Most Popular
| Top Stories | Commented | Featured |
ATI Radeon HD 5570 Review featured
AMD's six-core Thuban to have feature like Turbo Boost?
Google to launch Twitter-like service for Gmail
Intel unveils Itanium 9300 series enterprise processors
Netflix to roll out 1080p streaming later this year
China closes major hacker ring, arrests three members
Sharp and Samsung end LCD patent suits with cross-licensing agreement
TS Community
| User Gallery | Recent Discussion |
 My Gaming Rig by Moltar |  New NASA Photo = New Desktop Pic by luvhuffer |
 Techmax v1 / pic 3 by cookieboy |  HAWAIIAN RAINBOW by earthlostangel |
TechSpot RSSHardware
Nvidia unveils Fermi-based Tesla GPUs
Nvidia has unveiled new Fermi-based Tesla products, the Tesla 20 series GPUs, which are aimed at the high-performance computing market. Designed for parallel computing, the company said its Tesla 20 series graphics chips reduce the cost of computing by delivering the same performance of a traditional CPU-based cluster at one-tenth the expense and one-twentieth the power consumption.
Nvidia's Tesla 20 series chips are reportedly equipped with features that speed up many applications, such as ray tracing, 3D cloud computing, video encoding, database search, data analytics, computer-aided engineering, and virus scanning. The new GPUs will also combine various parallel computing features that have never been incorporated on a single device before.
Among the features is support for the next-generation IEEE 754-2008 double precision floating point standard, ECC, multi-level cache hierarchy with L1 and L2 caches, and support for the C++ programming language. The Tesla 20 series also supports as much as one terabyte of memory, concurrent kernel execution, fast context switching, 10x faster atomic instructions, 64-bit virtual address space, system calls, and recursive functions.
Nvidia's new graphics lineup will consist of the Tesla C2050 and C2070, as well as the S2050 and S2070. The first two will be workstation-friendly single GPU PCIe 2.0 cards with up to 3GB and 6GB of GDDR5 memory and performance in the range of 520GFlops to 630GFlops. The latter two will be aimed at datacenters, boasting four GPUs, up to 12GB and 24GB of GDDR5 memory, and a performance of 2.1TFlops to 2.5TFlops.
In order of mention, the Tesla 20 series GPUs will cost $2,499, $3,999, $12,995, and $18,995 -- all will arrive in the second quarter of 2010.
Nvidia's Tesla 20 series chips are reportedly equipped with features that speed up many applications, such as ray tracing, 3D cloud computing, video encoding, database search, data analytics, computer-aided engineering, and virus scanning. The new GPUs will also combine various parallel computing features that have never been incorporated on a single device before.
Among the features is support for the next-generation IEEE 754-2008 double precision floating point standard, ECC, multi-level cache hierarchy with L1 and L2 caches, and support for the C++ programming language. The Tesla 20 series also supports as much as one terabyte of memory, concurrent kernel execution, fast context switching, 10x faster atomic instructions, 64-bit virtual address space, system calls, and recursive functions.
Nvidia's new graphics lineup will consist of the Tesla C2050 and C2070, as well as the S2050 and S2070. The first two will be workstation-friendly single GPU PCIe 2.0 cards with up to 3GB and 6GB of GDDR5 memory and performance in the range of 520GFlops to 630GFlops. The latter two will be aimed at datacenters, boasting four GPUs, up to 12GB and 24GB of GDDR5 memory, and a performance of 2.1TFlops to 2.5TFlops.
In order of mention, the Tesla 20 series GPUs will cost $2,499, $3,999, $12,995, and $18,995 -- all will arrive in the second quarter of 2010.
Related Stories
User Comments (13)
Post a comment| red1776 on November 16, 2009 3:44 PM | at least this one doesn't seem to have wood screws holding
it together. 
|
| poundsmack on November 16, 2009 4:01 PM | the future is looks bright for nvidia
|
| raybay on November 16, 2009 4:16 PM | Well, it is not that sure what the nVidia future is like... but we wish them well as they go through their travails. |
| technochicken on November 16, 2009 4:33 PM | If the first two are aimed at high power workstations, then isn't nVidia competing with itself? They already has the Quadro series for workstations, and I'm sure they will update that line when the new GPU is released. |
| 9Nails on November 16, 2009 4:44 PM | These cards are more for render farms than desktop computing right? I'm guessing in this direction because with all the memory that these cards come with, a single DVI connector looks inadequate for a Quadro type replacement. (Unless it's one of those weird ones that have an octopus cable adapter.) |
| Guest on November 16, 2009 5:28 PM | Nvidia's future is dictated by the current state of the world economy. |
| Guest on November 16, 2009 7:07 PM | Nvidia should partner up with Vray / MentalRay people to help those engines render faster. They need to release more specific information before I set a budget for next year! |
| Timonius on November 16, 2009 8:24 PM | Oh man, I wish that I could wish that I could afford one of these babies! Suh-weeeeet! |
| BlindObject on November 16, 2009 8:41 PM | If they only made such art for gaming... |
| Guest on November 16, 2009 10:09 PM | Anyone want to comment on the statement that there will come in Q2 2010? That's... pretty late! |
| Guest on November 17, 2009 12:47 AM | these are way tooo cheap to be a part of my PC |
| ET3D on November 17, 2009 12:48 AM | "one terabyte of memory"? That's a lot. Also, I agree Q2 sounds a little late, and they haven't yet released any info of their consumer level chips. Hopefully that's coming soon. |
| Puiu on November 17, 2009 5:35 AM | I was expecting them to have more than 2.5 TFLOPs. Doesn't AMD's 5870 have 2.72 TFLOPs? Let's just hope that the 300 series will be good enough to compete with the 5000 series. I want those high prices of dx11 cards to drop a bit before i buy something. |
