Microsoft Pressing Intel for 16-Core Atom

how bout asking amd for a 16 core zacate? it was highly scalable wasn't it?

A 16 core Atom seems like a neat idea, but the Atom's are lackluster performace-wise. A 16 core Atom against a Xeon LV? That seems like bring a knife to a gunfight.....even though it's a 16-bladed knife.

This is like a little kid asking dad for a jetpack, since he flies airplanes. Wants are cool, Microsoft, but realism often takes over after your 10th birthday. How about asking for a 16 core Ivybridge without the graphics core? I'd love to undervolt a sandybridge and compare it to an atom. Toms, take it away.

I think MS is asking for the wrong thing. I think what they really want is a Terascale based CPU like the 80 core or 48 core CPu that Intel has shown off. The 80 core, at the time did the same work as 130 CPUs while only using 62w load.

That is what they want because the Terascals is highly customizable unlike Atom.

Why would you use atom? It's a piece of crap processor, easily outclassed by AMD's Zacate Fusion APU (which, as mentioned earlier, is scalable).

...yet Microsoft went to intel instead of AMD.

intel runs the world, get used to it

correct me if im wrong, but isnt server side of things not dependent on speed but on how much data it can process? the 16 core may = the current ones in data processing power, but would use what is it, 1/5 the power?




arent those experimental and no where near mass production?




because code would need to be rewritten to take advantage of an apu, and that is a massive ammount of work, on par with porting it to arm.

Yeah the 48 core terrascale CPU isn't designed to be put into production, it's just for researchers.

Server code can't really run on the GPU part of a APU, at least not web/file server code. The GPU part of a APU can only run massively parallel floating point code, web/file servers don't use much if any floating point math so the APU would be of no help.

I could see a atom with 16 cores doing well in a cloud computing server. A major tenet of cloud computing is to use low power/low cost CPUs in large numbers and so a low power 16 core atom would be a good fit.

What about Tilera's Processors? those are 100 cores in a die.
http://www.tilera.com/products/pro [...] -Gx_Family

I've heard about those before. They look extremely powerful, although operating systems would have to be adapted and software recompiled.

lets put this in a way they people can understand better. a watt per core.

xenon 3.33 (3500-series "Bloomfield") would be at best, 16.25 watt per core, and at worst 32.5 watt per core, depending on if you count logical as being as good as a physical, i dont know how that works server side, but im assuming its better than desktop.
the best xeon, (6500/7500-series "Beckton") which i doubt they use, comes to 16.25 watt per core, on physical, and 8.125 watt per core if you count logical.

now the atom side, i went with the top model, and the lowest power model.

the lowest power, daimondvill dual core ht, comes in at 4watt per core, or, 2 watt if you go logical
and the best one, pineview, which comes in at 6.25 watt, and 2.75 watt pr core.

now lets take a look here, at a 1000 cores and in best conditions (these servers i assume use it in the 10000+cores per service minimum, probably that amount of cpus) now servers on the scale of hotmail and the like dont upgrade often, at least i dont think they would, but they make big upgrades, so Microsoft may be using even more inefficient processors.

but here, 1kwatt is 1000watt. and kwatt here will be measured in hours, and where i live 1kwatt equals 11 cents.

xeon 16.25kwatt = 1.7825$ an hour
xeon 8.125kwatt = .89375$ an hour
atom 2kwat = .22$ an hour
atom 2.75kwatt .3025$ an hour

lets blow this out, to 100000 cores to give you a better picture.
xeon = 178.25$ an hour 4,278$ a day 132,618$ a month 1,561,470$ a year
xeon = 89.375$ an hour 2,145$ a day 66,495$ a month 782,925$ a year
atom = 22$ an hour 528$ a day 16,368$ a month 192,720$ a year
atom = 30.25$ an hour 726$ a day 22,506$ a month 264,990$ a year

and this is powering the cpu alone, and those numbers are probably higher than i just mentioned. i just expanded it to the point people can see the real world difference

Did Microsoft wisper quietly or well use ARM... Sice they are writing Windows 8 to support ARM I can only assume server support for Windows Server will follow. I think this was a rather loud hint.

i should also point out that arm apparently has tapped out a 28nm chip, cabable of up to 2.8ghz, and current arm cpus use less than 4 watts total, some older ones useing less than a watt, so i cant say how many watts they can put out, but what i do know is this, an arm chip can be put in a ipod, or other devices, and make nearly no heat, compared to competitors, so putting lets say 20 or so of those on a motherboard (if possible, o know we have 4 cpu mbs right now, but i think they dont have more due to size constraints on the cooling they need, and arm are SIGNIFICANTLY smaller than its competitors.

But will it play Crysis????

yes and no. yes in the sense that its more than fast enough, and no, in the sense that the code wouldn't execute off that chip

So in article we talk about "Intel's Atom, AMD's Bobcat or Via's Nano." in the title only the (vastly inferior compared to competitors) Atom is mentioned.

Does AMD's Bobcat have a faster processor than Atom? I know it has a graphical processor in the APU but would that even get used in the server? If the answers are no and no then Bobcat is only better for consumers.

i don't get MS, aren't they using virtualization for that problem?

Yes Bobcat is faster than atom on the CPU side according to passmark, just compare the scores of the e-350 to the fastest desktop Atoms
E- 350 gets a 798
Atom D525 gets a 718
http://www.cpubenchmark.net/cpu_list.php

Its funny how Microsoft first slap Intel in the face by announcing they will run Windows on ARM, and then they ask Intel for favors. They probably ran a serious estimate to see if they could switch to ARM over 10 years. Once they see the current technical difficulties with ARM for servers (no 64 instruction set, and other). Then they probably rethought "well maybe we should stick with Intel for a while longer(on servers), but get then to do something about the heat in this (server) room!"

How else to say, we don't need you, but we "might" need you. There's still AMD, but they're being "polite".

I love it! This article is basically a variation of my theory...that the future of computers is multi-core MICROCONTROLLERS...NOT a CPU as we currently think of them. They will do away with graphics cards first. Then hard drives will cease being serial ATA and will all be PCI or some such. Then ram and bios will migrate to the CPU. Everthing is gradually converging onto one single die...ie a microcontroller.

RAM will not be integrated with the CPU any time soon, the manufacturing processes are too different. At least, they are with DRAM. MRAM or memristor based memory may change that in the future. Plus, RAM needs vary so much by application that being able to install different amounts of RAM is important for both cost and performance reasons. Someday, maybe, but I doubt it. RAM and storage will be separate for the foreseeable future, and likely indefinitely.

The fundamental performance limit of the Atom CPU is it's an in-order execution unit. Hyper-threading makes that less of a factor in a multi-threaded environment, but it will remain slower than the OoO Bobcat for each thread and slower overall. And it looks like the Bobcat will do it at lower power than the Atom. The VIA NANO is similarly fast, but not as power efficient. If AMD addresses memory bottlenecks on the current Bobcat based APUs, the Bobcat could completely smoke the Atom. If they also added SMT capabilities Atom would be completely outclassed.

Why recode anything? You don't have to rewrite software just to make Brazos actually work, in the same way you don't have to rewrite software just to make Atom work. They're both x86-64 architectures albeit on a smaller, more simplified scale.

hope these guys get a cache bump

Right now there's only ARM, or a board loaded with (64 or so) Z-series Atom processors.

I would love a quad core Atom, with independent core voltage and speeds, and core shut down of inactive cores.
As well as CPU and integrated graphics turbo boost technology.
If not turbo boost, at least allow users to overclock the atom by 20% (~2Ghz)!

Don't work with servers much, do you?

Actually..this won't happen. otherwise it would've happened a long time ago. This might happen for phones and other small devices, but for desktops, there will always be mainstream boards for sale with upgradeable features. Customers want the ability to upgrade, so the only thing manufacturers are going to do is make things more compact and faster, but still upgradeable.

AFAIK, Atom is much less power efficient than Nehalem. A 60W quad-core L-series Xeon will beat a 16-core Atom with the same TDP senseless on both multithreaded and (especially) single-threaded workloads. It seems that this is a continuation of the usual Micro$oft policy - "more shit for the consumers!"

P.S. Don't say that single-threaded performance is irrelevant for servers - responce time is often as important as throughput.