Digitimes: Windows-on-ARM platform to join notebook competition in June 2013

cbn

Lifer
Mar 27, 2009
12,968
221
106
http://www.digitimes.com/news/a20111128PD225.html

Windows-on-ARM platform to join notebook competition in June 2013

Monica Chen, Taipei; Joseph Tsai, DIGITIMES [Tuesday 29 November 2011]

Windows on ARM (WoA), a combination of Windows 8 and ARM-based processors, is expected to make an official appearance at the end of 2012 and will try to compete in the notebook market as soon as June 2013, according to sources from notebook vendors.

Since players such as Nvidia and Qualcomm have been enhancing their ARM-based processors' power consumption and performance, if their processors can successfully pair up with Windows 8 and receive Windows software support, the WoA platform may soon be able to compete against Intel and AMD.

The sources pointed out that players with ARM-based processors are aggressive about WoA platform and are hoping that the platform will be able to raise their share in the tablet PC market as well as help them enter the notebook industry that has been dominated by Wintel.

The ARM CPU players are already aggressively cooperating with notebook players such as Asustek Computer and Lenovo and are set to launch WoA-based notebooks to test the water in mid-2013 with expectations to see the platform take off in 2014 and further grab share from Wintel in 2015 to become the second platform of the notebook market

The sources believe that WoA platform's advantages over low power consumption and price will provide strong competitiveness, but its biggest problems will be software support and cooperation with notebook vendors. If both problems can be resolved, the platform is expected to received strong attention from notebook players, especially second-tier and white-box players.

However, since Intel will also launch its 22nm Ivy Bridge processors that consume less power than previous generation CPUs, have quick response and stronger security and will launch Haswell-based processors in 2013 with even more advanced designs, competition between the two camps will become the main focuses of the Windows 8 generation.

Maybe we will see the first ARM Windows Notebooks powered by 64 bit ARMv8 SOCs?
 

cbn

Lifer
Mar 27, 2009
12,968
221
106
It shouldn't be long before Google follows suit with their own Android Laptop OS (besides Chrome OS).

Now all we need is a good open Laptop Chassis standard to accommodate all these relatively small SOCs.

Can this work?

Or do we need Intel to step in and work something up for the industry?

An atom SOC or perhaps a 22nm Celeron would fit in any space an ARMv8 SOC would fit. Furthermore, this would allow Intel to breakaway from MS and offer Meego or Android Laptop OSes for their own chips.
 
Last edited:

Cerb

Elite Member
Aug 26, 2000
17,484
33
86
Can this work?
Not no, but Hell no. it doesn't even work with x86. The device won't necessarily need to be tied to an OS, though. You can run different OSes on your PC, and there only reason that would be more difficult on upcoming tablets and notebooks running ARM SoCs would be poor documentation and/or driver support.
 

cbn

Lifer
Mar 27, 2009
12,968
221
106
Not no, but Hell no. it doesn't even work with x86.

Yes, Intel tried a standardized Laptop platform back in 2008 called "Rich Creek 2"---> http://hexus.net/business/news/channel/12788-intel-rich-creek-2-pictured/

I just wonder if the lack of success was due to the fact the standardized laptop platform was tied to the Windows Operating system.

With Larger OEMs like HP and Dell getting discounts on MS Operating systems I can't see the cost scaling for small system builders. I think the Operating system would have to be free (or included in the price of the hardware) to make Notebook chassis standard work.
 

Cerb

Elite Member
Aug 26, 2000
17,484
33
86
Yes, Intel tried a standardized Laptop platform back in 2008 called "Rich Creek 2"---> http://hexus.net/business/news/channel/12788-intel-rich-creek-2-pictured/

I just wonder if the lack of success was due to the fact the standardized laptop platform was tied to the Windows Operating system.
No, it's the physical needs, feature wants, and branding wants. If you can't tell the vendor from 10-20 feet away, the big Western companies would consider that a design failure, unless they copied a competitor's look (HP and Dell have been known to do that from each others' business notebooks).

Then, companies love being able to charge extra for proprietary accessories, like docking stations and batteries.

Finally, notebooks, even today, try to cram a ton of functionality into a small space, and standardized design would necessarily increase that space. Consumers want smaller and lighter, so no go (your notebook must be bigger and heavier to support many different CPUs and GPUs, compared to designing it for a specific range). Standardization only needs to work within a narrow product series. Since everyone uses the same standards for most IO (SD, ExpressCard, VGA, DVI, DP, HDMI, USB, RJ45, etc.), it works out just fine.

Standardization has pros and cons, and the cons outweigh the pros in notebooks' mechanical and electrical design, just as they do phones. However, operating systems do not need to be part of that non-standardization. It can be that way, but that is out of laziness (or apathy, or specifically wanting to brand a product via the OS), not because the hardware platform needs something that only Windows can provide.
With Larger OEMs like HP and Dell getting discounts on MS Operating systems I can't see the cost scaling for small system builders. I think the Operating system would have to be free (or included in the price of the hardware) to make Notebook chassis standard work.
Free OSes work in non-standard notebooks. Also, smaller vendors do exist, though are more prominent in Europe than the US, and they do cost more. They typically have some specific value not provided by most vendors (a fully featured laptop guaranteed to run any decent Linux distro, FI, so you don't have to research your hardware specs). Smaller vendors indeed cannot compete with $350 big vendor notebooks.

You may have to research a little to be sure that you'll have sufficient HW support, since Windows support is assumed, but that's generally not too much additional time to take. It's only a little bit harder than with a desktop PC, really (with a desktop, you can add a card if the situation is bad enough, typically).

While 3D support remains behind on AMD's part, AMD and Intel are plenty good enough about either creating and maintaining Linux support (which is enough for other FOSS OSes to re-implement, if they desire), or offering sufficient documentation for users to do a good job of it. 3rd party chips are where it can get iffy (OTOH, I've seen my fair share of POS driver support on Windows for no-name chips, such that I'd rather just know it's a brick before tracing BSODs to that HW).

Given common lack of future driver support under Windows by some companies, such research is even warranted for Windows (if you have to use the vendor for any new drivers, instead of the chip maker, consider that hardware a big if).
 
Last edited:

cbn

Lifer
Mar 27, 2009
12,968
221
106
No, it's the physical needs, feature wants, and branding wants. If you can't tell the vendor from 10-20 feet away, the big Western companies would consider that a design failure, unless they copied a competitor's look (HP and Dell have been known to do that from each others' business notebooks).

Then, companies love being able to charge extra for proprietary accessories, like docking stations and batteries.

Finally, notebooks, even today, try to cram a ton of functionality into a small space, and standardized design would necessarily increase that space. Consumers want smaller and lighter, so no go (your notebook must be bigger and heavier to support many different CPUs and GPUs, compared to designing it for a specific range). Standardization only needs to work within a narrow product series. Since everyone uses the same standards for most IO (SD, ExpressCard, VGA, DVI, DP, HDMI, USB, RJ45, etc.), it works out just fine.

Standardization has pros and cons, and the cons outweigh the pros in notebooks' mechanical and electrical design, just as they do phones. However, operating systems do not need to be part of that non-standardization. It can be that way, but that is out of laziness, not because the hardware platform needs something that only Windows can provide.

Yes, but netbooks are practically dead anyway. Furthermore, many of them already shared a common chassis for several years.

Making them cheaper to own/upgrade could breath life into this dying segment that many OEMs would probably like to abandon.

In fact, that might be a contributing reason why "Rich Creek 2" failed. It focused on the higher end Intel processor laptops rather than on the slower moving product lines like Atom.
 
Last edited:

cbn

Lifer
Mar 27, 2009
12,968
221
106
Anyone have opinions on how the ARMv8 Notebooks will pan out in terms of performance?

Will MS use a lighter codebase making them seem more surefooted and responsive compared to a bobcat or Atom notebook?

Or will we see a Windows 8 port that actually penalizes the CPU compared to an Android or Apple ARMv8 Laptop operating system?
 

cbn

Lifer
Mar 27, 2009
12,968
221
106
You may have to research a little to be sure that you'll have sufficient HW support, since Windows support is assumed, but that's generally not too much additional time to take. It's only a little bit harder than with a desktop PC, really (with a desktop, you can add a card if the situation is bad enough, typically).

While 3D support remains behind on AMD's part, AMD and Intel are plenty good enough about either creating and maintaining Linux support (which is enough for other FOSS OSes to re-implement, if they desire), or offering sufficient documentation for users to do a good job of it. 3rd party chips are where it can get iffy (OTOH, I've seen my fair share of POS driver support on Windows for no-name chips, such that I'd rather just know it's a brick before tracing BSODs to that HW).

Given common lack of future driver support under Windows by some companies, such research is even warranted for Windows (if you have to use the vendor for any new drivers, instead of the chip maker, consider that hardware a big if).

If the Android Laptop OS (and supporting software/drivers) were bundled with the standardized 10.1" netbook mainboard I'd imagine these problems wouldn't exist.

Now what type of companies would have an incentive to do this? Maybe ones like Gigabyte that do not have a conflict of interest with making devices?
 

Cerb

Elite Member
Aug 26, 2000
17,484
33
86
Yes, but netbooks are practically dead anyway. Furthermore, many of them already shared a common chassis for several years.
(snip)
In fact, that might be a contributing reason why "Rich Creek 2" failed. It focused on the higher end Intel processor laptops rather than on the slower moving product lines like Atom.
Laptop = notebook = netbook = ultrabook. I don't know of any major OEM that has publicly shared a portable computer chassis with a competitor, even if you could go find a replacement board or frame piece from another vendor and install it (they do sometimes contract it out). That's desktop territory, and it has only been due to consumers disliking them being too proprietary there (Dell and HP did try!), that most big vendor PCs are mostly standards-compliant. In portable computers, they never let standardized internal hardware take hold.

As to Atom, Intel neglected Atom, and also feared that Atom would eat into other products' niches (that is, margins). If there's one thing I hate about Intel, it's dividing the market up into neat little holes, of which each user and product must fit in one of. If Atom had gotten faster, it would have fared better. If Atom and low-power chipsets together were affordable without design restrictions, Atom could have flourished (see Zacate).

Making them cheaper to own/upgrade could breath life into this dieing segment that many OEMs would probably like to abandon.
How? It is the OEM's best interest to sell you a product that they make money on. They don't make money when you buy an upgrade part. Businesses in general also prefer to chase volume and margin, so they create and burst bubbles, which the netbook was no exception to. If making them cheaper to buy sells more, they'll do that (full-size Zacate). If making them better can get them better prices, they'll do that (Thinkpad, Lattitude, Probook, etc.). Making them easy to upgrade with 3rd party components is pretty much not on their radar, except where absolutely necessary.

Will MS use a lighter codebase making them seem more surefooted and responsive compared to a bobcat or Atom notebook?
No. Not because they won't use a lighter codebase (less storage makes what storage exists more precious, and mobile devices use flash), but because that has little to do with responsiveness, beyond what lightening up you can do on your average x86 PC. It's about scheduling, including preventing IO locks whenever possible, which will be about the same for ARM and x86. Nearly everything else involving responsiveness is up to programmer of your application (MS has done pretty much everything possible except a clean scheduler re-implementation, to get Windows and Windows programs to run as well as possible).

Or will we see a Windows 8 port that actually penalizes the CPU compared to an Android or Apple ARMv8 Laptop operating system?
Not a chance. If we're lucky, we will get to see a playing field about as level as x86 v. PPC was.

If the Android Laptop OS (and supporting software/drivers) were bundled with the standardized 10.1" netbook mainboard I'd imagine these problems wouldn't exist.
True. However, this is just as true with completely non-standard designs. Linux doesn't care any more than Windows if you have 2 USB ports, or if you have 6. They don't care what dimensions your motherboard is. They don't care if your computer takes in 120V, 19V, 16V, 12V, 9V, 7.5V, 5, 3.3V, etc..
 

cbn

Lifer
Mar 27, 2009
12,968
221
106
Laptop = notebook = netbook = ultrabook. I don't know of any major OEM that has publicly shared a portable computer chassis with a competitor, even if you could go find a replacement board or frame piece from another vendor and install it (they do sometimes contract it out). That's desktop territory, and it has only been due to consumers disliking them being too proprietary there (Dell and HP did try!), that most big vendor PCs are mostly standards-compliant. In portable computers, they never let standardized internal hardware take hold.

I don't think something like this would appeal to major OEM (HP, Dell, Samsung).

Open Netbook standard would be geared towards the the likes of Antec, Corsair, Silverstone, etc. (or any white box chassis for that matter)

As to Atom, Intel neglected Atom, and also feared that Atom would eat into other products' niches (that is, margins). If there's one thing I hate about Intel, it's dividing the market up into neat little holes, of which each user and product must fit in one of. If Atom had gotten faster, it would have fared better. If Atom and low-power chipsets together were affordable without design restrictions, Atom could have flourished (see Zacate).

If you mean Microsoft's design restrictions....yes, I agree. The fact MS will only allow 1GB memory (about $3.50 worth of RAM) to be used with its relatively heavy (by ARM standards) Windows 7 Starter OS is simply due to the fact they have the power of a near monopoly.

How? It is the OEM's best interest to sell you a product that they make money on. They don't make money when you buy an upgrade part. Businesses in general also prefer to chase volume and margin, so they create and burst bubbles, which the netbook was no exception to. If making them cheaper to buy sells more, they'll do that (full-size Zacate). If making them better can get them better prices, they'll do that (Thinkpad, Lattitude, Probook, etc.). Making them easy to upgrade with 3rd party components is pretty much not on their radar, except where absolutely necessary.

I am talking about Intel, AMD, VIA, Qualcomm, TI, Nvidia making more money. Not the device makers.

True. However, this is just as true with completely non-standard designs. Linux doesn't care any more than Windows if you have 2 USB ports, or if you have 6. They don't care what dimensions your motherboard is. They don't care if your computer takes in 120V, 19V, 16V, 12V, 9V, 7.5V, 5, 3.3V, etc..

Do you see my point though? Why buy a whole new laptop if the darn thing can just be engineered to take a replacement part?

Of course, this hurts Giant OEMs who want to sell us a completely new laptop every time we want a new CPU. But the whole point is take their failing netbook strategy away from them and let the smaller companies have a shot.

Speaking of failing strategies. Samsung announced this year they would no longer be making netbooks.
 

Cerb

Elite Member
Aug 26, 2000
17,484
33
86
rictions....yes, I agree. The fact MS will only allow 1GB memory (about $3.50 worth of RAM) to be used with its relatively heavy (by ARM standards) Windows 7 Starter OS is simply due to the fact they have the power of a near monopoly.
MS helped out, no doubt about it, but Intel kept the form factor absolute, and limited the RAM. Want to make a 15" (MSI did, IIRC)? Then you're gonna pay a mint, and get the high-power version of the 945. Want 4GB? Get a CULV.

The netbook, being artificially restricted, died like a bubble, when large form factor notebooks and AIO computers should have grown out of it. Instead, that wasn't allowed to happen until Zacate.

I am talking about Intel, AMD, VIA, Qualcomm, TI, Nvidia making more money. Not the device makers.

Do you see my point though? Why buy a whole new laptop if the darn thing can just be engineered to take a replacement part?
How would that work, though? Everything is soldered in, typically, except storage. The initial price would be raised significantly, along with size, for the ability to swap out parts. In the mean time, IO and storage standards will march on. I don't know what VIA is going to do, or right now, AMD, but Intel, Qualcomm, and NVidia have quite a bit to gain by having customized hardware, even if it runs the same OS and application binaries as the other guys. They can work with OEMs to make devices using their chips use them to their best, and I don't see anything wrong with that, whether it's x86 or ARM.

Occasionally people will replace the CPU, if possible, but it's not always possible to do. As it is, you'd buy a new one because the old one is broken or obsolete. Storage and other IO standards move on all the time, and the same thing happens on standardized desktops: people end up, whether a part can be replaced or not, getting a whole new PC, or very near it, because upgrading isn't worth the cost, even when it is an option.

Technology, even when we have the standards, is moving fast enough that buying something better than you need today, and upgrading storage (HDD/SDD and RAM) is generally the best way to not need a whole new system in just a couple years. In just two years, my mobo, RAM, and CPU were entirely obsolete, in my standard ATX desktop.

There's definitely more flexibility in the desktop, but the overriding problem is how consumers at large think of technology, which helps drive the development of it. They aren't going to upgrade more than storage, RAM, and external peripherals in a laptop even if competitors could have parts swapped around, because they want something new and different. Changing feature sets and sockets to make each part slightly cheaper and more effective, instead of locking those things in, FI, is not merely accepted, but often even desired (the 1-pin-off mobile variants are just for market segmentation, though :)).

If I'm going to spend a decent chunk of change, I'm going to try to make it count; and a product with a usable lifespan of one or two years doesn't cut it. But, if being connected on the go were a part of making money for me, I can think of several good notebooks that would fit the bill, just not <=$400 ones in a sale flier ($300-400 might get a good used 3-4 year old notebook).

That market exists, both for new and used, but it doesn't make the bean counters excited like the Joneses and their friends do. They want a newer better faster $250-400 notebook, or a $800 notebook that's as good as last year's $1500 one, etc.. Netbooks are a case of barely legal collusion, but the rest of the notebook market is very much driven by what consumers want, and they are, by and large, happy with disposable computers. Once tablets get good enough (IoW, normal people consider non-iPads), the same will likely be true there, too.

Ultimately, computers are becoming like any other commodity. Why should car makers be required to have interchangeable parts? Why should washer and dryer manufacturers be required to? Etc.. It can be a very wrong way of thinking (the disposable commodity part, at least), but the issue is social, political, and economic, not a matter of standardization. Too few people would take advantage of forward-compatibility of a chassis/board for it to be worth making future chips/boards backwards-compatible.

Of course, this hurts Giant OEMs who want to sell us a completely new laptop every time we want a new CPU. But the whole point is take their failing netbook strategy away from them and let the smaller companies have a shot.
Why not just do what AMD did? Zacate can move about any form factor. No need to try to push consumers into little grid boxes they don't want their money to fit into. If you want a faster CPU, you need to pay more upfront. If you want upgradability across generations, you would be paying even more, and losing out on new features, if not a chunk of performance. May as well just pay more now, and not use it all, yet.

Speaking of failing strategies. Samsung announced this year they would no longer be making netbooks.
The few people I've talked to who have one love them, so hopefully they're getting out of 'netbooks' as in the MS/Intel defined devices, not out of small notebooks.
 

Idontcare

Elite Member
Oct 10, 1999
21,110
64
91
Game, set, match

http://www.youtube.com/watch?v=0PiJeiaMq3E

This is the future. Soon, desktops only for gaming/professional apps.

Edit: ARM+OpenGL.

I see a big market for this device but I don't see it displacing desktops anymore than the currently available super-cheap Via Nano systems are displacing desktop sales.

If desktops were $2k and this came along and offered a compelling price/capability alternative then I could see it making inroads. But people can already build desktops that are $300-$400 without sacrificing all the things they would sacrifice for the cotton candy device.

Personally I am still seeing potential with further convergence mechanisms, the dockable smartphones and so on. But truly replacement/displacement devices that do nothing to converge existing stuff is a dead-end IMO.
 

jones377

Senior member
May 2, 2004
463
64
91
I see a big market for this device but I don't see it displacing desktops anymore than the currently available super-cheap Via Nano systems are displacing desktop sales.

If desktops were $2k and this came along and offered a compelling price/capability alternative then I could see it making inroads. But people can already build desktops that are $300-$400 without sacrificing all the things they would sacrifice for the cotton candy device.

Personally I am still seeing potential with further convergence mechanisms, the dockable smartphones and so on. But truly replacement/displacement devices that do nothing to converge existing stuff is a dead-end IMO.

It seems kinda pointless to me. More like a "we can do this now" sort of product. Better to have an all in one cellphone that can also act as a PC when docked. Everyone needs a cellphone anyway right? I'm thinking a cellphone/PC that you can add accessories to make it a fullblown PC. like a wireless HDMI/DVI reciever, wireless keyboard/mouse, wireless NAS, cloud service, etc. Those accessories should still be cheaper than a separate PC and everyone still needs a cellphone, right? :)
 

cbn

Lifer
Mar 27, 2009
12,968
221
106
I am still seeing potential with further convergence mechanisms, the dockable smartphones and so on.

With respect to "Standardized Laptop Chassis" I do think "lapdocks" will and should be the gold standard (for obvious reasons).

Anyone can attach or swap different smartphones on a "lapdock"....essentially changing the Laptop's CPU/mainboard and OS in a mere 2 to 3 seconds, no technical skill needed.

But what direction will lapdocks go? Maybe Hard drives inside of them? Qualcomm Mirasol Displays for the Lapdock? How about provisions (due to the extra battery) for overclocking the phone's CPU/GPU when it is docked?

Lastly, what company will drive Lapdocks the most? Motorola is the obvious one, but I can see Intel wanting this to happen also. (X86 smartphone just begs to be docked so it can access all the traditional apps developed for keyboard and large screen)
 

exar333

Diamond Member
Feb 7, 2004
8,518
8
91
With respect to "Standardized Laptop Chassis" I do think "lapdocks" will and should be the gold standard (for obvious reasons).

Anyone can attach or swap different smartphones on a "lapdock"....essentially changing the Laptop's CPU/mainboard and OS in a mere 2 to 3 seconds, no technical skill needed.

But what direction will lapdocks go? Maybe Hard drives inside of them? Qualcomm Mirasol Displays for the Lapdock? How about provisions (due to the extra battery) for overclocking the phone's CPU/GPU when it is docked?

Lastly, what company will drive Lapdocks the most? Motorola is the obvious one, but I can see Intel wanting this to happen also. (X86 smartphone just begs to be docked so it can access all the traditional apps developed for keyboard and large screen)

Companies like proprietary because of $$$$.

Standardizing reduces their profits...accessories are HIGH margin.
 

cbn

Lifer
Mar 27, 2009
12,968
221
106
How would that work, though? Everything is soldered in, typically, except storage. The initial price would be raised significantly, along with size, for the ability to swap out parts. In the mean time, IO and storage standards will march on. I don't know what VIA is going to do, or right now, AMD, but Intel, Qualcomm, and NVidia have quite a bit to gain by having customized hardware, even if it runs the same OS and application binaries as the other guys. They can work with OEMs to make devices using their chips use them to their best, and I don't see anything wrong with that, whether it's x86 or ARM.

Occasionally people will replace the CPU, if possible, but it's not always possible to do. As it is, you'd buy a new one because the old one is broken or obsolete. Storage and other IO standards move on all the time, and the same thing happens on standardized desktops: people end up, whether a part can be replaced or not, getting a whole new PC, or very near it, because upgrading isn't worth the cost, even when it is an option.

Technology, even when we have the standards, is moving fast enough that buying something better than you need today, and upgrading storage (HDD/SDD and RAM) is generally the best way to not need a whole new system in just a couple years. In just two years, my mobo, RAM, and CPU were entirely obsolete, in my standard ATX desktop.

Since these small SOC are BGA, a new mainboard would also have to be purchased to get the CPU/GPU.

But I can't imagine this being too expensive. $65- $75 for a new SOC/mainboard is a lot cheaper than paying $299 for another low quality netbook.

The key factor is making the cheap hardware and software easy to swap.

IMO, this is why using Microsoft Windows fails. Nobody is going to pay $100 for a single use OEM license or $180 for a multi-use Retail license and then install that OS on a cheap $65 SOC Mainboard.

High end hardware would be the goal of someone spending that kind of money. The economics of the MS OS situation is probably one good reason we only saw larger size notebooks like ASUS C90 or OCZ "DIY Notebook" (with their heavy and difficult to remove coolers) billed as upgradeable.
 

Blitzvogel

Platinum Member
Oct 17, 2010
2,012
23
81
Could be interesting, but ARM as a practical platform for high end consumption and digital creation has a very long way to go. Even quad core ARMs are hardly competition for something like Bobcat. Yes, Bobcat has no where near the power efficiency for the same clock, but you're talking about 2 full 64 bit OS capable x86 cores here + 80 Radeon SPs that will deliver better performance on any front than even a quad ARM device like Tegra 3.

That makes me wonder though, where is the secret sauce in ARM design? Lower issue/ALU, more shallow pipelines? What does scare me is that everything will be designed to work via an ARM type of way, instead of a full x86 type structure. It will be like consoles vs computers, and porting code between the two. The difference this time around will not only be lack of congruent performance in entertainment and video games, but purely practical applications too.
 
Last edited:

cbn

Lifer
Mar 27, 2009
12,968
221
106
Could be interesting, but ARM as a practical platform for high end consumption and digital creation has a very long way to go.

Google buying Motorola gives me confidence plans exist for Android on the Lapdock and deskdock.

With that will come the appropriate large screen and keyboard/mouse/trackpad apps for content creation.

Even quad core ARMs are hardly competition for something like Bobcat which will just utterly destroy ARM both in per core performance and graphics.

I'm sure Bobcat is faster than the Current Cortex A9s, but the Cortex A15 MP Core looks to be a strong improvement over A9. (I've seen claims of 50&#37; better IPC coupled to much higher clocks for the top end A15s)

With ARMv8 the CPUs will be even stronger. (Which opens up a new line of questions of how various companies will target their power saving/process tech strategies for Mobile, but still allow strong performance when ARMv8 is docked to the larger Lapdock battery or even better yet... AC power from a wall outlet.)
 

Blitzvogel

Platinum Member
Oct 17, 2010
2,012
23
81
Google buying Motorola gives me confidence plans exist for Android on the Lapdock and deskdock.

With that will come the appropriate large screen and keyboard/mouse/trackpad apps for content creation.



I'm sure Bobcat is faster than the Current Cortex A9s, but the Cortex A15 MP Core looks to be a strong improvement over A9. (I've seen claims of 50&#37; better IPC coupled to much higher clocks for the top end A15s)

With ARMv8 the CPUs will be even stronger. (Which opens up a new line of questions of how various companies will target their power saving/process tech strategies for Mobile, but still allow strong performance when ARMv8 is docked to the larger Lapdock battery or even better yet... AC power from a wall outlet.)

I too wonder how all these improvements can be made without throwing efficiency out the door in the process in order to catch up to Bobcat. It's like trying to take an F-16 and make it equal to an F-22. You just need a whole new design. Not only does the F-16 lack two engine nacelles, the single nacelle isn't big enough for one of the Raptor's engines. The nose and frontal area of the F-16 isn't big enough for the Raptor's APG-77 radar or same suite of electronics. F-16 doesn't have thrust vectoring, etc, etc, etc.......... So how do you take down one Raptor (in most minimalist sense)? Get a bunch of F-16s (like 10 or so).
 
Last edited:

cbn

Lifer
Mar 27, 2009
12,968
221
106
I too wonder how all these improvements can be made without throwing efficiency out the door in the process in order to catch up to Bobcat.

You make a good point.

I think this chart (originally posted by IDC) does a good job of illustrating the diminishing returns of pursuing IPC (from a performance per watt standpoint).

05.jpg
 
Feb 19, 2009
10,457
10
76
Whatever thats coming needs x86 support because as soon as you get into notebooks, thats where a LOT of consumers use their notebooks for work. Lots of work related software only supports x86 windows.

I know because at my work place, there's a few apple worshippers with their ipads and macbooks.. who have to work on a desktop PC because our stuff don't run on apple. :D
 

Blitzvogel

Platinum Member
Oct 17, 2010
2,012
23
81
You make a good point.

I think this chart (originally posted by IDC) does a good job of illustrating the diminishing returns of pursuing IPC (from a performance per watt standpoint).

05.jpg

YES! I've seen that chart (and all the other ones he's made). He makes a great point about the relevancies of node size/process/IPC/and power consumption.