NVIDIA has good laugh with "FASTRA" Super Computer

@ 2010/02/03

Muehahaha
Comment from Kougar @ 2010/02/05
Quote:
Originally Posted by wutske View Post
The first CUDA cards introduced a while ago also sucked at DP, so Tesla cards aren't the reason why the 'older' cards don't do DP very well.
DP is indeed interesting for scientific groups who want accuracy, but sometimes real-time processing is more important that a bit more accurate results.
Why are you comparing to older cards? GF100 hasn't yet been sold as CUDA/Tesla cards, so it's rather silly to compare to past models. GF100 was designed especially for DP performance unlike GT200.

DP FMA rate:
GT200 offers 88.5 Gflops
GF100 Tesla offers 744 Gflops
GF100 GeForce offers 186 Gflops
(Source TheTechReport)

I'm not sure what percentage of the market writes for SP FMA and runs calculations twice versus DP FMA and runs them once, but there are enough organizations that built GT200 GeForce computing platforms that this will certainly affect quite a few of them.
Comment from wutske @ 2010/02/05
Quote:
Originally Posted by Kougar View Post
As most scientific computing groups consider accuracy to be paramount most of them use double-precision workloads. It was specifically done to ensure groups couldn't build FASTRA computers without paying more for Tesla cards. They'd get better performance from RV870 instead of GeForce now.
The first CUDA cards introduced a while ago also sucked at DP, so Tesla cards aren't the reason why the 'older' cards don't do DP very well.
DP is indeed interesting for scientific groups who want accuracy, but sometimes real-time processing is more important that a bit more accurate results.
Comment from Kougar @ 2010/02/04
Quote:
Originally Posted by wutske View Post
Then don't use DP
As most scientific computing groups consider accuracy to be paramount most of them use double-precision workloads. It was specifically done to ensure groups couldn't build FASTRA computers without paying more for Tesla cards. They'd get better performance from RV870 instead of GeForce now.
Comment from jmke @ 2010/02/04
DP is useful because it delivers more accurate results

Quote:
Refers to a type of floating-point number that has more precision (that is, more digits to the right of the decimal point) than a single-precision number. The term double precision is something of a misnomer because the precision is not really double. The word double derives from the fact that a double-precision number uses twice as many bits as a regular floating-point number. For example, if a single-precision number requires 32 bits, its double-precision counterpart will be 64 bits long.

The extra bits increase not only the precision but also the range of magnitudes that can be represented. The exact amount by which the precision and range of magnitudes are increased depends on what format the program is using to represent floating-point values.
Comment from wutske @ 2010/02/04
Quote:
Originally Posted by Kougar View Post
Won't be the case anymore... double-precision performance has been limited to 1/4th performance that is capable on GeForce cards... have to buy Tesla models for full performance. RV870 will perform better than GeForce models because of this in DP work.
Then don't use DP
Comment from Kougar @ 2010/02/03
Quote:
Originally Posted by thorgal View Post
OK, so they can stop laughing now. Tones considered Tesla's for FASTRA II, but it was 5x as expensive for the same performance.
Won't be the case anymore... double-precision performance has been limited to 1/4th performance that is capable on GeForce cards... have to buy Tesla models for full performance. RV870 will perform better than GeForce models because of this in DP work.
Comment from Massman @ 2010/02/03
Actually, Tegra is the mobile name ... not terga
Comment from thorgal @ 2010/02/03
OK, so they can stop laughing now. Tones considered Tesla's for FASTRA II, but it was 5x as expensive for the same performance.
Comment from jmke @ 2010/02/03
was wrong with product name, it's Tesla
terga = mobile product

$1200 per card


$4000 for 1U server with 2xTesla's


build server $6000


etc
Comment from Massman @ 2010/02/03
The long wait
Comment from thorgal @ 2010/02/03
And its price ?
Comment from jmke @ 2010/02/03
this is a picture of NVIDIA Tesla product line :-)
Comment from thorgal @ 2010/02/03
More facts ?

 

 

Copyright © 2001-2011 Madshrimps, All rights reserved.
Graphic Design by Dennis Kestelle, Programming by Maarten Menten, Overall Site design by John Meys

All information and graphics contained in Madshrimps are sole property of the Madshrimps crew and may not be reproduced or copied in any manner without written permission from us.

Madshrimps