[Q] ARM vs x86 in consumer space in 10 years

[5150Joker]

Microsoft had the opportunity with Windows RT to extend the Windows ecosystem to ARM. However they decided to:
1) Artificially lock-out desktop app execution unless properly signed
2) discontinue Windows for ARM (at least from product perspective, internally they certainly still are compiling Windows for ARM)

In particular 1) was a stupid move, as you just had to recompile Win32 Apps to run under Windows RT. CLR/.Net apps were even running as is. Microsoft had a full featured Windows for ARM but stumbled over their own stupid decisions. Would not surprise me if 10 years down most would be running desktop apps under ARM, but it would not be Windows because Microsoft missed the boat and locked itself into x86 territory.

It's true that they messed up with Windows RT but that still doesn't prevent them from rectifying that mistake by building a Windows 10 version of it that can run x86 apps (even if emulated). If Sony/MS pick an ARM SoC next generation, then it would open the doors for compatibility with tablets and desktops since ARM would cover all three of them.

The key is publishers supporting Vulkan and we're seeing it being adopted for Android and if Sony and Nintendo's next consoles support it, then that could open the doors for ARM in the desktop world. The question is how efficient would ARM be in 5-10 years on the desktop vs Intel and if it would be worth the cost. At the rate ARM processors are progressing, it's possible they could start catching up to Intel on the desktop side.

We had MIPS consoles and PowerPC consoles, didn't change a thing in PC. Picture this- there is an ARM port of SteamOS, but it offers 1/10th of the games of the x86 version due to lack of compatibility. Which do you buy?

Yes but back then we didn't have a world filled with ARM tablets and smartphones. If you can produce a product that ports between console/tablet/desktop easily, that should be appealing to developers and gamers. Look at the current console designs, they're specialized PC parts and console ports are available on the PC at the same time now with very little changes needed. I don't see why that couldn't happen with ARM designs in the future. It's pretty evident that ARM is the future in the consumer market and x86's days are numbered. I don't see Intel overcoming this challenge despite how much money it has. In fact, it wouldn't surprise me to see Intel producing ARM CPUs in house at some point in the future.

 

[MarkizSchnitzel]

..unless Intel fully adopts ARM themselves, it's a safe bet to assume absolutely no compatibility with x86 apps, or at least, anything speed critical.

How likely would that be? I mean, if they had to, I'm sure they will, but would they be able to, being so inert?

If the next generation consoles make the logical leap to an SoC that uses an ARM processor then that could pave the way for a Windows version that runs ARM.

That was in the back of my mind as well, MS breaking from x86. I'd welcome it, simply for competition bringing benefits to consumers.

There were 10 billion ARM enabled SOC's/CPU's sold 2013. Intel sells less than 500 million CPU's per year. So ARM camp is more than 20 times bigger today (volume wise).

I guess I needed to be more precise, I was thinking CPUs in devices like laptop, hybrid and desktop form factors.

well the open post only talks about consumer space, not your definition

Sorry about that, should have clarified.

Microsoft had the opportunity with Windows RT to extend the Windows ecosystem to ARM. However they decided to:
1) Artificially lock-out desktop app execution unless properly signed
2) discontinue Windows for ARM (at least from product perspective, internally they certainly still are compiling Windows for ARM)

In particular 1) was a stupid move, as you just had to recompile Win32 Apps to run under Windows RT. CLR/.Net apps were even running as is. Microsoft had a full featured Windows for ARM but stumbled over their own stupid decisions. Would not surprise me if 10 years down most would be running desktop apps under ARM, but it would not be Windows because Microsoft missed the boat and locked itself into x86 territory.

Thanks for this view, I was thinking something along these lines when the RT came out. it was so limited in every way imaginable, it never had a chance, but the idea sounded really attractive.

 

[ShintaiDK]

x86 sits on 2/3rds of all global MPU revenue. Just to put things into perspective.

 

[Arachnotronic]

x86 sits on 2/3rds of all global MPU revenue. Just to put things into perspective.

Yeah, people tend to forget this.

ARM and its licensees have won smartphones, a large market for compute to be sure. But Intel, IMO, will continue to reign supreme in notebooks/desktops and the data center.

 

[MarkizSchnitzel]

x86 sits on 2/3rds of all global MPU revenue. Just to put things into perspective.

Exactly the kind of info I was hoping for. Thanks.

 

[5150Joker]

x86 sits on 2/3rds of all global MPU revenue. Just to put things into perspective.

They haven't been properly challenged yet in the notebook/desktop market, that's why. I think that will change in time once ARM becomes a viable alternative. Short of quantum computing, Intel can't go too much further with node shrinks to stay ahead of the industry.

 

[ShintaiDK]

They haven't been properly challenged yet in the notebook/desktop market, that's why. I think that will change in time once ARM becomes a viable alternative. Short of quantum computing, Intel can't go too much further with node shrinks to stay ahead of the industry.

Oh they can. Specially if companies like Apple joins their foundry wagon. Then ARM(besides Apple) get stuck on much higher nodes. because TSMC/Samsung cant afford lower.

Its all a cash and revenue game.

 

[Thala]

Thanks for this view, I was thinking something along these lines when the RT came out. it was so limited in every way imaginable, it never had a chance, but the idea sounded really attractive.

Thing is, contrary to popular belief, that Windows RT was not limited at all. Some people think of it as dumbed down Windows for mobile while in fact it was a full featured Windows somewhere in between Windows Home and Pro. It is an ARM native build (Thumb 2) of all the Windows SW components with very few exceptions.
Thing is, Windows RT artificially rejects to load a desktop app, if it was not signed properly. Once developers found out how to circumvent this restrictions quite a few (open source) desktop apps were compiled for Windows RT with essentially a press of a button.

The key is publishers supporting Vulkan and we're seeing it being adopted for Android and if Sony and Nintendo's next consoles support it, then that could open the doors for ARM in the desktop world. The question is how efficient would ARM be in 5-10 years on the desktop vs Intel and if it would be worth the cost. At the rate ARM processors are progressing, it's possible they could start catching up to Intel on the desktop side.

Precisely. With Microsoft dropping Windows RT it is now up to the Linux world to open the desktop for ARM which includes APIs like Vulkan. Regarding performance ARM has overtaken Atom and is approaching Core territory in particular with recent CPUs like A9x. There is not question of if, but only when we will see ARMs in the desktop space.

 

[Thanatosis]

Yeah, people tend to forget this.

ARM and its licensees have won smartphones, a large market for compute to be sure. But Intel, IMO, will continue to reign supreme in notebooks/desktops and the data center.

I would have to agree with you on data center, but I think we have yet to see a real concerted effort on the part of an ARM licensee to build a quality notebook.

There is good reason to stick with intel in server and data center because they have a large performance advantage and a much more robust support structure plus lots of good software. ARM SoCs today are outperforming intel's competition at notebook TDPs and with all the noise about going fanless there is even more reason for Apple or other vendors to try out ARM in their notebooks. I think we may see the start of this transition in 2016.

 

[cbn]

We had MIPS consoles and PowerPC consoles, didn't change a thing in PC. Picture this- there is an ARM port of SteamOS, but it offers 1/10th of the games of the x86 version due to lack of compatibility. Which do you buy?

You know Valve has that SteamLink coming out for $50.

http://store.steampowered.com/app/353380/

But I keep on wondering how long before they starting adding additional logic to the ARM SoC in that that device so it plays games natively as well.

P.S. IMHO some competition to this evolving ARM console would be recycled products like this ---> http://forums.anandtech.com/showthread.php?t=2447324 (running some kind of low end video card, perhaps this one --> http://forums.anandtech.com/showpost.php?p=37593874&postcount=58 )

 

[NTMBK]

You know Valve has that SteamLink coming out for $50.

http://store.steampowered.com/app/353380/

But I keep on wondering how long before they starting adding additional logic to the ARM SoC in that that device so it plays games natively as well.

But that has the exact same problem I just described- there are no games for it. If you want a device that does that, you are much better off with an x86 SoC which can run the Steam back catalogue, and has a built in h.264 decoder which can decode the video stream just fine.

 

[cbn]

But that has the exact same problem I just described- there are no games for it. If you want a device that does that, you are much better off with an x86 SoC which can run the Steam back catalogue, and has a built in h.264 decoder which can decode the video stream just fine.

But If the goal is simply streaming then $50 for an ARM SteamLink box is cheaper than any new x86 box.

To get to $50 with x86, a person would need a sale priced refurbished Windows 7 Core 2 SFF desktop (which granted are a lot more powerful than a $50 ARM stream box, but are used hardware and make more noise....though the noise is not too bad IMO).

The question is what does Valve do with SteamLink II (or whatever they call the next device). 1.) Do they keep the hardware weak and use a die shrink to lower the price below $50? Or 2.) do they use the die shrink to add more logic and make the console more capable for $50? I'm thinking #2 is more likely to happen (especially considering they are most likely using off the shelf ARM hardware and these SoCs can only get so cheap).

 

[NTMBK]

But If the goal is simply streaming then $50 for an ARM SteamLink box is cheaper than any new x86 box.

Agreed, but that isn't what I was arguing against- I was arguing against your theoretical "able to run games locally" ARM machine.

The question is what does Valve do with SteamLink II (or whatever they call the next device). 1.) Do they keep the hardware weak and use a die shrink to lower the price below $50? Or 2.) do they use the die shrink to add more logic and make the console more capable for $50? I'm thinking #2 is more likely to happen (especially considering they are most likely using off the shelf ARM hardware and these SoCs can only get so cheap).

Just adding logic isn't going to turn it into a gaming capable device. You also need to add RAM to run the games in, and storage large enough to install modern games in. That's not a $50 device any more!

They will probably either a) keep using the same ARM SoC for years, or b) add the capability to decode higher resolution/framerate feeds, and improve the networking performance of the device. Playing non-existent ARM games seems like such an utterly pointless addition.

 

[Exophase]

So ARM loses on software performance, but wins on performance per watt?......but overall, this balances the situation out for VDI in some ways.

Two sides of the same coin. Any overhead from software translation that hurts performance a lot will also hurt perf/W a lot.

Intel is attacking the full perf vs perf/W spectrum in servers, even utilizing Atoms at one end. They also have a ton of institutional experience with servers. I don't see anyone being competitive with x86 running on ARM. For translation to be an acceptable proposition it has to be ran on hardware that was designed with that translation in mind like Denver.

 

[cbn]

Just adding logic isn't going to turn it into a gaming capable device. You also need to add RAM to run the games in, and storage large enough to install modern games in. That's not a $50 device any more!

They will probably either a) keep using the same ARM SoC for years, or b) add the capability to decode higher resolution/framerate feeds, and improve the networking performance of the device. Playing non-existent ARM games seems like such an utterly pointless addition.

Yes, if SteamLink doesn't have H.265 already the next version of the machine should.

Now as far as native gameplay goes, there are a lot of low requirement Indie games on Steam (with some needing less than 1GB for the install). They might be able to include some of those (as ARM ports) with not much in the way of hardware increase at some future point in time.

With that mentioned, some of Valve's own games like Team Fortress 2 have relatively high install footprints (somewhere around 16GB). Not sure when we would see an SteamLink console capable of playing that game natively (my E8500/4GB/HD5450 machine can play TF2 comfortably at 720p, and with 2GB RAM it would work fine as well).

 

[cbn]

So ARM loses on software performance, but wins on performance per watt?......but overall, this balances the situation out for VDI in some ways.

Two sides of the same coin. Any overhead from software translation that hurts performance a lot will also hurt perf/W a lot.

Yep, so the ARM Server running the x86 emulation needs to be inherently better in terms of performance per watt (with acceptable single thread). This, in order to counteract the loss in efficiency from the translation.

 

[TheELF]

We had MIPS consoles and PowerPC consoles, didn't change a thing in PC. Picture this- there is an ARM port of SteamOS, but it offers 1/10th of the games of the x86 version due to lack of compatibility. Which do you buy?

We had powerPC PCs back in 1995
the acorn archimedes
the commodore amiga
and even the powermac yup the whole damned osx ran on risc/arc up to leopard,
none of them survived,why? because you can't really push arc technology into performing, you can't get good speeds out of them,they suck at single thread and that is a very important part of gaming,unless you want the future to consist of eye candy games where you only press the x button whenever the screen flashes.

I mean look at the tablet/smartphone market,single speed seems to have hit the max already and now they start trying to push 8+ cores just to sucker people into believing that they will get better performance.

 

[Zodiark1593]

We had powerPC PCs back in 1995
the acorn archimedes
the commodore amiga
and even the powermac yup the whole damned osx ran on risc/arc up to leopard,
none of them survived,why? because you can't really push arc technology into performing, you can't get good speeds out of them,they suck at single thread and that is a very important part of gaming,unless you want the future to consist of eye candy games where you only press the x button whenever the screen flashes.

I mean look at the tablet/smartphone market,single speed seems to have hit the max already and now they start trying to push 8+ cores just to sucker people into believing that they will get better performance.

Apple seems to think not, though as none of us had any part in designing their CPUs, we don't know what black magic was involved, or if there are concessions that were made somewhere (more likely).

 

[5150Joker]

Oh they can. Specially if companies like Apple joins their foundry wagon. Then ARM(besides Apple) get stuck on much higher nodes. because TSMC/Samsung cant afford lower.

Its all a cash and revenue game.

Samsung as a conglomerate dwarfs Intel and the Samsung Electronics unit posted a Q3 2015 profit of $6.3 billion vs Intel's $3.1 billion. Intel isn't the giant it used to be and in a world where ARM is quickly catching up, Intel will have to contend with Samsung, Apple, Qualcomm, NVIDIA and to a smaller extent AMD. I don't think it can take all those players on long term and win. They were foolish for not licensing out x86 to more companies and in the end that's going to be their downfall. As a consumer, I can't wait for the competition to reignite, the CPU world has been stagnant for far too long.

 

[Arachnotronic]

Samsung as a conglomerate dwarfs Intel and the Samsung Electronics unit posted a Q3 2015 profit of $6.3 billion vs Intel's $3.1 billion. Intel isn't the giant it used to be and in a world where ARM is quickly catching up, Intel will have to contend with Samsung, Apple, Qualcomm, NVIDIA and to a smaller extent AMD. I don't think it can take all those players on long term and win. They were foolish for not licensing out x86 to more companies and in the end that's going to be their downfall. As a consumer, I can't wait for the competition to reignite, the CPU world has been stagnant for far too long.

FYI, the Samsung profit number is operating income not net profit; the comparable number for Intel here is $4.2 billion.

Much of Samsung's profit comes from the sale of memory and mobile devices; very little of that profit comes from selling logic semiconductors.

Also, NVIDIA is basically out of the race against Intel in its core MPU business (Tegra is 100% automotive now), Qualcomm is laying people off because it needs to bring its expenses into check (since the profitability of its mobile chip business has plunged into the abyss), and AMD is on its deathbed with people getting excited by a measly $370M cash infusion from selling off yet more assets.

Intel is in a very strong financial position and it can deliver these kinds of profits while at the same time investing at a multiple to what these other companies do in semiconductor logic-related R&D.

Intel isn't invincible but the notion that it's just going to roll over and die because a lot of people are trying to compete with it doesn't seem that well thought out IMO.

 

[Arachnotronic]

They were foolish for not licensing out x86 to more companies and in the end that's going to be their downfall.

You haven't really followed this industry for long, have you? Back in the heyday of the PC (when PCs were booming like phones are now), everybody had an x86 license: TI, IDT, Cyrix, VIA, AMD, IBM, etc.

One by one these players exited the market because it just wasn't economically viable; there's only so much x86 MPU revenue to go around and if you don't capture a large enough part of it, it makes no sense to continue to try to invest to build products.

FWIW, it seems very counterintuitive that "licensing x86 to more companies" thereby increasing competition and diluting the value of the x86 compatibility advantage that Intel has would be anything but a bad thing.

 

[imported_ats]

x86 sits on 2/3rds of all global MPU revenue. Just to put things into perspective.

And the corollary for phones is that Apple makes 90+% of all the profits.

AKA, its not so much that people are paying money to use ARM as they are doing it because its very very cheap. That foothold can just as easily be run over by something being even cheaper. There are vast amount of spaces that use ARM that could just as easily switch to using RISC-V in the future.

 

[imported_ats]

They haven't been properly challenged yet in the notebook/desktop market, that's why. I think that will change in time once ARM becomes a viable alternative. Short of quantum computing, Intel can't go too much further with node shrinks to stay ahead of the industry.

Oh how people forget history...

Fun fact...

ARM was originally in a desktop. That's where it started.

 

[imported_ats]

Precisely. With Microsoft dropping Windows RT it is now up to the Linux world to open the desktop for ARM which includes APIs like Vulkan. Regarding performance ARM has overtaken Atom and is approaching Core territory in particular with recent CPUs like A9x. There is not question of if, but only when we will see ARMs in the desktop space.

Um, when is not in question. It was 1987. ARM eventually dropped out of the market when they couldn't compete.

 

[dark zero]

Yeah, people tend to forget this.

ARM and its licensees have won smartphones, a large market for compute to be sure. But Intel, IMO, will continue to reign supreme in notebooks/desktops and the data center.

Not at all seeing that the Notebook and the Desktop are dying extremely faster than they can recover and the Chromebook is also adding the insult to the injury. Only in the military and the Datacenter areas have influence and strong revenues.