AMD CPU speculation... and expert conjecture

[cdrkf]

From the EETIMES article I mentioned before:

Keller suggested AMD will consolidate its work on separate Bulldozer and Jaguar x86 cores into one processor line in the future.

Kind of scarey if you think about it. ARM can't do it, Intel can't do it, but somehow AMD will do it. A one size fits all core. The holy grail of computing.

There sure were some harsh comments regarding Keller on there.

I think people are reading rather allot into that comment there.

The facts we know- AMD are now designing a new x86 and ARM core. Keller has mentioned that they don't intend to have 2 x86 cores any more.

The long and short of that is to my mind is:
The 'big' x86 cores scale down to about 15w reasonable easily (with decent perf) as proven by Intel and to a lesser extent AMD (low power Kaveri is actually looking pretty good).

ARM cores scale up to about 25W currently- so there's you're overlap. I think the point is there isn't much benefit to AMD having the low power 2 - 25W x86 parts when they have already stated they can achieve a 4x efficiency improvement with their new ARM part.

If they release a much better big x86 core, with better ipc et al- it will probably scale nicely down as low as thin and light laptops. For anything smaller they do ARM.

 

[juanrga]

The facts we know- AMD are now designing a new x86 and ARM core. Keller has mentioned that they don't intend to have 2 x86 cores any more.

Exactly.

ARM cores scale up to about 25W currently- so there's you're overlap. I think the point is there isn't much benefit to AMD having the low power 2 - 25W x86 parts when they have already stated they can achieve a 4x efficiency improvement with their new ARM part.

Only to mention that 25W ARM SoC is the 8-core version. The posterior 16-core version will have a higher TDP.

http://www.bit-tech.net/news/hardware/2013/06/19/amd-to-launch-arm-processor-seattle-in-2014/1

 

[juanrga]

FD-SOI offers some power consumption benefits for low-performance products. For high-performance products FINFETS are required. Consequently all foundries chose FINFETs for the 14nm node except STMicro that chose FD-SOI for 14nm.

As everyone knows Globalfoundries original 14nm FINFET process (14XM) was canceled recently and replaced by Samsung FINFET 14LPP/LPE process. Both are FINFET on bulk processes.

We already knew that STMicro signed a 14nm FD-SOI partnership with Globalfoundries for providing a second-foundry source for volume production in case STMicro couldn't satisfy client demands.

STMicro signed now a 28nm FD-SOI partnership with Samsung for the same reason:

While ST is already qualified for volume production at its Crolles facility, they needed more production volume. "ST couldn't support it [28nm FD-SOI] so they needed a partner," said Len Jelinek, IHS's research director for semiconductor manufacturing, in an interview with EE Times. "Choosing Samsung was a logical move."

What products will be made on the 28nm FD-SOI process at STMicro/Samsung foundries?

Low-power, low-margin wearables and smartphone chips.

Why is STMicro predicting a higher demand from its customers?

Because it predicts explosion of the cheap smartphone market in China and Taiwan.

Will we see a massive migration of cheap 40nm phone chips to the new 28nm FD-SOI?

I guess no.

Why?

28-nm FD-SOI is supposed to afford a 30% faster performance. "But that's at maximum frequency," said Freeman, "and no device operates at maximum frequency all the time. It's more likely 15 to 20%" performance increase."

What does this mean for Samsung future 14nm high-performance plans?

Nothing

"I would characterize it as an interim step for Samsung until they get to 14nm," Kevin Krewell of Tirias Research told EE Times. "FD-SOI offers power benefits but you don't get the strength." He explained that the process does work well for wearables where the idle power is most important. FD-SOI gives just enough performance to handle a wearable's work and small display but it reduces the power to extend the battery life that is more important in wearables, which spend most of their lives in sleep mode. Having an efficient sleep mode that doesn't leak power is ideal. FD-SOI, with its lower voltages than bulk CMOS, could achieve that.

http://www.eetimes.com/document.asp?doc_id=1322353

What does this partnership between Samsung and STMicro mean for AMD products?

Nothing.

AMD will migrate from current 28nm bulk to 20nm bulk the next year, and then to 14/16 FINFET (bulk). As I predicted. Keller has already confirmed that the new K12 core is being designed on 14/16 FINFET. Keller admits that the 14/16 FINFET process looks fine, and he is happy with it.

 

[de5_Roy]

Mullins can’t find design wins
http://vr-zone.com/articles/mullins-cant-find-design-wins/77479.html
sigh. doesn't look like a fair competition.

Final Thoughts on AMD's Core Technology Update Presentation
http://www.pcper.com/reviews/Processors/Final-Thoughts-AMDs-Core-Technology-Update-Presentation
found seattle's secret sauce - massive 4MB shared L2 cache and 8MB L3 cache. amd didn't even bother to tweak kyoto, just slapped together two kabinis and called it a day. the jaguar-based cpu clusters have 2MB shared L2 cache, no L3 ... and nearly half of the die is occupied by the igpu that server applications won't likely use.

 

[Cazalan]

@cazalan: does this mean the previous alliance of st-ericsson is dissolved and samsung will use st's booster tech in their 28nm process? it seemed like samsung hit issues with their own 20nm process... and this makes me question glofo and samsung's 14nm bulk's viability.

The 14/20nm is simply more expensive but that's been known for a while, and they still have to ramp wafer capacity. What it says mainly is that their 28nm node will remain viable and continue bringing in revenue for Samsung. Of course that doesn't affect GF yet unless they choose to cross license that too.

 

[juanrga]

Final Thoughts on AMD's Core Technology Update Presentation

It was not all that long ago that AMD was essentially considered road kill, and there was a lot of pessimism that Rory Read and Co. could turn AMD around. Now after a couple solid years of growth, a laser-like focus on product development based on the IP strengths of the company, and a pretty significant cut of the workforce, we are seeing an AMD that is vastly different from the one that Dirk Meyers was in charge of (or Hector Ruiz for that matter). Their view for the future takes a pretty significant turn from where AMD was even 8 years ago. x86 certainly has a future for AMD, but the full-scale adoption of the ARM architecture looks to be what finally differentiates this company from Intel.

Exactly! the old AMD is gone.

Their goal is not only to develop and sell x86 based processors, but also be a prime moving force in the ARM market.

ARM also has the ability to scale in performance from handhelds up to the server world, but so far their introduction into servers and HPC solutions has been minimal to non-existent. This is an area that AMD hopes to change, but it will not happen overnight. A lot of infrastructure is needed to get ARM into that particular area. Ask Intel how long it took for x86 to gain a handhold in the lucrative server and workstation markets.

ARM is in the same position that Intel was decades ago. Then the old-RISC guys (alpha, SPARC, MIPS...) did claim that x86 couldn't scale up to servers/HPC. Not only x86 scaled up but replaced them from top500 list now. Today some people claims that ARM cannot scale up. Not only ARM is scaling up, but ARM will replace x86 as discussed at last Supercomputer Conference

http://dl.acm.org/citation.cfm?id=2503281

Yes infrastructure is needed. Precisely ARM has developed the SBSA (Server Base System Architecture) in collaboration with AMD, Canonical, Citrix, Linaro, Microsoft, Red Hat, SUSE, Dell, HP, AppliedMicro, Broadcomm...

Seattle is AMD first product compliant with SBSA

http://cdn.pcper.com/files/imagecache/article_max_width/review/2014-05-14/amd_02.png

We do not know much about AMD’s ARM design other than it is codenamed K12. We have no idea what kind of power range this product will span, but my guess is that it will be relatively low powered. Think 65 watts and below.

I expect higher TPDs (but below 100W). However, it is worth mentioning this is the expected TDP for future node process. Kaveri is a 95W APU when made on 28nm, if Kaveri was replicated in the same node used for K12, then the TDP would be reduced from 95W to about 45W approx.

Thus when correcting for the different node scaling, the "relatively low powered" 65W TDP is in reality representing a more powerful chip than current top A10 Kaveri.

I doubt AMD will ever produce another FX-8000 or FX-9000 type of part ever again. From here on out it looks like much more power efficient APUs will be the desktop offerings. We can only hope that AMD reaches the same level of IPC and power efficiency that Intel has with Haswell. AM3+ is the last socket of its kind, and from here on out we will have products that will have to fit in FM2+’s 100 watt TDP envelope.

Indeed! As I predicted no more inefficient 220W FX OC CPUs.

However, it seems the old FX-brand will be reused for high-performance APUs/CPUs: A4 --> A6 --> A8 --> A10 --> FX, with the FX brand starting at 8-cores and extending up to 16-cores.

 

[juanrga]

The funny thing about all this ambidextrousness, it was part of Dirk Meyers strategy and he was dropped like a hot potato.

The funny part is that ARM already did talk about this so early like 2011:

One possibility is that AMD could amend its Fusion architecture to include both x86 and ARM CPU cores plus graphics cores in a heterogeneous multiprocessor. However, it is also possible that AMD may choose to abandon difficult and expensive attempts to develop its own multicore x86 architecture in competition with Intel and, instead, allow ARM to provide its cores.

http://www.eetimes.com/document.asp?doc_id=1259378

 

[etayorius]

So if ARM replaces x86... what about PC Gaming and all the x86 Games? New Tech is always cool, but what we gona do when all PCs are ARM... 98% of software is on x86.

I don`t mind to eventually move into ARM as long as ARM manages to run (not emulate) x86 software, i care about my PC games everything else is secondary for me.

 

[Cazalan]

So if ARM replaces x86... what about PC Gaming and all the x86 Games? New Tech is always cool, but what we gona do when all PCs are ARM... 98% of software is on x86.

I don`t mind to eventually move into ARM as long as ARM manages to run (not emulate) x86 software, i care about my PC games everything else is secondary for me.

Smart companies will develop games with cross platform APIs. Unreal engines 3 and 4 can both run in a web browser even.

 

[harly2]

Juanrga....

Really enjoy lots of your points, thanks for the info you have posted. But if I may....I notice over at semi-accurate you don't do all the self promotion with all the "as I predicted" and "like I said"... I I I. Would you mind talking more like that here. Not sure if your using this forum as your own for an ego boost or something but Its getting to me. I go around every few days and read all the AMD news and just keep running into you doing this here. You quote someone and then rant about how you said it. Everything you find is available at numerous sites. And if you do predict something and its right that's cool, know one is viewing you as an oracle or something, just take or give info without the self promotion please. Again thanks for the digging and the info but its clear that you do this for yourself, don't pretend your doing the world a favor. Its your hobby and that's cool, in this forum on this site do you mind just stopping the constant self promotion. Not sure if you think some of the old semi-accurate guys would put you in your place or something but you definitely talk in a different tone here like everyone should bow down.

PS: you're going around telling everyone the new AMD x86 ARC is based on Puma+ from an article that is being translated, and the article is vague, I read it. Pretty sure plenty of people have shot that down. We have access to all the same info you do... The arc is new not based on any current architecture, AMD is taking ideas..chill, and we know very little. Thanks for all the informative posts you have made and will make.

 

[griptwister]

From a Business standpoint, what AMD is doing is ingenius. Dominating a market and growing it. If they can do what Intel is doing now with the x86 market, They can re-join the x86 market and offer more competition. But, If ARM takes off to the point where x86 is no longer necessary That would be interesting. If some how there was a way of programming to rid of CPU usage all together when gaming, What would the need be for x86.

On the blue side of the world, Intel is coming up with CPUs that are capable of hitting 5Ghz on Air. I'm looking forward to these bad boys.

Also, That GPU that AMD is coming up with that's supposed to go against to 780 Ti. To my understanding, AMD is doing stacked RAM before Nvidia Volta? Haha! If so, I look forward to upgrading my GPU.

 

[harly2]

I agree on that the x86 cores will be massively preferred by Windows desktop users, but the non-Windows users could use the ARM version. There is an interesting discussion at other forums, with some people speculating about the possibility of Apple using the K12 in a future Mac Pro.

People have been "speculating" on forums about Apple using (or buying!) AMD since Jesus was a boy. Yawn.

Apple were actually interested in going with AMD APUs- they were looking at the original A8 but unfortunately AMD were about a year too late (it was delayed repeatedly).

Apple do use some AMD products as well- namely their graphics cards in a number of different things. I do think however that Apple are looking to use more of their own processor designs again (I think the move to 64bit on the A7 was future looking, I'm almost certain we'll see a Macbook Air based around some derivative of A7 in the near future).

Not possible, well it is and maybe you are correct but ARM is iOS and x86 is OS X. They would need to port OS X to ARM, and emulate old software on A7, that takes away some power. There is a few articles around the net about porting OS X to ARM but that would be useless right now and there is other rumors that apple ditched the idea, just think of the headache. I think If you see A7/A8 in air it will be iOS moving up now that it's 64bit, not OS X going to ARM.

The full power of both cores for an A7 is still so so for PC level device. Plus the PowerVR GPU would be put under a lot of stress outside of mobile world.

 

[griptwister]

Juanrga....

Really enjoy lots of your points, thanks for the info you have posted. But if I may....I notice over at semi-accurate you don't do all the self promotion with all the "as I predicted" and "like I said"... I I I. Would you mind talking more like that here. Not sure if your using this forum as your own or for an ego boost or something but Its getting to me. I go around every few days and read all the AMD news and just keep running into you doing this here. You quote someone and then rant about how you said it. Everything you find is available at numerous sites. And if you do predict something and its right that's cool, know one is viewing you as an oracle or something, just take or give info without the self promotion please. Again thanks for the digging and the info but its clear that you do this for yourself, don't pretend your doing the world a favor. Its your hobby and that's cool, in this forum on this site do you mind just stopping the constant self promotion. Not sure if you think some of the old semi-accurate guys would put you in your place or something but you definitely talk in a different tone here like everyone should bow down.

PS: you're going around telling everyone the new AMD x86 ARC is based on Puma+ from an article that is being translated, and the article is vague, I read it. Pretty sure plenty of people have shot that down. We have access to all the same info you do... The arc is new not based on any current architecture, AMD is taking ideas..chill, and we know very little. Thanks for all the informative posts you have made and will make.

Good post dude. I agree with you. I feel like Juan fires shots in the dark until he hits something. And when he does, he makes a big commotion about it. Really, this is a discussion, not a debate. Seeing someone tell everyone else they're wrong all the time get's irritating. It's getting worse than the Bill Nye and Ken Ham debate up in here. :lol:

 

[harly2]

http://img1.mydrivers.com/img/20140516/s_a3847e06fde04444bb6d668ec312caef.jpg
AMD held APU14 today in China. According to this talk, AMD plans to not move away from desktop. Rather, talks even about growth. Who here was saying that AMD plans to abandon desktop?

there's a bit of a marketing hype there, but they're not wrong. in countries like china, south korea, malayasia and indonesia afaik (but can't confirm atm) pcs have a very good base. china only recently lifted the console ban while possessing a well-established competitive pc gaming culture. in u.s., pc markets are saturated, so amd and intel are looking elsewhere.

@cazalan: does this mean the previous alliance of st-ericsson is dissolved and samsung will use st's booster tech in their 28nm process? it seemed like samsung hit issues with their own 20nm process... and this makes me question glofo and samsung's 14nm bulk's viability.

edit: who else here thinks that keller is being lined up and rory reed's successor and koduri being the cto? i mean, if keller becomes the ceo, and him being a former insider, the cpu core/ghz wars may heat up again.... just saying...

In my eyes Lisa Su is the Star.

 

[Cazalan]

From a Business standpoint, what AMD is doing is ingenius. Dominating a market and growing it. If they can do what Intel is doing now with the x86 market, They can re-join the x86 market and offer more competition. But, If ARM takes off to the point where x86 is no longer necessary That would be interesting. If some how there was a way of programming to rid of CPU usage all together when gaming, What would the need be for x86.

With the exception of Applied Micro (AMCC) which is already shipping a custom 8x ARMv8 core quite similar to Seattle, they're in a good position for now. They even say the same things.

"This will give the industry more choice and enable a move away from the one size fits all approach that has existed in the marketplace." - MiTAC, Computex 2013

 

[juanrga]

So if ARM replaces x86... what about PC Gaming and all the x86 Games? New Tech is always cool, but what we gona do when all PCs are ARM... 98% of software is on x86.

I don`t mind to eventually move into ARM as long as ARM manages to run (not emulate) x86 software, i care about my PC games everything else is secondary for me.

Software can be ported. Apple has made two changes in architecture: first from Motorola to PowerPC and then to x86.

I have always said that migration to ARM don't will be achieved in two or three years, but in the long run. And whereas servers and HPC will be the first ones to migrate, traditional Windows desktop users will be probably the last on migrate to ARM.

When the desktop ARM market was big enough, game developers will release ARM versions. Older x86 games will be played on simulation mode. No game developer will port the old games that you have already purchased.

It is worth mention that some years ago Carmack speculated with the idea that current game consoles would be based in ARM cores. It seems that both Microsoft and Sony tested ARM cores but found that, the then available 32bit cores, couldn't provide the same performance than x86

ARM-based architectures will soon get as powerful as AMD’s Jaguar cores, but not when Sony or Microsoft needed them for their new consoles.

It is easy to imagine AMD developing some custom SoC for a future powerful game console with K12 cores, stacked RAM and a whole of GCN cores. During an old Quakecon conference, John Carmack mentioned the possibility of an ARM based generation:

We can imagine something that maybe had 16 ARM cores with a whole bunch of PowerVR graphics cores, and you could have made a pretty good console with that,” he suggested. “I suspect it was the 64-bit not being quite cooked enough on there that really made the difference, and the fact that huge amounts of memory are obviously a big deal for the console platforms now.

 

[juanrga]

Some analysts see Apple migration to ARM as inevitable:

http://appleinsider.com/articles/12/11/06/arm-based-macs-seen-as-inevitable-but-apple-unlikely-to-switch-anytime-soon

Shaw Wu with Sterne Agee said in a note to investors on Tuesday that he believes it will take Apple some time to optimize its Mac OS X operating system for the ARM processors currently found in the iPad and iPhone.

Intel processors are currently much more powerful for running intensive Mac applications, as well as for development. But he believes that ARM processors will eventually become powerful enough to replace Intel chips.

In his view, making Apple's entire product line based on custom-built ARM-based processors would simplify the architecture of its devices, and also help to create a more seamless experience for users.

Wu's take was issued in response to a report that surfaced on Monday from Bloomberg, which indicated that Apple's engineers are confident that the company's A-series custom chip designs will one day be powerful enough to run the company's desktop and laptop machines. ARM-based silicon in Apple devices is currently limited to iOS devices.

Monday's report also suggested a change to ARM processors is not likely to take place "in the next few years." But it also portrayed a shift to proprietary chip designs as an "inevitable" transition for the company in the future.

 

[jdwii]

Juan Microsoft or GTFO with this Arm crap(i find it funny he is using apple now like how many Apple computers(not lame tablets) do you see?
In the U.S x86 will be number 1 for the next 10 years at least if not longer Arm is a troll for anything that needs high-end work or gaming or compatibility with older parts. Amd can say what they want but really the market is what matters.

 

[jdwii]

I remember i just asked my boss yesterday about Arm and he laughed and said that is low-end crap and couldn't even run a print server let alone anything that actually matters i guess in the U.S we can afford are electricity Bill

 

[-Fran-]

jwdii... Remember Windows RT? For all the hate and sorrow WinRT had going on, it was a Windows operating system running on Tegra (2 or 3, I can't remember). So yeah, Juan might be using Apple as an example (a very justified one), but his point stands valid: You can port software anytime you want. Most of those calls are political calls, not technical ones. Windows can't be in "ARM PCs", because there are no ARM CPU vendors like AMD or Intel so consumers can actually assemble one. AMD wants that to change, and it looks very promising.

About how capable ARM is, you might to read the paper Cazalan linked and I linked again some pages back. Your boss is very wrong in his appreciation. You might want to point that out to him, so he's not left in the dust when change comes knocking on his door.

Also, about the RAM speed thing... We don't call the speed of CPUs by their bus speed and call the multiplier from outside or separate, right? Potato.

Cheers!

EDIT: Typos.

 

[colinp]

Juanrga....

Really enjoy lots of your points, thanks for the info you have posted. But if I may....I notice over at semi-accurate you don't do all the self promotion with all the "as I predicted" and "like I said"... I I I. Would you mind talking more like that here. Not sure if your using this forum as your own for an ego boost or something but Its getting to me. I go around every few days and read all the AMD news and just keep running into you doing this here. You quote someone and then rant about how you said it. Everything you find is available at numerous sites. And if you do predict something and its right that's cool, know one is viewing you as an oracle or something, just take or give info without the self promotion please. Again thanks for the digging and the info but its clear that you do this for yourself, don't pretend your doing the world a favor. Its your hobby and that's cool, in this forum on this site do you mind just stopping the constant self promotion. Not sure if you think some of the old semi-accurate guys would put you in your place or something but you definitely talk in a different tone here like everyone should bow down.

Completely agree with this sentiment.

To be fair, though, Juanrga isn't the only one around here with a huge ego and who keeps repeating the old "I was right, bow down to my superior intellect". Gamerk and 8350rocks have both been repeat offenders, maybe there are others. This thread would be all the nicer if everyone would cut it out.

 

[etayorius]

Juanrga....

Really enjoy lots of your points, thanks for the info you have posted. But if I may....I notice over at semi-accurate you don't do all the self promotion with all the "as I predicted" and "like I said"... I I I. Would you mind talking more like that here. Not sure if your using this forum as your own for an ego boost or something but Its getting to me. I go around every few days and read all the AMD news and just keep running into you doing this here. You quote someone and then rant about how you said it. Everything you find is available at numerous sites. And if you do predict something and its right that's cool, know one is viewing you as an oracle or something, just take or give info without the self promotion please. Again thanks for the digging and the info but its clear that you do this for yourself, don't pretend your doing the world a favor. Its your hobby and that's cool, in this forum on this site do you mind just stopping the constant self promotion. Not sure if you think some of the old semi-accurate guys would put you in your place or something but you definitely talk in a different tone here like everyone should bow down.

Completely agree with this sentiment.

To be fair, though, Juanrga isn't the only one around here with a huge ego and who keeps repeating the old "I was right, bow down to my superior intellect". Gamerk and 8350rocks have both been repeat offenders, maybe there are others. This thread would be all the nicer if everyone would cut it out.

We can also throw noob in the mix, Gamerk and Noob will never ever give the slightest credit to AMD, Juan and 8350Rocks just keep throwing predictions, the only difference between them is that Juan smear it into everyone face`s when he`s right, and he usually is... while 8350Rocks seems to fail most of his predictions due to his AMD insider constantly giving him massively wrong info.

 

[juanrga]

It is worth mentioning that Windows RT fiasco has little to do with ARM, but is due to Microsoft mismanagement. The same mismanagement behind PS4 outperforming XB1 or linux destroying windows server

SA has an interesting article about all that

http://semiaccurate.com/2014/05/15/microsoft-now-irrelevant-computing-want-know/

 

[juanrga]

jWindows can't be in "ARM PCs", because there are no ARM CPU vendors like AMD or Intel so consumers can actually assemble one. AMD wants that to change, and it looks very promising.

Not only AMD. Nvidia also want ARM from phones to supercomputers:

ARM is already the standard architecture for mobile devices. Project Denver extends the range of ARM systems upward to PCs, data center servers, and supercomputers. ARM’s modern architecture, open business model, and vibrant eco-system have led to its pervasiveness in cell phones, tablets, and other embedded devices. Denver is the catalyst that will enable these same factors to propel ARM to become pervasive in higher-end systems.

Denver frees PCs, workstations and servers from the hegemony and inefficiency of the x86 architecture. For several years, makers of high-end computing platforms have had no choice about instruction-set architecture. The only option was the x86 instruction set with variable-length instructions, a small register set, and other features that interfered with modern compiler optimizations, required a larger area for instruction decoding, and substantially reduced energy efficiency.

Denver provides a choice. System builders can now choose a high-performance processor based on a RISC instruction set with modern features such as fixed-width instructions, predication, and a large general register file. These features enable advanced compiler techniques and simplify implementation, ultimately leading to higher performance and a more energy-efficient processor.

 

[jdwii]

jwdii... Remember Windows RT? For all the hate and sorrow WinRT had going on, it was a Windows operating system running on Tegra (2 or 3, I can't remember). So yeah, Juan might be using Apple as an example (a very justified one), but his point stands valid: You can port software anytime you want. Most of those calls are political calls, not technical ones. Windows can't be in "ARM PCs", because there are no ARM CPU vendors like AMD or Intel so consumers can actually assemble one. AMD wants that to change, and it looks very promising.

About how capable ARM is, you might to read the paper Cazalan linked and I linked again some pages back. Your boss is very wrong in his appreciation. You might want to point that out to him, so he's not left in the dust when change comes knocking on his door.

Also, about the RAM speed thing... We don't call the speed of CPUs by their bus speed and call the multiplier from outside or separate, right? Potato.

Cheers!

EDIT: Typos.

Yes and it was slower than anything now image that on a server controlling the login process of 500 computers.