Top GPU models

Message boards : Number crunching : Top GPU models
Message board moderation

To post messages, you must log in.

AuthorMessage
Jeffery

Send message
Joined: 24 May 99
Posts: 17
Credit: 40,925,202
RAC: 6
United States
Message 1803345 - Posted: 18 Jul 2016, 20:13:32 UTC

It looks as if the Top GPU models page was updated today. See it here: https://setiathome.berkeley.edu/gpu_list.php

I'm looking at the NVIDIA Widows column and am wondering if it is accurate. How can number 10 be the (0.632) GeForce GTX 780 Ti, and number 2 be the (0.817) GeForce GTX 670?

The 780 Ti beats the 670 on CUDA cores and memory speed, and is just a bit slower on clock speed. See specs:

780 Ti http://www.geforce.com/hardware/desktop-gpus/geforce-gtx-780-ti/specifications
670 http://www.geforce.com/hardware/desktop-gpus/geforce-gtx-670/specifications

Is the 670 really better thanks the 780 Ti for SETI?
ID: 1803345 · Report as offensive
Cruncher-American Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor

Send message
Joined: 25 Mar 02
Posts: 1513
Credit: 370,893,186
RAC: 340
United States
Message 1803348 - Posted: 18 Jul 2016, 20:37:47 UTC

Those scores aren't really rating the cards themselves, I believe, but how much work they do for SETI in aggregate.

Perhaps I am wrong on this, and I would appreciate a correction and explanation, if due.
ID: 1803348 · Report as offensive
Jeffery

Send message
Joined: 24 May 99
Posts: 17
Credit: 40,925,202
RAC: 6
United States
Message 1803349 - Posted: 18 Jul 2016, 20:43:00 UTC - in response to Message 1803348.  

I'd love to learn more. I appreciate the Top GPU models page but was confused by the ranking since it contradicted the most often quoted performance stats of graphics cards.

I'm sure someone here will know why the rankings are the way they are.
ID: 1803349 · Report as offensive
Profile Mike Special Project $75 donor
Volunteer tester
Avatar

Send message
Joined: 17 Feb 01
Posts: 34253
Credit: 79,922,639
RAC: 80
Germany
Message 1803354 - Posted: 18 Jul 2016, 21:25:55 UTC

Those scores are allways theoretical.
What card is in use doesn`t necessarily mean how efficient it is used.
My R9 380 for example has at leat a 50% higher output than the scores would suggest.


With each crime and every kindness we birth our future.
ID: 1803354 · Report as offensive
Profile Stubbles
Volunteer tester
Avatar

Send message
Joined: 29 Nov 99
Posts: 358
Credit: 5,909,255
RAC: 0
Canada
Message 1803369 - Posted: 18 Jul 2016, 23:27:06 UTC - in response to Message 1803349.  

I'd love to learn more. I appreciate the Top GPU models page but was confused by the ranking since it contradicted the most often quoted performance stats of graphics cards.

I'm sure someone here will know why the rankings are the way they are.

Take a look at Shaggie's great work!!!
that should help you out
ID: 1803369 · Report as offensive
Jeffery

Send message
Joined: 24 May 99
Posts: 17
Credit: 40,925,202
RAC: 6
United States
Message 1803393 - Posted: 19 Jul 2016, 2:18:44 UTC - in response to Message 1803369.  

Thank you! @Shaggie76's work is really appreciated.
ID: 1803393 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13720
Credit: 208,696,464
RAC: 304
Australia
Message 1803446 - Posted: 19 Jul 2016, 7:47:09 UTC - in response to Message 1803438.  
Last modified: 19 Jul 2016, 7:59:06 UTC

I wasn't aware of the big difference between Nvidia and AMD on vlar files.

http://setiathome.berkeley.edu/workunit.php?wuid=2214140528

1080 vs Fury X

The Nvidia maybe running two at a time. I am only running one at a time but the CPU time is a big difference.

Al is running 4 WUs at a time.

EDIT, that gives him (roughly)
4.8 VLARs per hour
and you around
2.6 VLARs per hour.
Grant
Darwin NT
ID: 1803446 · Report as offensive

Message boards : Number crunching : Top GPU models


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.