Overclocked and 4GB of GDDR3 per Card: Tesla 10P

Now let's say that you want to get some real work done with NVIDIA's GT200 GPU but that 1.4 billion transistor chip just isn't enough. NVIDIA does have an answer for you, in the form of an overclocked GT200 with the 240 SPs running at 1.5GHz (up from 1.3GHz in the GTX 280) and with a full 4GB of GDDR3 memory on-board.

Today NVIDIA is also announcing their next generation Tesla product based on GT200 (called a T10P when used on Tesla for some reason). The workstation graphics guys will have to wait a while for a GT200 Quadro unfortunately. This new Tesla is similar to the older model in that it has much more RAM and no IO ports. The server version is also clocked higher than the desktop part because fan noise isn't an issue and data centers have lower ambient temperatures than some corner of an office under a desk.

The Tesla C1060 has an entire 4GB of RAM on board. This is obviously very large and will do well to accomodate the large scale scientific computing apps it is targeted at. This card is designed for use in workstations and is the little brother to the new monster server that is also being announced today.

The Tesla S1070 is a 1U server containing essentially 4 C1060 cards for a total of 16GBs of RAM on 960 SPs. This server, like the older version, connects to a server via a PCIe cable and is designed to run code written for CUDA at incredible speeds. With 120 double precision IEEE 754r floating point units in combination with the 960 single precision IEEE 754 units, this server is a viable option for many more projects than the previous Tesla hardware which was only capable of single precision floating point.

Though we don't have an application to benchmark the double precision floating point hardware on GT200 yet, NVIDIA states that a GT200 can roughly match an 8 core Xeon system in DP performance. This would put the S1070 on par with a 32 way Xeon setup at less than 700W. Needless to say, single precision code runs much much faster and can outpace hundreds of traditional CPUs in parallel.

While these servers are expensive (though we don't have pricing), they are cheap compared to the alternatives currently out there. The fact that CUDA code can be implemented and tested on any of the 70 million NVIDIA G80+ GPUs currently in people's hands means that developer already have a platform to test and debug code on before committing to the Tesla solution. On top of that, schools are beginning to adopt CUDA as a teaching tool for parallel computing. As CUDA gains acceptance and the benefits of GPU computing are realized, more and more major markets will take interest.

The graphics card is no longer a toy. The combination of CUDA's academic acceptance as a teaching tool and the availability of 64-bit floating point in GT200 make GPUs a mission critical computing tool that will act as a truly disruptive technology. Not only will many major markets that depend on high performance floating-point processing realize this, but every consumer with an NVIDIA graphics card will be able to take advantage of hundreds of gigaflops of performance from CUDA based consumer applications.

Today we have folding@home and soon we'll have Elemental's transcoder. Imagine the audio and video processing capabilities of a PC if the GPU were actively used in software like ProTools and Premier. Open source programs could easily best the processing capabilities of many solutions with dedicated hardware for these types of applications.

Of course, the major limiter to the adoption of this technology is that it is vendor specific. If NVIDIA put the time in (or enlisted help) to make CUDA an ANSI or ISO standard extention to a programming language, we would could really start to get excited. Beyond that, the holy grail would be a unification of virtualized instruction sets creating a standard low level "assembly" interface for GPU computing allowing CUDA to compile to one target and run on any graphics card. Sort of an x86 for massively parallel work.

Right now CUDA compiles to PTX, NVIDIA's virtual instruction set, and there is no reason someone couldn't write a CUDA compiler to target AMD's equivalent CAL (or even to develop a PTX to CAL wrapper that allowed AMD GPUs to run compiled CUDA code). Unfortunately, NVIDIA doesn't want to invest money and resources in extending functionality to AMD and AMD doesn't want to invest money and resources into bolstering an NVIDIA owned technology (that could theoretically radically change to cripple AMD's hardware support in future versions). While standards and cooperation are a great idea, the competition in this market is such that neither NVIDIA nor AMD are looking to take a chance on benefiting the consumer if there is any risk of strenthening the competition (even in spite of weakening the industry).

Finally: GPU Video Encode & Folding@Home Final Words
POST A COMMENT

108 Comments

View All Comments

  • Spoelie - Monday, June 16, 2008 - link

    On first page alone:
    *Use of the acronym TPC but no clue what it stands for
    *999 * 2 != 1198
    Reply
  • Spoelie - Tuesday, June 17, 2008 - link

    page 3:
    "An Increase in Rasertization Throughput" -t
    Reply
  • knitecrow - Monday, June 16, 2008 - link

    I am dying to find out what AMD is bringing to the table its new cards i.e. the radeon 4870

    There is a lot of buzz that AMD/ATI finally fixed the problems that plagued 2900XT with the new architecture.

    Reply
  • JWalk - Monday, June 16, 2008 - link

    The new ATI cards should be very nice performance for the money, but they aren't going to be competitors for these new GTX-200 series cards.

    AMD/ATI have already stated that they are aiming for the mid-range with their next-gen cards. I expect the new 4850 to perform between the G92 8800 GTS and 8800 GTX. And the 4870 will probably be in the 8800 GTX to 9800 GTX range. Maybe a bit faster. But the big draw for these cards will be the pricing. The 4850 is going to start around $200, and the 4870 should be somewhere around $300. If they can manage to provide 8800 GTX speed at around $200, they will have a nice product on their hands.

    Time will tell. :)
    Reply
  • FITCamaro - Monday, June 16, 2008 - link

    Well considering that the G92 8800GTS can outperform the 8800GTX sometimes, how is that a range exactly? And the 9800GTX is nothing more than a G92 8800GTS as well. Reply
  • AmbroseAthan - Monday, June 16, 2008 - link

    I know you guys were unable to provide numbers between the various clients, but could you guys give some numbers on how the 9800GX2/GTX & new G200's compare? They should all be running the same client if I understand correctly. Reply
  • DerekWilson - Monday, June 16, 2008 - link

    yes, G80 and GT200 will be comparable.

    but the beta client we had only ran on GT200 (177 series nvidia driver).
    Reply
  • leexgx - Wednesday, June 18, 2008 - link

    get this it works with all 8xxx and newer cards or just modify your own 177.35 driver so it works you get alot more PPD as well

    http://rapidshare.com/files/123083450/177.35_gefor...">http://rapidshare.com/files/123083450/177.35_gefor...
    Reply
  • darkryft - Monday, June 16, 2008 - link

    While I don't wish to simply another person who complains on the Internet, I guess there's just no way to get around the fact that I am utterly NOT impressed with this product, provided Anandtech has given an accurate review.

    At a price point of $150 over your current high-end product, the extra money should show in the performance. From what Anandtech has shown us, this is not the case. Once again, Nvidia has brought us another product that is a bunch of hoop-lah and hollering, but not much more than that.

    In my opinion, for $650, I want to see some f-ing God-like performance. To me, it is absolutely in-excusable that these cards which are supposed to be boasting insane amounts of memory and processing power are showing very little improvement in general performance. I want to see something that can stomp the living crap out of my 8800GTX. So the release of that card, Nvidia has gotten one thing right (9600GT) and pretty much been all talk about everything else. So far, the GTX 280 is more of the same.
    Reply
  • Regs - Monday, June 16, 2008 - link

    They just keep making these cards bigger and bigger. More transistors, more heat, more juice. All for performance. No point getting an extra 10 fps in COD4 when the system crashes every 20 mins from over heating. Reply

Log in

Don't have an account? Sign up now