Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
![]() |
World Community Grid Forums
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
No member browsing this thread |
Thread Status: Active Total posts in this thread: 23
|
![]() |
Author |
|
branjo
Master Cruncher Slovakia Joined: Jun 29, 2012 Post Count: 1892 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
When it comes to GPU-computing, I prefer to not come across surprises. There is not even one text mention of 'OpenCL' under the Catalyst driver 13.1 release-notes each for Windows and Linux. Is that a case of AMD doing an under-promise_over-deliver act so that a cruncher is set up for a pleasant suprise when he/she discovers GPU-crunching improvements that are not mentioned in the release-notes for Catalyst 13.1 driver? Hmm... ![]() ... ; ; andzgridPost#822 ; I have not read release note, but is you check AMD Driver Support page, you can find OpenCL(tm) Driver ver. 10.0.1084.4 in Description of 13.1. Catalyst Software Suite ![]() Cheers ![]() ![]() Crunching@Home since January 13 2000. Shrubbing@Home since January 5 2006 ![]() [Edit 1 times, last edit by branjo at Jan 24, 2013 4:27:03 PM] |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
It would have been as easy as just typing in the Google search box "ATI 13.1 drivers and OpenCL", and this top hit pops out at the AMD website... http://sites.amd.com/us/game/downloads/pages/radeon_win7-64.aspx ...
---------------------------------------- Display Driver ver. 9.012 OpenCL(tm) Driver ver. 10.0.1084.4 Catalyst Control Center ver. 2012.1219.1520.27485 Skipping forward. [Edit 1 times, last edit by Former Member at Jan 24, 2013 4:42:28 PM] |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I have not read release note, but is you check AMD Driver Support page, you can find One thing is, all those version numbers for the software items in that Catalyst package/suite do not tell a story about what improvements (or possibly new issues) are in store behind those numbers. A higher-numbered version does not guarantee that improvements were baked in. This is also true in the hardware front. Witness, for example, the industry practice of re-badging existing GPUs with higher numbers. All said, the only sure way to remove room for any ambiguity is to have the fixes, improvements, known-issues, and the likes documented in the release-notes and there is absolutely nothing for the Catalyst 13.1 package's release-notes that mentions anything that has a bearing with GPU-computing. If there are in fact improvements (directly or indirectly) for GPU-computing using the Catalyst 13.1 package, it surely is not documented.OpenCL(tm) Driver ver. 10.0.1084.4 in Description of 13.1. Catalyst Software Suite ![]() Cheers ![]() P.S. SekeRob, you dropped the ball again: confusing software description with release-notes. ; ; andzgridPost#823 ; [Edit 1 times, last edit by Former Member at Jan 24, 2013 5:19:49 PM] |
||
|
BladeD
Ace Cruncher USA Joined: Nov 17, 2004 Post Count: 28976 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
It's been well known for some time that the newer drivers increased framerates for some games by 20% or more.
---------------------------------------- |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
L.S. SekeRob, you dropped the ball again: confusing software description with release-notes. ; ; andzgridPost#823 ; When it comes to GPU-computing, I prefer to not come across surprises. ![]() [Edit 1 times, last edit by Former Member at Jan 24, 2013 5:29:14 PM] |
||
|
branjo
Master Cruncher Slovakia Joined: Jun 29, 2012 Post Count: 1892 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
I have not read release note, but is you check AMD Driver Support page, you can find One thing is, all those version numbers for the software items in that Catalyst package/suite do not tell a story about what improvements (or possibly new issues) are in store behind those numbers. A higher-numbered version does not guarantee that improvements were baked in. This is also true in the hardware front. Witness, for example, the industry practice of re-badging existing GPUs with higher numbers. All said, the only sure way to remove room for any ambiguity is to have the fixes, improvements, known-issues, and the likes documented in the release-notes and there is absolutely nothing for the Catalyst 13.1 package's release-notes that mentions anything that has a bearing with GPU-computing. If there are in fact improvements (directly or indirectly) for GPU-computing using the Catalyst 13.1 package, it surely is not documented.OpenCL(tm) Driver ver. 10.0.1084.4 in Description of 13.1. Catalyst Software Suite ![]() Cheers ![]() P.S. SekeRob, you dropped the ball again: confusing software description with release-notes. ; ; andzgridPost#823 ; [ot][jk]You are right. I have almost the same problem with every new day. But I do not try to avoid it ![]() Cheers and NI! ![]() ![]() Crunching@Home since January 13 2000. Shrubbing@Home since January 5 2006 ![]() [Edit 2 times, last edit by branjo at Jan 24, 2013 9:41:29 PM] |
||
|
coolstream
Senior Cruncher SCOTLAND Joined: Nov 8, 2005 Post Count: 475 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Thanks for the replies so far. My cards are XFX and Sapphire, so I have downloaded both versions along with the AMD version for comparison. I have calculated Hash values for all:
----------------------------------------AMD CRC32: ED7E4799 MD5: 6E3E13F59D9A6631B81D7E574044E656 SHA-1: C06DAA8A4D709A26C8770A3F959B0B4C42BABF38 Sapphire CRC32: B2D58C6B MD5: 6EBB49B2D15066F7F4AA05B433622672 XFX CRC32: B2D58C6B MD5: 6EBB49B2D15066F7F4AA05B433622672 SHA-1: D12062F459188ADA182FE92B1A07FD36A118BD59 Its strange that the AMD should have different values. Digital signatures of ALL versions are the same however. One new thing I have discovered is that there is no longer the option to check for updated drivers from within CCC. Instead, at the end of installation, you are taken to AMD site and invited to sign up for driver updates alert. (confirmed with XFX and Sapphire installation) I can only presume that since both of these driver update files are now giving update alerts from AMD, there must be no difference between them even if the AMD version had different Hash values. I am going to use the individual versions for the relevant cards in the meantime. ![]() Crunching in memory of my Mum PEGGY, cousin ROPPA and Aunt AUDREY. |
||
|
Rickjb
Veteran Cruncher Australia Joined: Sep 17, 2006 Post Count: 666 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
Upgraded to 13.1 for my 7870 a couple of days ago. Hadn't noticed any changes from the earlier version, but now that someone pointed it out I think screen responsiveness for general usage is improved.
Currently running 9 HCC1 WUs, fed by all 4 cores of a Q9650. Agree that AMD's lack of release notes is really slack. I run chip manufacturers' drivers for most devices, because the versions on the device integrators' sites often lag behind. And I don't necessarily want branding and advertising. I wonder whether WCG's use of AMD GPUs is now a significant enough proportion of the total usage of AMD's devices for GPGPU that management-level liason between WCG and AMD would be mutually beneficial. |
||
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Think what's at WCG is not even a fraction, except if WCG[IBM] were to speak to them and tell them of the science application, the flappy ears would instantaneously perk up... but, you'd never get to know if they would. Anyone ever went to the AMD support forums and doing a sales pitch of the HCC-GPU success-story?
----------------------------------------edit: Rickjb, what's the actual system juice use differential at wallsocket of your 7870 between idle and running 9, ideally measured with this 13.1 driver suite? [Edit 1 times, last edit by Former Member at Jan 25, 2013 8:40:06 AM] |
||
|
Falconet
Master Cruncher Portugal Joined: Mar 9, 2009 Post Count: 3295 Status: Offline Project Badges: ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---------------------------------------- AMD Ryzen 5 1600AF 6C/12T 3.2 GHz - 85W AMD Ryzen 5 2500U 4C/8T 2.0 GHz - 28W AMD Ryzen 7 7730U 8C/16T 3.0 GHz |
||
|
|
![]() |