what happens with Windows RT/8…

    “I have aggregated the sales growth by 4-year periods, 1999-2003,
    2003-2007, and 2007-2011, in order to compare similar periods, not sure
    what’s hard to understand about that.”

    Just pointing out you’re exaggerating the percentages from what the article actually shows. Like stating 60% when it was much less. While you’re also ignoring or apparently not aware that declines start to happen before we actually notice them. Along with how different factors can play off each other to create a larger effect. Since it’s not any one factor that lead to the present slow period for PC sales but multiple factors.

    So you’ve basically taken a overly simplified view that ignores many factors in order to come to your conclusion and that’s why it’s inaccurate!

    “Yes, but your claim was that the market was already maturing, but sales
    growth was actually increasing before the smartphone hit the scene!”

    They were, maturing doesn’t mean the market stops growing. Just that it slows down. Most indicators of decline are actually measuring the decline of growth and not actual shrinking of the market.

    Unlike the mobile market that has emerging markets to grow the market into more parts of the world. So even in a recession they are able to show considerable growth because of the shear number of people they can offer them to, but the same is not true of other products that are pretty much stuck in the mainstream western markets, some even to particular regions.

    Low pricing is one of the reasons why Netbooks saw a similar rapid growth, but it never took from other laptop sales and eventually reached market saturation itself.

    Similarly, it’s only a matter of time before the mobile market also reaches saturation and growth will slow down.

    The rapid obsolescence and decreasing pricing is helping to push that time further out but it’ll eventually reach it nonetheless. Thus the importance of growth into different parts of the market… At least until whatever time someone manages to re-invent the market and have the cycle begin again but that’s easier said than done.

    “Your assumption is that you need Wintel or a desktop to do work or be
    creative. Plenty of people get creative work done on a Mac, arguably
    more so than Windows now. ”

    It’s no assumption that people need the performance of a PC and the capabilities of a desktop OS to really get things done. Even people who like to use tablets still get laptops and for doing real work the desktop is still the best both ergonomically and efficiently.

    The fact that peripherals like keyboards are one of the main accessories bought for tablets is telling by itself! Along with services like Citrix provides for running Windows remotely.

    Even something as simple as the web browser is still being worked on for mobile devices. Support for things like HTML5, etc are still better on desktops version of the browsers.

    The Chrome port is pretty much the first time that is starting to change, but it still lacks many of the things supported in the desktop version. Like no WebGL support, no WebView support, no Native Client support, apps and extensions are not supported, among other more minor things they left out of the version for Android. So it’s still a crippled version!

    While if you’re talking of actual Macs, then you should be aware that Apple uses Intel for their PC hardware and have been doing so for a number of years now.

    MacBook Air, MacBook Pro, iMacs, and Mac Pros are all running on Intel hardware!

    While it doesn’t matter if they use OSX, Windows, or GNU/Linux as they all allow for more than any mobile OS does. It also doesn’t matter whether they use Intel or AMD as both are x86 solutions that provide more performance than ARM can offer.

    So let’s not confuse the so called biased term of “Wintel” with what’s actually being discussed!

    “What does this have to do with anything? Whether it’s PC companies who
    do well in mobile or new companies is completely irrelevant to this
    discussion.”

    Apple isn’t the same as other PC companies for one thing. They never did very well in the PC market. It was mobile products like the iPod that saved the company and helped, along with their growing number of other mobile devices to grow them into what they are now. Even with the success of the MacBook Air, they’re still beginning to trim the number of OSX products they make.

    So using them as the poster child of how to compare the PC market and mobile market is flawed to begin with! There are plenty of companies doing better than Apple in the PC market.

    Apple also doesn’t like to share their market share. So comparing them to companies that do is also flawed. Unlike Apple products, people have the option to get different versions of the competing products from a variety of other companies. So profits are split among those various companies but not for Apple.

    Never mind other factors, like MS has pretty much lost the server market, lost the search engine to Google, and is still trying to recover from the damage Vista did to them. All of which have nothing to do with the rise or fall of the PC market.

    Along with the previous points on how MS is primarily a software company and shouldn’t be compared to a company that also profits on hardware.

    Really, do I have to walk you through all this? It should be pretty obvious from what I already have pointed out.

    “The fact remains that they prefer mobile devices to PCs and if these new
    mobile devices didn’t exist, many’d be buying a lot more PCs, so yet
    again there is a direct link. ”

    No, that’s not a fact but your opinion and I’ve clearly punched a lot of holes in your reasoning for that opinion.

    You just don’t seem to want to account for those discrepancies. Like pointing out
    that the emerging markets don’t apply to PC sales but do to mobile.
    This isn’t because they all prefer mobile but because they can’t all afford or
    support pretty much anything else.

    Some of those emerging markets are countries without reliable power. So would be lucky to be able to even charge a phone, let alone run a desktop.

    While also the point of calling them emerging markets is the fact they’re new markets. So increase the number of potential consumers but only for products those markets can afford and support. Making comparisons unfairly disproportional unless they’re actually factored for…

    Along with the other points like how much of the mobile growth is actually just replacing units at a faster rate than PC users do, which again doesn’t mean they want mobile more but just have to replace them more often.

    Besides, I also pointed out that the only comparison can be made are to recent trends because only recently have mobile and PC market started to overlap.
    Everything previous is thus separate markets with no relevance to each
    other!

    “None of this has anything to do with ARM CPUs using more power, as you
    are talking about the GPUs and a completely separate factor of a display
    arms race. ”

    Wrong, everything that goes into the system power usage matters and if you’ve forgotten we are talking about SoCs!!! The GPU, CPU, etc are all together!

    I also didn’t just point to the GPU but also the RAM and the CPU. Like the reasons why some phones come with quad cores but others come with dual cores but are otherwise the same model with the same parts.

    This is all because they are pushing the limits of power for mobile devices. Between the demand for ever increasing performance and getting the most run time possible is pushing the limits of what’s possible now.

    Really, why do you think they’ve been working on both low power and high power processors? They can’t easily be designed for both! So they instead switch between them to improve efficiency.

    Examples like the Cortex A5 actually show how ARM is starting to reach its design limits. Since they can’t simply make one processor capable of both high performance and low power usage. So instead they make two types.

    The Cortex A5 will usually get coupled with a A15 to provide both high performance when needed and low power usage when not.

    Mind the A5 is also a much more basic processor. It goes back to In Order Processing for example. So it provides the lower power usage with a compromise in performance.

    Similarly they’re employing methods like cycling the number of active cores as needed. So SoC’s like the Tegra 3 only use all cores when operating at full power but otherwise uses fewer cores or none and switches to the 5th low power core to conserve power. Otherwise it would use more power than the Tegra 2.

    This also means run times are becoming deceptive as running at full power will of course result in less run time than running at low power.

    “My point was that the benchmarks show that it was only the screen and
    associated display hardware that drove the new iPad’s higher power usage
    and that the ARM CPU basically did not contribute.”

    Things like the memory are shared, it’s not just the GPU and higher resolution also makes the CPU work harder as not everything is handled by the GPU.

    While as also pointed out we’re talking about a SoC, with everything integrated.

    Besides, any good system requires both good CPU performance and good GPU performance. You can have plenty of one or the other but without both the performance isn’t balanced and you’ll run into more things that won’t run well because of it.

    Like gaming, some games require more CPU performance than GPU. While others need the opposite. Without a well balanced system then you’ll won’t be able to handle both extremes well.

    The general demand for higher performance also means you can’t go by what just the minimum requirements may be.

    What’s okay for one usage isn’t always okay for others. The PC market covers a very wide range of usages and it requires both flexibility and capability to really compete in that market or you’ll just wind up with another niche product.

    “News flash, the new iPad runs very hot!”

    Warmer is not hot, the iPad will never burn you! If it really ran hot then it would need a active cooling system, IE a case fan or similar.

    But that does prove my point that they have to keep the TDP to a certain minimum otherwise they will need a fan, and is one of the reasons why they have to balance power usage… Along with things like maintaining run time, etc.

    Really, tablets are sealed! They have no vents and rely on passive cooling which is very inefficient. Since, unless air is moving then it’s can act almost like a insulator as you have to wait for the warmed air to rise before cooler air replaces it and that’s a slow process and only gets worse if the ambient air temperature is close to the heat source temperature.

    Lithium batteries also don’t respond well to excess heat and the iPad is mostly battery! In addition to the fact the LEDs need to be kept relatively cool too.

    Besides, not all of the heat for the iPad 3 is coming from the screen. The lower left-hand side of the iPad, held in portrait mode, is where it gets the most noticably warmer.

    ” I’ll believe Intel’s power usage claims when I see it. Considering no
    smartphone runs on Intel chips now, obviously nobody believes them yet.
    ;)”

    There are couple of phones running on Intel already and benchmarks and tests have already been done. So you need to get up to date on the status of things!

    Early next year the Z2580 will raise the Medfield with dual core and a SGX544 based GMA that’s essentially comparable to the iPad 3’s SGX543MP4 but that’s before Intel starts rolling out it’s first next gen architecture for the ATOM by the later half of 2013.

    All ATOMs are going SoC from the 22nm update onward and they’ll be releasing a new version every single year for faster than Moore’s Law advancement rate.

    The 22nm will right off the bat offer up to 8GB of RAM support, support for newer standards like UBS 3.0, a Intel based GMA that at the very least will appeal better to Linux users than closed driver supported GPU’s like Imagination offers, among many other changes.

    If Intel can keep ahead of the FAB advancement curve then they stand a chance of being more than just competitive with ARM. While factors like whether desktop OS will start being used by mobile devices could determine which starts to get the edge.

    Time will tell, in either case the competition will be good for the market. Since a monopoly of either will not be good for the consumers.

  • Joakim

    OK, no point responding in detail when you cannot do simple math, ie aggregating 10-15% growth over 4 years to get 60% growth for a 4-year period, and keep blaming the ARM CPU in the new iPad for the power increase when it is well-known that it is the new retina display and backlight/GPU that is driving all the power increase. Apple just announced that their new A6 ARM CPU in the iPhone 5 actually uses less power than the previous one. You keep making these silly claims and then try to distract from how wrong you are by changing the subject to a whole bunch of completely unrelated issues, that you bring up for no reason, like whether it matters whether it’s a PC company that’s now successful in mobile or not. Obviously it’s a combination of a PC company like Apple and an established mobile company like Samsung using a new OS like Android, but why you would ever bring this up, who knows, other than to distract from all the things you’re wrong about by burying those issues we were actually discussing in a bunch of useless and irrelevant new issues that you bring up for no reason.

    I asked you twice to put up or shut up by predicting how well Windows 8 would do and both times you completely ignored my request, typical of how your responses are full of a bunch of irrelevant blather without addressing the issues we were actually talking about and why it’s not worth discussing this stuff with you.

  • CyberGusa

    “OK, no point responding in detail when you cannot do simple math, ie aggregating 10-15% growth over 4 years to get 60% growth for a 4-year period”

    Sorry but that’s just spinning the facts, and again all you’re doing is ignoring all the factors that lead to those figures and assuming it’s all the same from then to now when I’ve pointed to multiple things that makes the markets different and why they don’t directly correlate. So no, you’re the one trying to oversimplify and getting the math wrong!

    “Apple just announced that their new A6 ARM CPU in the iPhone 5 actually uses less power than the previous one.”

    Yes, because it’s being based on the newer Cortex A15 but that architecture isn’t going to be replaced for another 2 years!

    ” You keep making these silly claims and then try to distract from how
    wrong you are by changing the subject to a whole bunch of completely
    unrelated issues, that you bring up for no reason, like whether it
    matters whether it’s a PC company that’s now successful in mobile or
    not.”

    Wrong, those are relevant for actual comparison. You’re trying to tie the different markets together for a direct correlation of cause and effect, but everything I’ve stated clearly points out that’s absolutely wrong!

    Obviously, you must be in denial because only in the last year has there even started to be any real overlap between the mobile and PC markets. Previously ARM didn’t have the performance needed to run a desktop OS and even now still needs a successful desktop OS to really penetrate into the PC market.

    “I asked you twice to put up or shut up by predicting how well Windows 8
    would do and both times you completely ignored my request”

    Windows 8 won’t be a failure but it’s uncertain if it will help or hurt ARM because RT is not the same as full Windows 8. This isn’t something that can be accurately predicted until people actually try the device and see if the limitations are too much or not!

    What we’ve discussed is how the markets have actually evolved up to now and everything I’ve stated has been relevant whether you want to admit it or not!

  • Joakim

    Yet more hand-waving with irrelevant asides and just wrong details: the new A6 CPU is not based on ARM Cortex A15, it’s a new Apple design that likely has more in common with Cortex A9 than A15. Of course, your claim has nothing to do with with my original claim, for which the A6 was provided as evidence, that new ARM cores get more powerful and more energy-efficient. And it’s silly to say that mobile only started to really “overlap” the PC market in the last year. Many people were only using their PCs for email and web browsing, so for them a smartphone or tablet have been much better for years, they do what their PC did and are mobile! Nobody cares about “penetrating” the desktop PC market, as I said, that market is dying and it’s likely the desktop won’t be around soon.

    As one of my first links showed, there were 5 times as many smartphones and tablets sold last year as desktops, the desktop is about to be obsoleted. I also noted that the home server market is about to take off one of these days and if Intel were smart they would try to regroup there, but considering they have done nothing to help that market take off yet, they appear too dumb to do so. Not only that, Microsoft and Intel are so scared of Android that Intel is not supporting linux on their new Clover Trail chip, likely at Microsoft’s request, that tells you all you need to know about Wintel right there.

  • CyberGusa

    No hand-waving at all, things like market momentum and other changing economic factors are just facts of how the market works.

    You’re so called 60% was hand-waving. Ignoring that most of it didn’t happen until 2 years after the start of the recovery, which is indicative of people finally becoming confident enough to shop for what they wanted instead of saving for just what they needed.

    Really, you talk as someone who knows practically nothing on how the market works and are trying to draw conclusions from things that aren’t even related and then have the nerve to try to dismiss the actual factors involved because they don’t mess with your already decided conclusions.

    You can believe that the desktop is obsolete, but in truth it’s only the old version of it that is and the reality is the desktop will evolve like any other technology we use over time. It’ll become more invisible but it’s not really going away.

    While laptops as well are going to go through a change but it remains that to do actual work that mobile devices are not enough. Mobile still means compromise and as long as it does it will never replace other product categories.

    You’re understanding of technology is also apparently flawed if you think Clover Trail’s lack of Linux support has much to do with MS. Fact is GNU/Linux isn’t optimized for mobile usage like Android and now Windows 8.

    There’s also over 600 distros of linux, but the main problem is 32nm ATOMs are presently using Imagination PowerVR GPU’s, which have never really supported Linux and barely support their clients/licensees that well either.

    Intel for example only managed to release a barely fully functional 32bit Windows 7 drivers in time for the Cedar Trail release earlier this year. But since then Intel has had to focus on getting the Windows 8 drivers ready in time for launch and so there have been no more Windows 7 driver updates for the PowerVR based GMAs!

    So it really isn’t a conspiracy why Intel isn’t supporting general Linux with Cedar Trail. But they will run Android, as they already do with Medfield, and Windows 8.

  • Joakim

    A couple of last corrections to your characteristic mistakes: Android uses the linux kernel, so if Clover Trail won’t run linux, it won’t run Android. Not sure why you mention Intel not supporting linux on Cedar Trail at the end, presumably you meant Clover Trail there. Frankly, Intel’s move is pointless, as the power state support they use as their weak excuse will be quickly reverse-engineered by the linux folks and Android will run on Clover Trail. Presumably they’re only doing it to scare off OEMs like Acer or Asus from trying any dual-boot setup, because then they will be on their own and Intel will not provide any “support” for Android on Clover Trail. It has nothing to do with your example of PowerVR GPUs, which btw almost all run on linux since Android uses the linux kernel and all of those GPUs run with Android, or with not having time: it’s clearly just a political move, and since it doesn’t really benefit Intel, the clear implication is that Microsoft demanded it, because they’re scared of Android.

  • Joakim

    It looks like Intel has seen the blowback from their decision and are now relenting, they claim they will provide support for linux on Clover Trail later on, and that they have another x86 chip platform aimed at linux/Android that will come out later. Sounds like a lot of PR aimed at quieting criticism of whether x86 is really an open platform anymore when they pull moves like this. Frankly, the very fact that Microsoft would pull a move like this- it has to be them, this exclusion does not help Intel at all- speaks to how desperate they are to avoid straight up comparison to Android on the same hardware, which is pretty damning of how confident they are in Windows 8.

  • CyberGusa

    Sorry but Clover Trail is based on Medfield, which already runs Android. But sharing a Kernel does not make Android the same as desktop Linux distros. For one thing Android isn’t a GNU/Linux distro, the only thing they share is the kernel!

    Second, Google isn’t limited to Open Source like most Linux Distros. They are perfectly fine with having non-Open Source modifications made to Android and are also fully willing to deal with closed driver support.

    Third, it’s “Imagination” providing the driver support for the PowerVR based GMA’s. So blaming Intel ignores who they’re stuck working with right now.

    Really, do you see desktop Linux on a lot of ARM devices? No you don’t and there are reasons for that ranging from lack of Open Source support and hardware fragmentation to the use of Closed drivers from companies that don’t support Linux in general.

    Many Linux distros are maintained and developed by advocates of Open Source, while of the over 600 different distros only a handful even have the resources to make deals with closed driver hardware makers and there’s no guarantee drivers made for one distro will work for them all!

    Really, it takes more than a Kernel to run a OS!

  • Vincent Randal

    What I saw was a tablet with a keyboard. No real software for getting any real work done. Let us know when there’s an Android device that does the all the work a real notebook can do – running Ubuntu VM’s compiling Android OS from AOSP source. This Archos is a toy.