SoylentNews

A $1,499 Supercomputer on a Card?

posted by n1 on Thursday April 10 2014, @07:39PM   
from the will-it-play-crysis-though dept.

Bytram writes:

A $1,499 supercomputer on a card? That's what I thought when reading El Reg's report of AMD's Radeon R9 295X2 graphics card which is rated at 11.5 TFlop/s(*). It is water-cooled, contains 5632 stream processors, has 8 GB of DDR5 RAM, and runs at 1018MHz.

AMD's announcement claims it's "the world's fastest, period". The $1,499 MSRP compares favorably to the $2,999 NVidia GTX Titan Z which is rated at 8 TFlop/s.

From a quick skim of the reviews (at: Hard OCP, Hot Hardware, and Tom's Hardware), it appears AMD has some work to do on its drivers to get the most out of this hardware. The twice-as-expensive NVidia Titan in many cases outperformed it (especially at lower resolutions). At higher resolutions (3840x2160 and 5760x1200) the R9 295x2 really started to shine.

For comparison, consider that this 500 watt, $1,499 card is rated better than the world's fastest supercomputer listed in the top 500 list of June 2001.

(*) Trillion FLoating-point OPerations per Second.

• not necessarily not necessarily (Score: 3, Informative) by takyon on Thursday April 10 2014, @08:23PM

  by takyon (881) <reversethis-{gro ... s} {ta} {noykat}> on Thursday April 10 2014, @08:23PM (#29690) Journal

  AMD: [anandtech.com]

        AMD Radeon R9 295X2        AMD Radeon R9 290X     AMD Radeon HD 7990      AMD Radeon HD 7970 GHz Edition
  FP64  1/8                        1/8                     1/4                     1/4

  NVIDIA: [anandtech.com]

        GTX Titan Black            GTX 780 Ti              GTX Titan               GTX 780
  FP64  1/3 FP32                   1/24 FP32               1/3 FP32                1/24 FP32

  "Today NVIDIA is letting its compute-at-home customers have their cake and eat it too with the GeForce GTX Titan Black. The Titan Black is a full GK110 implementation, just like the GTX 780 Ti, with all of the compute focused-ness of the old GTX Titan. That means you get FP64 performance that's only 1/3 of the card's FP32 performance (compared to 1/24 with the 780 Ti)."

  Double-precision floating-point format [wikipedia.org]
  FLOPS - Floating-point operation and integer operation [wikipedia.org]

  --
  [SIG] 10/28/2017: Soylent Upgrade v14 [soylentnews.org]

  Starting Score:	1		point
  Moderation		+1
  Informative=1, Total=1
  Extra 'Informative' Modifier		0
  Karma-Bonus Modifier		+1
  Total Score:		3

• Re:not necessarily (Score: 1) by opinionated_science on Thursday April 10 2014, @08:39PM

  by opinionated_science (4031) on Thursday April 10 2014, @08:39PM (#29700)

  ahh , thank you for the summary of the crippled cards ;-)

  From what I understand DP=1/2 SP from a memory bandwidth point of view.

  Is there any technical reasons that the best FP64 performance is 1/3 FP32, other than marketing?

  Parent