AMD Stream Processor launched, uses GPU power for general tasks

[tuteja1986]

http://www.tgdaily.com/2006/11/14/amd_stream_processor/

1GB DDR 3
R580 core

$2600

So is this worth it ;(

 

[Maximilian]

A review on this will be interesting.

 

[sandorski]

wow, I want my Co-processor!

 

[IEC]

<-- wishes he had a math co-processor for his brain

 

[Duvie]

Originally posted by: tuteja1986
http://www.tgdaily.com/2006/11/14/amd_stream_processor/

1GB DDR 3
R580 core

$2600

So is this worth it ;(

Read the article a bit more and place some context with that 2600 dollar price tag...Dont just throw it out for shock value...for its intended use in its market it is considered a deal...1/3 the cost for potentially 3 times the performance of a competitors accelerator board.

 

[myocardia]

Definitely not, IMO. For that price, you could buy four X1950XTX's, along with two C2D SLI motherboards, plus two E6600's. Each E6600 plus two X1950XTX's would come quite close to the performance of a single stream processor plus a Xeon/Opteron, I would think, if not bettering it.

 

[hardwareking]

I wonder how much better it'll perform at Folding@Home than a regular x1900 xt.

 

[myocardia]

Absolutely nowhere near 10 times as good, that's for sure. Unfortunately, it costs 10x as much as a single X1900XT.

 

[biostud]

Key to a successful implementation are applications that actually can take advantage of the processor. AMD simplifies the creation of such software by providing a new thin hardware interface which it calls CTM (for "close to metal"). The company claims that the technology can increase processing application performance by as much as eightfold more than traditional 3D application programming interfaces. CTM promises "unfettered access to the native instruction set and memory" of the GPU.

This will probably only work with the stream processor and not regular r580 boards, also the article mention a modified memory controller, this way they can charge $2600 for the processor.

 

[Lonyo]

Originally posted by: myocardia
Definitely not, IMO. For that price, you could buy four X1950XTX's, along with two C2D SLI motherboards, plus two E6600's. Each E6600 plus two X1950XTX's would come quite close to the performance of a single stream processor plus a Xeon/Opteron, I would think, if not bettering it.

Yeah, now add in some extra cost for power consumption and removing heat.
Performance per watt is where is important too, and 2 complete Crossfire systems are going to be a lot lower in performance per watt terms than a stream processor.

 

[TanisHalfElven]

now i really wonder if the things this THING can do, cannot be done better by G80 if nvidia provides the resources to devs.

 

[myocardia]

Originally posted by: Lonyo

Originally posted by: myocardia
Definitely not, IMO. For that price, you could buy four X1950XTX's, along with two C2D SLI motherboards, plus two E6600's. Each E6600 plus two X1950XTX's would come quite close to the performance of a single stream processor plus a Xeon/Opteron, I would think, if not bettering it.

Yeah, now add in some extra cost for power consumption and removing heat.
Performance per watt is where is important too, and 2 complete Crossfire systems are going to be a lot lower in performance per watt terms than a stream processor.

I fail to see how heat would be a big deal, since the X1950's exhaust their heat from the case. And power consumption would not make a big difference to anyone either, unless they're buying 1000's of these, it seems to me.

[/quote]Originally posted by: Biostud
This will probably only work with the stream processor and not regular r580 boards, also the article mention a modified memory controller, this way they can charge $2600 for the processor. [/quote]
This never crossed my mind, and I'm sure you're completely right about that.

 

[Munky]

Originally posted by: tanishalfelven
now i really wonder if the things this THING can do, cannot be done better by G80 if nvidia provides the resources to devs.

Of course the g80 will perform better, it all comes down to shader throughput. Also, I think this is just a regular r580 core with 1GB of mem and some special drivers/software. The hardware certainly doesnt seem to warrant a $2000+ pricetag compared to a regular x1900xt.

If Nvidia chooses to compete in this market (and I believe they will) then this could fuel even greater demand for high end gpu's, and certainly keep AMD developing new high end "steam processors". And when Intel decides to enter the market, things will get even more interesting.

 

[Lord Banshee]

I believe Nvidia has already decided to enter this market with the g80. It is called Cuda.
http://developer.nvidia.com/object/cuda.html

It isn't exactly the same but i do not see the benefit from AMD/ATI's way compared to the cheaper easier method employed by Nvidia. I have never used a GPU to do math for the purpose of math so i don't really know the insights.

 

[myocardia]

Originally posted by: Lord Banshee
I have never used a GPU to do math for the purpose of math so i don't really know the insights.

Haha, neither have any of us, that's why we're all guessing.:laugh:

edit: I of course meant hypothesizing.

 

[biostud]

ars technica take on it:
http://arstechnica.com/news.ars/post/20061115-8230.html

NVIDIA's G8800 presentation stressed that the company's overall approach to their new GPU product was that it's primarily a stream processor that can also be used as a GPU, and not just a GPU that can be used as a stream processor. I have to plead guilty to falling for the spin in my article on the architecture, but thankfully I was set straight via e-mail by someone who knows way more than me about this topic. There's plenty of dedicated, graphics-specific circuitry on the G8800?this thing was definitely designed from the ground up to push pixels, and it would be a lot slimmer if it didn't have to produce beautifully rendered 3D graphics. So from a hardware perspective, the G8800 is definitely a GPU with a unified shader model, but it has a shiny new stream-processing toolchain layered on top of it in addition to the existing graphics toolchain.

AMD was clearly watching the G8800 launch, and it probably reaffirmed their own plans to go ahead and play the "hey, it's not a mere GPU; it's a dedicated stream processor" with the launch of the AMD Stream Processor. The AMD Stream Processor is a repackaged ATI X1900 (R580) GPU with 1GB of GDDR3 DRAM on a standard PCIe daughtercard. In fact, it even looks like a typical graphics card, but it sells for a hefty $2,599.

AMD has been beta testing this hardware, and the related Close to Metal (CTM) initiative described below, for some time now. The card promises users 360GFLOPS of stream processing power.

In a lot of ways, this "is it mostly a stream processor, or is it just a GPU posing as a dedicated stream processor" discourse is just marketing-driven bunk. All modern GPUs are "stream processors" at their very core, but they have plenty of graphics-specific circuitry included in them that handles different parts of the rendering process. GPUs also come with an extensive toolchain to make graphics programming easier?a collection of APIs, languages, and other types of software abstraction that enable the programmer to use them specifically for rendering graphics.

Metal and remixes
To go with the "stream processor" rebranding effort, AMD has also launched its "Close to Metal" (CTM) initiative. No, this is not a VH1 80's hair band retrospective. What it is, is AMD opening up the very low-level parts of the R580 architecture that you'd typically never see as a graphics developer, because it's masked by the drivers. The point of doing this is to get academics and others who want to use GPUs as stream processors to write that missing "stream processing" abstraction layer themselves, and then to tweak it for maximum performance and maintain it.

One thing that I've learned over the years is that companies "open" up a technology (or parts of a technology) for one of three reasons:

Putting it out to pasture: they're through making money off of it, and they just want to give it away in the hopes that someone will do something useful out with it and that it won't go to waste. In the software world, this is called "abandonware."
Trying to create a de facto standard: they'd like everyone to use this, because they invented it and their competitor did not. This would presumably give them some kind of benefit if it becomes a de facto standard. Sun's Java is a cautionary tale here. Everybody uses Java, but the benefits to Sun's bottom line are pretty hard to quantify. For a hardware example of this phenomenon, see IBM's "opening" of the POWER architecture.
Getting someone else to do your work for you: they've got a foundation, but they're missing some critical pieces that they'd like "hobbyists" and "academics" other open source types to supply.
My initial, and admittedly somewhat superficial, impression of CTM is that it's a mix of numbers 2 and 3, but mostly 3. AMD/ATI want the HPC community to start buying their GPUs to do stream processing, so instead of enticing them with a complete toolchain (à la NVIDIA) they hope to suck them in by dropping the entire Lego box in their laps and inviting them to make cool things with all the parts.

I think AMD's approach may actually work, in this case. HPC users write their own code, and many of them live and die by a few cycles saved here and there. So many will appreciate having access to the bare metal. AMD claims that at least 60 companies are signed on to help develop tools for the stream processor via CTM.