Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Microsoft prepares tablet OS, prays that someone will make a decent tablet processor

QuizzicalQuizzical Member LegendaryPosts: 25,346

As you likely know, Windows 8 has a number of design features targeted at tablets.  Microsoft is scheduled to launch it later this month.  But just because the OS is ready (and there are doubts about that as well) doesn't mean that the hardware will be.

For desktops and laptops, of course, Windows 8 should run on the same hardware as Windows 7.  That's not a problem.  This thread is about tablets.

Microsoft is trying to hedge its bets, with both an x86 version and an ARM version, with the latter called Windows RT.  Let's take them one at a time.

Only three companies have an x86 license, so Microsoft will have to hope that at least one of them makes a good tablet processor for Windows 8.  On the Intel side, we mainly have Clover Trail Atom.

This is basically a tablet variant of Cedar Trail Atom, which was such a colossal flop that Intel wanted no attention for it whatsoever.  The initial launch consisted of adding the parts to a list on Intel's web site, without so much as a press release.  The press release did come later, but not review samples, as Intel didn't want to get buried in a barrage of negative reviews for a part that lost to AMD's older Brazos platform in just about every way imaginable.

Intel also has ULV Ivy Bridge parts available.  But 17 W is a bad idea in a tablet.

On the bright side, Haswell is coming next year.  Intel is putting a heavy focus on bringing power consumption down in Haswell, and there will commonly be 10 W bins of it.  Haswell will probably bring impressive CPU performance for a 10 W TDP, and plenty of CPU performance for a tablet.

The trouble is, to get low power for a tablet, you need integrated graphics.  The next respectable graphics product that Intel launches will be their first.  And Haswell probably won't be it.  Intel has already announced that API support won't even catch up to aging Radeon HD 5000 and GeForce 400 series cards.  Haswell graphics are mostly derivative of Intel's last few disastrous generations (Intel HD anything).

Broadwell is supposedly going to bring a huge overhaul on the graphics side.  But that's not coming until 2014.  Besides, until Intel ships a good video driver for some older generation product, you shouldn't trust them to have a good one for a newer generation product.  We're getting close to 15 years of waiting on the former, and still no luck.

Next up, we have AMD.  Here, Hondo will be a killer product for tablets.  It sports two enhanced Bobcat cores on a 28 nm process node with AMD integrated graphics in a 4.5 W TDP.  It was also cancelled (along with Wichita and Krishna, the laptop parts) due to serial delays of Global Foundries' 28 nm process node.  Oops.

Rumor is that AMD has reappropriated the name and Hondo will be a 40 nm product that is basically the older Desna chip with some bits removed that aren't relevant to tablets, in order to save on power consumption.  AMD is supposedly going to launch something or other on October 9.  But a warmed-over version of a laptop chip that is nearly two years old isn't exactly ideal.

On the bright side, Temash is coming, and is far more optimized for tablets.  It will feature up to four Jaguar cores on a Global Foundries 28 nm process node, probably with GCN graphics.  But that's not until next year, so it won't be in time for Windows 8.

AMD also has Trinity for laptops, and could squeeze that into a tablet.  But as with Ivy Bridge, 17 W in a tablet is a bad idea.  AMD has a 19 W bin of Trinity that is a quad core with 6 SIMD engines.  Could they make a 10-12 W bin that is a dual core with three SIMD engines (currently being sold as A6 versions of Trinity)?  Likely.  Should they?  That's still quite a bit of power for a tablet, even if they could do it.

And then there is VIA.  Were you even aware that VIA made x86 processors?  VIA has a significant share of the market for ultra-low budget x86 processors in embedded systems, but nothing with the energy efficiency needed for a tablet.

So what about ARM?  There are problems here, too.  The ARM processors on the market are basically targeted at cell phones at best, or even lower power alternatives otherwise.  If you put a cell phone chip into a tablet, you get cell phone performance.  Tablets should be able to have bigger batteries and more heat dissipation than cell phones, which would allow for higher performance.  But ARM Cortex A9 doesn't get there.

ARM Cortex A15, on the other hand, does.  And lots of ARM partners have chips that use ARM Cortex A15 cores.  They've shown them off at trade shows, but everything seems to be delayed, and I'm not aware of any that are in commercial products yet.  There is the Samsung Exynos 5250, the ST-Ericsson Nova A9600, Nvidia Tegra 4, and TI OMAP something or other.

All coming--but all not here yet.  Some of them were supposed to be here by now, too.

Well, maybe not all coming.  TI is reportedly shutting down its OMAP division and cancelling everything.  Oops.

What about custom ARM cores based on the Cortex A15?  Well, there is the Qualcomm Krait.  Qualcomm showed it off earlier this year, and performance is impressive.  But it's still not here.  My guess is that the Qualcomm Krait will be the first of the good tablet processors to show up at retail, but I really don't know.

There is also Apple, which is tight-lipped about upcoming CPU plans.  The Apple A6 is even a custom chip, rather than a Samsung chip with an Apple label on it like previous generations.  But don't expect to see an Apple chip show up in a Windows RT device.

And it's not like ARM vendors are going to wait for Windows RT to launch products.  Google Android dominates the ARM phone/tablet market, and it's available today.  Furthermore, there isn't much reason to believe that Windows RT will change that.

One huge advantage that Windows 8 has is that it will run software designed for Windows 7, Vista, XP, and so forth.  There is a huge software base already in place.  For ARM chips, Apple and Google have a huge software base, and Windows RT has... nothing.  Now add to that a high price tag and rumors that potential partners are furious that Microsoft is launching their own tablets and may opt out of making Windows RT tablets at all.  The Windows brand name is universally recognized, but Windows RT doesn't really have anything going for it that various versions of Windows Phone didn't, and none of those turned out well.

«1

Comments

  • RidelynnRidelynn Member EpicPosts: 7,383

    Your under the assumption that tablets have to compete with PCs - especially in terms of GPU power.

    They don't. Intel graphics, as poor as they are, are still miles ahead of what tablets are using today - it's not all about the GPU, you've got to consider the total package - in a tablet it's the SOC, in a PC it's the APU (or CPU+GPU).

    The iPhone 5 A6 custom SOC is the hardware to beat these days (in a phone or tablet). It's hard to compare graphics scores from iOS to anything else (especially since Apple is closed-lipped about the specs).

    iPhone 5 common benchmarks
    http://www.pcmag.com/article2/0,2817,2410034,00.asp

    Tablets don't have to come close to competing with Desktops in terms of total power. They are always going to be extremely power constrained (who wants to carry about 5lbs of batteries in a hand-held device?), heat dissipation (no room for any fans or heatsinks, unless your doing it wrong) and input mechanisms (most tablets have some form of keyboard hookup, but your still limited to touch gestures for everything else - they are nice, but I miss my mouse sometimes). Even an ultrabook will be miles ahead of the tablet arena (if they are doing it right).

    Tablets don't have to compete with PCs, especially in horsepower. They need to complement them.

  • JetrpgJetrpg Member UncommonPosts: 2,347

    My question is tablet WHY.

    I look at a tablet as an ultra thin laptop with its screen turned around and glue to the base of itself. = <-- like that, oh and with a  touch screen.

    I don't know it seems like tablets want to be something they cannot currently be. Maybe the replacment for laptops , that are less functional and far more expensive. In otherwords, generally a bad idea unless your real rich.

    Windows 8 seem to be MS attept toi seem more high tech then apple, but its a lot of what no one wants, but few a few super rich (and they don't really care) and a few fanboies ... those fanboies never stop beign entertianing.

    This is central to the issue of CPU for them , what are they... how much power do you need on them? How energy efficiant do they have to be, etc?

    Its hard to answer these questions because their rather pointless currently.

    "Society in every state is a blessing, but government even in its best state is but a necessary evil; in its worst state an intolerable one ..." - Thomas Paine

  • QuizzicalQuizzical Member LegendaryPosts: 25,346

    What can you do with a tablet right now?  E-reader, sure.  You can read e-mail, and perhaps some light web browsing.  But they're not a credible replacement for a laptop, and adding a keyboard dock doesn't change that.

    But the performance that you can get in a tablet with a given level of power consumption is about to go way, way up.  Several chips are going to be an enormous advance over all currently available products.  And they're chips that were supposed to be out by now, but they all got delayed and/or cancelled.

    Right now, you can build a tablet around a cell phone chip, in which case, you get cell phone performance.  The larger form factor of a tablet should enable considerably better performance than that.  Alternatively, you can build a tablet around a laptop chip.  One problem here is that laptops need a ton of stuff that doesn't do anything other than waste power in a tablet.  A related problem is that many laptop chips can't have their power reduced far enough to be sensible for use in a tablet.

    That leaves a big hole in the middle.  It's a hole that is about to be filled, but the chips aren't here yet.

  • JetrpgJetrpg Member UncommonPosts: 2,347
    Thats my point.

    "Society in every state is a blessing, but government even in its best state is but a necessary evil; in its worst state an intolerable one ..." - Thomas Paine

  • CleffyCleffy Member RarePosts: 6,412

    Probably the stupidest thing I have heard about Windows 8 is this.  Windows 8 32-bit.

    On tablets, I wish they would make a new tablet PC with a laptop chip like an A10.  There has not be a graphically significant tablet PC in a long time which are basically tablets with a keyboard.  Last best one used a nVidia 6250.  Rest have used Intel IGP.

  • DOGMA1138DOGMA1138 Member UncommonPosts: 476
    Originally posted by Quizzical

    What can you do with a tablet right now?  E-reader, sure.  You can read e-mail, and perhaps some light web browsing.  But they're not a credible replacement for a laptop, and adding a keyboard dock doesn't change that.

    But the performance that you can get in a tablet with a given level of power consumption is about to go way, way up.  Several chips are going to be an enormous advance over all currently available products.  And they're chips that were supposed to be out by now, but they all got delayed and/or cancelled.

    Right now, you can build a tablet around a cell phone chip, in which case, you get cell phone performance.  The larger form factor of a tablet should enable considerably better performance than that.  Alternatively, you can build a tablet around a laptop chip.  One problem here is that laptops need a ton of stuff that doesn't do anything other than waste power in a tablet.  A related problem is that many laptop chips can't have their power reduced far enough to be sensible for use in a tablet.

    That leaves a big hole in the middle.  It's a hole that is about to be filled, but the chips aren't here yet.

    No sufficient for what? I've seen an ARM based tablet running an early build of Microsoft Office 2013 at ThinkNext this year and it was running it just aswell as any laptop would.

    There are Atom(and some ULV) based Windows 7 tables for quite a while, i toyed with a chinese one that could run some games pretty decently, played StarCraft 2 on it on low to medium as a joke for a while. The problem with those was the weight which was about as much as a UM notebook or a netbook at the time, and the branded ones were rather expensive.

    I'm running chrooted version of Ubuntu 12 on my Galaxy Note 2, and Galaxy Tab 2 on a daily bases mainly for office,  eclipse and gimp and it runs it very well, and considering that it's running it with the Android OS in the backround(or well inside it) its even more impresive.

    Windows 8 RT has quite low hardware requirements, and Windows 8 it self runs very well on machines with low amount of RAM and a limited proccessor, i did several trials with it on a VM and with 1GHZ proccessor, and 256MB of RAM using metro only apps it worked rather well..

     

     

  • DraemosDraemos Member UncommonPosts: 1,521
    Originally posted by Jetrpg

    My question is tablet WHY.

    I look at a tablet as an ultra thin laptop with its screen turned around and glue to the base of itself. = <-- like that, oh and with a  touch screen.

    I don't know it seems like tablets want to be something they cannot currently be. Maybe the replacment for laptops , that are less functional and far more expensive. In otherwords, generally a bad idea unless your real rich.

    Windows 8 seem to be MS attept toi seem more high tech then apple, but its a lot of what no one wants, but few a few super rich (and they don't really care) and a few fanboies ... those fanboies never stop beign entertianing.

    This is central to the issue of CPU for them , what are they... how much power do you need on them? How energy efficiant do they have to be, etc?

    Its hard to answer these questions because their rather pointless currently.

    Actually, a lot of people would love the convenience of a tablet with the power of a laptop.    I don't really know what makes you think MS is trying to appear more "high tech" than Apple, they're both tech companies that are highly geared towards innovation.  If Microsoft didn't do it, it's only a matter of time until Apple or Google did.

  • TheLizardbonesTheLizardbones Member CommonPosts: 10,910

    I have a Nook "tablet" and it is useful, but the limitation of what I can do isn't really in the speed of the processor. The limitations are all centered around the touch screen interface. Simple touch operations are better than trying to click with a mouse, but anything more complex becomes time consuming. I am a touch typist and I can type somewhere between 70 and a 100 words per minute, depending on what I'm doing. I would be amazed if I could type 20 words per minute on a tablet. Even if spreadsheets or software development environments were available on a tablet, it would be pointless for me to even try using them because it would be so slow and painful to use the touch screen interface to do those kinds of things.

    This is all from an Android perspective. I have no idea how Windows performs in a tablet environment. It could be painfully slow and might require a much more powerful processor than what is currently available. This would not surprise me. I think the tablets are still going to suffer from the interface they must use though.

    I can not remember winning or losing a single debate on the internet.

  • ShakyMoShakyMo Member CommonPosts: 7,207
    "killer processor"

    Helps if you have a light weight pained Android rather than what ever bloatware Microsoft are cooking up. Your processor doesn't need to be quite as killer then.
  • QuizzicalQuizzical Member LegendaryPosts: 25,346
    Originally posted by Cleffy

    Probably the stupidest thing I have heard about Windows 8 is this.  Windows 8 32-bit.

    On tablets, I wish they would make a new tablet PC with a laptop chip like an A10.  There has not be a graphically significant tablet PC in a long time which are basically tablets with a keyboard.  Last best one used a nVidia 6250.  Rest have used Intel IGP.

    Some businesses have very old proprietary software that won't run on a 64-bit OS and that they really don't want to scrap.  Rather than telling them to keep running Windows 7 forever or finding a way to get a 16-bit program to run on a 64-bit version of Windows 8 (I have no idea how hard this would be), Microsoft makes a 32-bit version of Windows 8.

    The trouble with a tablet based on an A10 APU is that a 25-35 W TDP is way, way too much for a tablet.  AMD's upcoming Temash chip will surely raise the performance level that you can find in a more tablet-friendly 4.5 W TDP.  I think Temash is supposed to have GCN graphics, though it will probably only be 1-2 compute units.

    I somewhat expect Nvidia's upcoming Tegra 4 to likewise have nifty graphics for a tablet; Internet rumors say it will support DirectX 11 and Open GL 4.x, which isn't something you'd implement if the graphics are only meant to display the desktop.  I couldn't find any source where Nvidia directly confirmed that, though, so I don't know if it's just speculation.

  • DOGMA1138DOGMA1138 Member UncommonPosts: 476

    If Microsoft cared enough for 16-bit support they would've upgraded their NTVDM to work in real 64bit mode to support X86-64 processors running in long mode, IA-64 processors can technically run it(and in the RC of Server 2008 for IA-64 they could) since they don't use  have  an " 8086 emulator"/legacy mode for 32/16bit support as the X86-64 ones do.

    I think the only reason that we still have Windows 8 32bit, is that Windows 8 RT is 32bit, and you still got enough corporate desktops out there that do not have 64bit processors.
     

  • RidelynnRidelynn Member EpicPosts: 7,383


    Originally posted by Quizzical
    What can you do with a tablet right now?  E-reader, sure.  You can read e-mail, and perhaps some light web browsing.  But they're not a credible replacement for a laptop, and adding a keyboard dock doesn't change that.

    I don't think they should be intended to be a laptop replacement. I think that is the mistake that Microsoft has been making with their tablets all along, since the very first tablets over 10 years ago.

    Laptops are meant to fully replace PC's, and be portable - for the most part. Tablets just need to supplement, or rather, compliment that.

    That's why tablets didn't take off until Apple came around - everyone was trying to use them to replace the PC. Apple decided to buck that trend, and use a tablet to compliment a PC. Email, e-books, web browsing make it a formidable portable appliance, but it doesn't need to compete with full-blown spreadsheets or data processing. Besides, that sort of heavy-duty computing is being offloaded to the cloud anyway, with just light interfaces to be able to interact with it (of which are fully capable of running on tablets, although the interface may still prove to be a challenge).

  • QuizzicalQuizzical Member LegendaryPosts: 25,346
    Originally posted by DOGMA1138
    Originally posted by Quizzical

    What can you do with a tablet right now?  E-reader, sure.  You can read e-mail, and perhaps some light web browsing.  But they're not a credible replacement for a laptop, and adding a keyboard dock doesn't change that.

    But the performance that you can get in a tablet with a given level of power consumption is about to go way, way up.  Several chips are going to be an enormous advance over all currently available products.  And they're chips that were supposed to be out by now, but they all got delayed and/or cancelled.

    Right now, you can build a tablet around a cell phone chip, in which case, you get cell phone performance.  The larger form factor of a tablet should enable considerably better performance than that.  Alternatively, you can build a tablet around a laptop chip.  One problem here is that laptops need a ton of stuff that doesn't do anything other than waste power in a tablet.  A related problem is that many laptop chips can't have their power reduced far enough to be sensible for use in a tablet.

    That leaves a big hole in the middle.  It's a hole that is about to be filled, but the chips aren't here yet.

    No sufficient for what? I've seen an ARM based tablet running an early build of Microsoft Office 2013 at ThinkNext this year and it was running it just aswell as any laptop would.

    There are Atom(and some ULV) based Windows 7 tables for quite a while, i toyed with a chinese one that could run some games pretty decently, played StarCraft 2 on it on low to medium as a joke for a while. The problem with those was the weight which was about as much as a UM notebook or a netbook at the time, and the branded ones were rather expensive.

    I'm running chrooted version of Ubuntu 12 on my Galaxy Note 2, and Galaxy Tab 2 on a daily bases mainly for office,  eclipse and gimp and it runs it very well, and considering that it's running it with the Android OS in the backround(or well inside it) its even more impresive.

    Windows 8 RT has quite low hardware requirements, and Windows 8 it self runs very well on machines with low amount of RAM and a limited proccessor, i did several trials with it on a VM and with 1GHZ proccessor, and 256MB of RAM using metro only apps it worked rather well..

    I've run Excel on a Core 2 Duo and Eclipse on an AMD E-350, and while they worked, performance was rather less than optimal.  Either of those on Atom or ARM Cortex A9 cores would have been just painful for what I do.  Maybe I have heavier-duty uses for them than you do (sometimes my desktop Core i7 chokes on Excel, too).  But then, those programs are what desktops are for, not laptops, let alone tablets.

    There's also the broader question of, all else equal, do you want more performance or less performance?  Because soon, you'll be able to get a lot more performance.  It's not just a new generation of hardware, but a combination of a full node die shrink, adding HKMG, and having the first architectures designed for tablets all at once.

  • QuizzicalQuizzical Member LegendaryPosts: 25,346
    Originally posted by Draemos
    Originally posted by Jetrpg

    My question is tablet WHY.

    I look at a tablet as an ultra thin laptop with its screen turned around and glue to the base of itself. = <-- like that, oh and with a  touch screen.

    I don't know it seems like tablets want to be something they cannot currently be. Maybe the replacment for laptops , that are less functional and far more expensive. In otherwords, generally a bad idea unless your real rich.

    Windows 8 seem to be MS attept toi seem more high tech then apple, but its a lot of what no one wants, but few a few super rich (and they don't really care) and a few fanboies ... those fanboies never stop beign entertianing.

    This is central to the issue of CPU for them , what are they... how much power do you need on them? How energy efficiant do they have to be, etc?

    Its hard to answer these questions because their rather pointless currently.

    Actually, a lot of people would love the convenience of a tablet with the power of a laptop.    I don't really know what makes you think MS is trying to appear more "high tech" than Apple, they're both tech companies that are highly geared towards innovation.  If Microsoft didn't do it, it's only a matter of time until Apple or Google did.

    And soon, you'll be able to have exactly that.  It will only be tablets that can hang with relatively slower laptops, but if you add a keyboard dock, it will be able to handle most (but not all) of the things that you might sensibly use a laptop for.

  • QuizzicalQuizzical Member LegendaryPosts: 25,346
    Originally posted by Ridelynn

    Laptops are meant to fully replace PC's, and be portable - for the most part. Tablets just need to supplement, or rather, compliment that.

    That's why tablets didn't take off until Apple came around - everyone was trying to use them to replace the PC. Apple decided to buck that trend, and use a tablet to compliment a PC. Email, e-books, web browsing make it a formidable portable appliance, but it doesn't need to compete with full-blown spreadsheets or data processing. Besides, that sort of heavy-duty computing is being offloaded to the cloud anyway, with just light interfaces to be able to interact with it (of which are fully capable of running on tablets, although the interface may still prove to be a challenge).

    Whether or not laptops are meant to fully replace desktops, they're terrible at it.  A tablet that could replace a laptop if you add a keyboard dock is a lot more plausible than a laptop that could replace a desktop.  It's not there yet, and laptops will always have advantages over tablets, but they're going to get narrower soon.  And it's actually the same problems in both cases:  performance, input options, upgrade/repair options, and so forth.

    As for Apple and the iPad, look at their ads.  They say, look at all of these cool things that you can do with an iPad.  They're things that you never wanted to do before seeing the ad, and if you stop to think about it after seeing the ad, they're things that you still don't want to do and wouldn't do even if you bought an iPad.  But Apple gets people to focus on what iPads can do, rather than what they can't.  That's the key to their success.

  • DOGMA1138DOGMA1138 Member UncommonPosts: 476

     


    Originally posted by Quizzical

    Originally posted by DOGMA1138

    Originally posted by Quizzical What can you do with a tablet right now?  E-reader, sure.  You can read e-mail, and perhaps some light web browsing.  But they're not a credible replacement for a laptop, and adding a keyboard dock doesn't change that. But the performance that you can get in a tablet with a given level of power consumption is about to go way, way up.  Several chips are going to be an enormous advance over all currently available products.  And they're chips that were supposed to be out by now, but they all got delayed and/or cancelled. Right now, you can build a tablet around a cell phone chip, in which case, you get cell phone performance.  The larger form factor of a tablet should enable considerably better performance than that.  Alternatively, you can build a tablet around a laptop chip.  One problem here is that laptops need a ton of stuff that doesn't do anything other than waste power in a tablet.  A related problem is that many laptop chips can't have their power reduced far enough to be sensible for use in a tablet. That leaves a big hole in the middle.  It's a hole that is about to be filled, but the chips aren't here yet.
    No sufficient for what? I've seen an ARM based tablet running an early build of Microsoft Office 2013 at ThinkNext this year and it was running it just aswell as any laptop would. There are Atom(and some ULV) based Windows 7 tables for quite a while, i toyed with a chinese one that could run some games pretty decently, played StarCraft 2 on it on low to medium as a joke for a while. The problem with those was the weight which was about as much as a UM notebook or a netbook at the time, and the branded ones were rather expensive. I'm running chrooted version of Ubuntu 12 on my Galaxy Note 2, and Galaxy Tab 2 on a daily bases mainly for office,  eclipse and gimp and it runs it very well, and considering that it's running it with the Android OS in the backround(or well inside it) its even more impresive. Windows 8 RT has quite low hardware requirements, and Windows 8 it self runs very well on machines with low amount of RAM and a limited proccessor, i did several trials with it on a VM and with 1GHZ proccessor, and 256MB of RAM using metro only apps it worked rather well..
    I've run Excel on a Core 2 Duo and Eclipse on an AMD E-350, and while they worked, performance was rather less than optimal.  Either of those on Atom or ARM Cortex A9 cores would have been just painful for what I do.  Maybe I have heavier-duty uses for them than you do (sometimes my desktop Core i7 chokes on Excel, too).  But then, those programs are what desktops are for, not laptops, let alone tablets. There's also the broader question of, all else equal, do you want more performance or less performance?  Because soon, you'll be able to get a lot more performance.  It's not just a new generation of hardware, but a combination of a full node die shrink, adding HKMG, and having the first architectures designed for tablets all at once.
    For things I'm comfortable doing on a tablet the performance suits me very well, it was an office 2013 for Windows RT so it was optimized, and although it was not a 20MB excel file with data connections, and more lines of VB than the base source of Windows NT was it still worked very well for you're every day office excel sheets. 

    Eclipse runs very well on my ARM tablet, or padphone, i'm not going to compile or debug applications on it, i use it when i need to review source code on the go, and i don't see my self actually using it in any other capacity on a tablet-like device even if i could.

    As for the hardware buzz words, you don't really need "die" shrinks, or going High-K to facilitate better performance in laptops, I've seen chips that go into ElOp seeking missiles that operate at much high performance envelopes than what is commercially available today, and not they were not FPGA's or dedicated DSP's.

    There is a reason why Intel did not license ARM, for them its easier to bring their processors to the thermal envelope of ARM SOC's than to bring ARM SOC's to what they consider to be the performance level of what they consider to be a desktop processor.

    I have quite a few friends that work at Intel's IDC in Haifa, i've seen quite a few cool things but don't count on Intel bringing what you think, or "want" to mobile platforms any time soon, the market does not want it, and well does not need it. I got one of the "Intel Phones" couple a months ago and it works well(and funny enough overheats less than my Samsung Nexus, or SGS2 even tho it is running on ATOM), but their mobile strategy and you'll see it over the coming months is moving towards SOC based solutions for mobile platforms and rightly so.

    And honestly if you ask me if i rather have 50% more performance, or 50% more battery life for a mobile device ill take the battery life any time, if you don't you really need to reconsider what are you using a tablet for if you have access to a power socket ;) And if you increase your PPW for a mobile platform it's much more sensible to have a longer battery life than allowing some one to run a much more demanding application on their mobile product.

     

    Edit: since i don't really give a squat about buzzwords and product lines i actually had to check why did you trash Ceder Trail so badly, and i don't see a reason it's hard to find a bad review about it, it the N2800 and N2500 give about the same performance as the E-350/A-450 with nearly half the power consumption, not bad for a stop gap product which was put out only to test the core architecture for their SOC line. As i said don't expect those to improve like you think they will, Intel is going SOC for mobile platforms(AMD will eventually also once they figure out how to get there), and Desktop LV-ULV for Laptops..

     

     

     

     

     

  • QuizzicalQuizzical Member LegendaryPosts: 25,346
    Originally posted by DOGMA1138

    For things I'm comfortable doing on a tablet the performance suits me very well, it was an office 2013 for Windows RT so it was optimized, and although it was not a 20MB excel file with data connections, and more lines of VB than the base source of Windows NT was it still worked very well for you're every day office excel sheets. 

     

    Eclipse runs very well on my ARM tablet, or padphone, i'm not going to compile or debug applications on it, i use it when i need to review source code on the go, and i don't see my self actually using it in any other capacity on a tablet-like device even if i could.

    As for the hardware buzz words, you don't really need "die" shrinks, or going High-K to facilitate better performance in laptops, I've seen chips that go into ElOp seeking missiles that operate at much high performance envelopes than what is commercially available today, and not they were not FPGA's or dedicated DSP's.

    There is a reason why Intel did not license ARM, for them its easier to bring their processors to the thermal envelope of ARM SOC's than to bring ARM SOC's to what they consider to be the performance level of what they consider to be a desktop processor.

    I have quite a few friends that work at Intel's IDC in Haifa, i've seen quite a few cool things but don't count on Intel bringing what you think, or "want" to mobile platforms any time soon, the market does not want it, and well does not need it. I got one of the "Intel Phones" couple a months ago and it works well(and funny enough overheats less than my Samsung Nexus, or SGS2 even tho it is running on ATOM), but their mobile strategy and you'll see it over the coming months is moving towards SOC based solutions for mobile platforms and rightly so.

    And honestly if you ask me if i rather have 50% more performance, or 50% more battery life for a mobile device ill take the battery life any time, if you don't you really need to reconsider what are you using a tablet for if you have access to a power socket ;) And if you increase your PPW for a mobile platform it's much more sensible to have a longer battery life than allowing some one to run a much more demanding application on their mobile product.

     

    Edit: since i don't really give a squat about buzzwords and product lines i actually had to check why did you trash Ceder Trail so badly, and i don't see a reason it's hard to find a bad review about it, it the N2800 and N2500 give about the same performance as the E-350/A-450 with nearly half the power consumption, not bad for a stop gap product which was put out only to test the core architecture for their SOC line. As i said don't expect those to improve like you think they will, Intel is going SOC for mobile platforms(AMD will eventually also once they figure out how to get there), and Desktop LV-ULV for Laptops..

    If all you're using Eclipse for is a text-reading program, then yeah, a tablet should be fine for that.  That's hardly a replacement for using Eclipse on a desktop for what it's really meant for, though.

    Die shrinks are what drives Moore's Law.  Do a full node die shrink and you can have twice as many transistors as before, with each using only about 70% as much power as before.  That gives you your choice of higher performance, lower power consumption, more features (which can mean reducing power consumption by combining things into fewer chips), or some of everything.  And that's always a big deal.  While you really need architectural changes to fully exploit what a new process node can do, even a simple die shrink is a big deal.

    Well of course Intel isn't going to license ARM for desktops and laptops.  If ARM becomes the dominant architecture, then Intel is just one of many processor companies.  So long as it's x86, Intel only has to worry about AMD and VIA.  And they don't have to worry very much about VIA.  Intel likes being the only one with the patents to produce an architecture, or at least, to have as little competition as possible.  That's why they foisted Itanium on the world.

    Well of course Intel's mobile strategy is moving toward SoCs.  That's part of optimizing chips for cell phones and tablets.  One major reason that earlier versions of Atom weren't able to convince any cell phone companies to bite is that they weren't SoCs.  How many SATA ports do you need in a tablet?  How about USB ports?  PCI Express lanes?  Put as many of them in a tablet as you would a desktop and they just sit there wasting die space and power without doing anything productive.  Having more chips than necessary having to communicate with each other likewise wastes space and power.

    You know what's better than having your choice of 50% more performance or 50% more battery life?  Having both.  In a desktop, it doesn't matter if a processor is burning a few watts at idle.  In a laptop, it's far from ideal, but not catastrophic.  In a tablet, it's a big problem.  Intel is claiming that Haswell will have only 5% of the idle power consumption of Ivy Bridge, for a variety of reasons.  AMD says that Jaguar cores will be able to shut down 98% of the core at idle, while Bobcat cores could only shut down 91%.  Changes like that are what you need in order to make a chip suitable for tablets.

    Meanwhile, except for games or watching videos, a processor is nearly always idle.  Battery life is thus mostly determined by idle power consumption.  To get more performance, you maybe burn more power during the brief periods when the processor is active.  But even this doesn't have to be that much; if double the power consumption while active is compensated for by finishing the task in half the time so that the processor is only active for half as long, then it burns the same energy as before.  That's perhaps overly optimistic (more performance -> higher clock speeds -> higher voltages -> worse energy efficiency), but there is some of that feedback effect.

    As for Cedar Trail, there aren't that many reviews of them, as Intel didn't send them out for media samples.  It roughly hangs with an AMD E-350 in well-threaded processor benchmarks, but it needs programs that scale well to four threads to fully exploit the processor.  Atom gains a lot more from hyperthreading than Intel's desktop architectures because Atom's scheduler has a lot more holes where a given thread wouldn't use a core.  E-350 only needs two threads to put both cores to good use, and completely crushes Atom in single-threaded performance.

    E-350 has an 18 W TDP, while Cedar Trail Atom is 10 W.  But about half of the E-350's TDP is for the GPU; my own testing found about a 7 W difference in total system power consumption between Prime95 and completely idle--and that's not all due to the processor, either.  Meanwhile, Atom barely allocates any of that power to the GPU, which is why E-350 has a GPU that easily gives several times the performance of Cedar Trail Atom.  I don't know if the new Atom can handle the normal Windows desktop, but previous versions of Atom couldn't.

    But if you're looking for energy efficiency, then the chip to compare isn't E-350, but Z-01.  That gets you 5/8 of the CPU performance and 1/2 of the GPU performance in less than 1/3 of the TDP.  That will hang with Cedar Trail Atom in CPU performance with one or two threads, still completely crush it in GPU performance, and while using less power.  Toss in that Brazos is cheaper than Atom, too, and you have a complete slaughter.  Atom's only real advantage is slightly lower idle power consumption.  No wonder Intel didn't send out media samples.

    If you take a chip that gets so throughly outclassed in laptops and put it in a tablet, would you expect performance (by any efficiency metric, not just raw performance) to suddenly be awesome?  I wouldn't.

  • WizardryWizardry Member LegendaryPosts: 19,332

    I am most certainly not an avid follower of IOS platforms but this seems a little behind the times?

    I am assuming this is to compete versus the Apple foundation of IOS devices?They already have the TABLET and there is already proven software in amny ways superior to what large developers are producing.

    Itablet Citadel  ONE guy made this whole map.Yes there will be limitations but i am ok with zoning into another map to make a complete world,i don't need seamless worlds,i just want GOOD  well done worlds.

    No question ,at least to me that Epic Games houses the best minds in gaming,however this tech proves how it is already here and done better than some of our best MMORPG's.

     

    Never forget 3 mile Island and never trust a government official or company spokesman.

  • DeivosDeivos Member EpicPosts: 3,692
    One person might have ported Epic Citadel for android, but more then one person made Epic Citadel. :p

    "The knowledge of the theory of logic has no tendency whatever to make men good reasoners." - Thomas B. Macaulay

    "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge." - Daniel J. Boorstin

  • DOGMA1138DOGMA1138 Member UncommonPosts: 476
    Originally posted by Quizzical
    Originally posted by DOGMA1138
    WALL OF TEXT
    WALL OF TEXT 2.0

     

     

    Not quoting this, too long(did not read it all i think too confusing) and buzzwords are killing meh...

    edit: bleh some one commented had to quote...

    You're view or Moores law is way too text book, die shrinks do not work as simply as that, since SiO replacements methods do no transfer immediately to smaller scale fabs, also many aspects of the die cannot be minimized, control asics, cache, and other elements do not scale as much as the transistors, also when you see a 32NM fab process it does not means the smallest gate transistor is 32NM


    And again, you can't have both 50% lower power consumption and performance, and if some how you could i would take 100% longer batter life, over a 50% 50% hybrid since it suits me for what i need from a tablet today.

    Also the "NM" does not matter, the number of nano meters in the fabrication process actually has nothing to do with the size of the transistors ;)

    The XX  nano meters you read on the press release is the average distance between identical discrete SRAM elements, that's it, In Intel's 65NM process for example, the average gate size of the MOFSET transistor about 30-35NM depending on point of time in the cycle, while the overall length of the transistor was around 160-180nm.

    High school computer science was quite long time ago but power consumption of the processor is more dependent on the design of the processor it self(e.g. implementation of sleeping gates), and the chemical process of the silica it
    self(strain, doping, oxidation etc`). 

    When any semi conductor manufacturer switches to it does it to increase the yield per wafer, in many cases actually have the same power consumption, or even higher one due to leakage, and not being to implement some chemical process into the higher resolution lithography.

    If you look at mil-spec chips some of them are still made at the 300nm range, with most of them today made around the 120-180 line of processes, yet they achieve much better power consumption, noise / EM internecine resistance, and even higher logical density in some cases than commercial IC's, when you care less about the cost per unit, and can afford to stick with a single process and perfect it you get much much better results. Intel has no reason to do so, since if can shrink your die size from 10mm to 5, you get 4 times more cores per wafer(based on a 300mm one).

    The N2800 has a 6.5W and the N2600 has a 3.5W TDP, also TDP is not power consumption so not sure why you mentioned it, the N2800 has a power consumption of about 8-10W when its under high load, the N2600 has about 5-7W, thats the one they released for tablets, and ultra thin netbooks, the D series was meant for low power desktops, and cheaper stuff, and has 12-14W power consumption.
    Furthermore i don't understand why are you making excuses for AMD? do you really care why common applications are running on Atom better? Also you are quite wrong on the whole scheduler part, the Atom is actually much worse, or at least was, it is not a CISC processor like the standard x86's are, its a RISC processor with an internal translator, HT is implemented in it quite differently than in the P6/Core I what ever line or processors.

    And again honestly you're view of the whole market is way different than what most consumer want, or need, i do not need a tablet i can play Crysis on, even if i had one i would not do it, if i have enough time to play Crysis ill do it on my PC, Laptop(Gaming one), or Console.

    Tablets, Smartphones, Laptops/Desktops, have their own role, and there is no need for any more overlap than the basic ability to access my data, and my basic workspace.
    Any improvements in a mobile SoC should go to making it smaller, and more power efficient, if you can gain passive performance advantages than be it, but don't expect any one to work in shoving a Core I7 into one any time soon, and by the time they will be able too the performance difference between that and what you'll have on mobile will remain pretty much the same.

     

     

  • GrayGhost79GrayGhost79 Member UncommonPosts: 4,775
    AMD has a prototype tablet using a trinity A6 APU running windows 8. As of right now nothing really looks to come close to matching it in performance. I believe it was built by Compal if you want to check it out.
  • QuizzicalQuizzical Member LegendaryPosts: 25,346
    Originally posted by Wizardry

    I am most certainly not an avid follower of IOS platforms but this seems a little behind the times?

    I am assuming this is to compete versus the Apple foundation of IOS devices?They already have the TABLET and there is already proven software in amny ways superior to what large developers are producing.

    Itablet Citadel  ONE guy made this whole map.Yes there will be limitations but i am ok with zoning into another map to make a complete world,i don't need seamless worlds,i just want GOOD  well done worlds.

    No question ,at least to me that Epic Games houses the best minds in gaming,however this tech proves how it is already here and done better than some of our best MMORPG's.

     

    Wait until you see what the next generation of tablet chips can do.  It's going to be a huge jump when you have chips actually meant for tablets, rather than just using a cell phone chip (not enough performance) or a laptop chip (too much power consumption).

    As for operating systems, I expect Apple to eventually be reduced to a niche player, just as happened with desktops and laptops and is far along the way in happening with cell phones.  Apple brings some important innovations, but then within a few years, competitors are able to match them with cheaper prices and more open platforms.  I somewhat expect Google Android to end up as the dominant tablet OS, for basically the same reasons that it has it has become the dominant cell phone OS.

    Windows 8 tablets will also occupy a niche, with their main selling point that they can run all of your normal Windows hardware.  You can use it like a tablet for web browsing, then plug in a keyboard and use it like a laptop and run all of your normal laptop software.  As some people keep insisting, that's not really what people want in a tablet, and to a considerable degree, that's true.  That's why Windows 8 tablets will only occupy a niche (for the people who want a tablet that is also a functional laptop), and the vast amount of software won't make Windows 8 the dominant tablet OS.

    Wandering off in a different direction, I think Nvidia has ambitions about Tegra-based laptops, eventually with discrete GeForce cards.  They haven't announced it, and it's really just speculation on my part.  But they're a graphics company, and need to survive in a world in which integrated graphics has already eaten up the low end of graphics and is moving toward eating up the mid-range.  And now that integrated graphics are built into the same chip as the CPU, you can't do integrated graphics if you don't do the processor.

  • QuizzicalQuizzical Member LegendaryPosts: 25,346
    Originally posted by GrayGhost79
    AMD has a prototype tablet using a trinity A6 APU running windows 8. As of right now nothing really looks to come close to matching it in performance. I believe it was built by Compal if you want to check it out.

    ULV Ivy Bridge crushes it in CPU performance, though it loses badly in GPU performance.  But they both suffer from the same problems:  17 W is really too much for a tablet.  And they both have a bunch of bits burning power without doing anything productive in a tablet.

  • GrayGhost79GrayGhost79 Member UncommonPosts: 4,775
    Originally posted by Quizzical
    Originally posted by GrayGhost79
    AMD has a prototype tablet using a trinity A6 APU running windows 8. As of right now nothing really looks to come close to matching it in performance. I believe it was built by Compal if you want to check it out.

    ULV Ivy Bridge crushes it in CPU performance, though it loses badly in GPU performance.  But they both suffer from the same problems:  17 W is really too much for a tablet.  And they both have a bunch of bits burning power without doing anything productive in a tablet.

    The battery life was actually pretty good. The GPU performance is why nothing comes close and the GPU is the biggest limiting factor with tablets. 

    I'm not saying there isn't a lot of room for improvement but I truly think the Llano/Trinity route is the best hope for viable tablet CPU's or rather the APU route in general. 

  • RidelynnRidelynn Member EpicPosts: 7,383


    Originally posted by Quizzical
    Originally posted by Wizardry I am most certainly not an avid follower of IOS platforms but this seems a little behind the times? I am assuming this is to compete versus the Apple foundation of IOS devices?They already have the TABLET and there is already proven software in amny ways superior to what large developers are producing. Itablet Citadel  ONE guy made this whole map.Yes there will be limitations but i am ok with zoning into another map to make a complete world,i don't need seamless worlds,i just want GOOD  well done worlds. No question ,at least to me that Epic Games houses the best minds in gaming,however this tech proves how it is already here and done better than some of our best MMORPG's.  
    Wait until you see what the next generation of tablet chips can do.  It's going to be a huge jump when you have chips actually meant for tablets, rather than just using a cell phone chip (not enough performance) or a laptop chip (too much power consumption).

    I have an original iPad. I have a not-insignificant list of gripes about it, but really, "Not fast enough" is not one of those gripes.

    I don't think the "next generation of tablet chips" is what the tablet revolution is waiting for to take off. The tablet revolution has already happened. It wasn't CPU horsepower that made it happen - it was a shift in thinking about the purpose of a tablet. It had failed for years because everyone just assumed that tablets would replace a laptop and take over many of the same functions.

    Apple may or may not be a significant player in the future - they don't really have a great track record of maintaining a lead in anything, but they have an outstanding record of innovating and capitalizing on that innovation (oddly enough, that also was tied directly to SPJ, and he's no longer there to drive that, so time will tell if anyone else can pick that up). That doesn't really have anything to do with the future of the tablet - it has everything to do with the fact that Apple single-handedly defined the tablet, and what consumers want to do with one (although it took them two tries, the Newton didn't quite have the technology yet). A smart phone processor proved to be more than enough horsepower, and a smartphone OS more than capable enough (from both Apple and Google) - what we are seeing now are just refinements and iterations.

    Microsoft may be able to catch up in the same manner they typically do - by throwing enough money and resources at it that they eventually get on the boat (Office and Xbox both were years behind their counterparts, but were able to become market leaders with this philosophy) - but that hasn't always worked for them. If they continue to try to push "Replacement for the PC" - which they may be doing with their general purpose OS - they will continue to lose, and lose badly no matter how much capitol or resources they throw behind it (ala Bob or Zune).

Sign In or Register to comment.