HD 4xxx gpu's still getting wu's? {So far, yes}

Moderators: mhouston, Site Moderators, PandeGroup

Re: HD 4xxx gpu's still getting wu's? {So far, yes}

Postby MtM » Wed Feb 29, 2012 4:23 pm

Yeah it was competing, I lowered cpu usage as much as possible with the ati variables. The ppd from the slot was less then 400 though, and the competition on a core costs me much more. I never turned in unit's after preferred deadline, the gpu made every unit in approx the same time.

Code: Select all
PRCG:       Project:5733 run:3 clone:564 gen:1162
unit:       0x113477904f2ea1b8048a023400031665
client:       MARVIN-PC
Slot:       00
Hardware:    RV710 [Radeon HD 4350]
Core:       0x11
Corestatus:    FINISHED_UNIT (100 = 0x64)
Compiler:       Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86
BoardType:    AMD
Downloaded:    2012-02-05T15:35:08
Download size:   96.94KiB
Download speed:   61.7KiB/s
Completed:    2012-02-07T03:18:20
Submitted:    2012-02-07T03:18:24
Upload size:    246.79KiB
Upload speed:    61.7KiB/s
ServerResponce:    WORK_ACK (400)
Credit:       511
PPD:       343.33
Avg TPF:    00:21:00
Min TPF:       00:20:40
Max TPF:    00:22:31
WorkServer:    171.64.65.102


PRCG:       Project:5738 run:0 clone:614 gen:1121
unit:       0x651064524f2c8e91046102660000166a
client:       MARVIN-PC
Slot:       00
Hardware:    RV710 [Radeon HD 4350]
Core:       0x11
Corestatus:    FINISHED_UNIT (100 = 0x64)
Compiler:       Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86
BoardType:    AMD
Downloaded:    2012-02-04T01:48:56
Download size:   94.94KiB
Download speed:   64.07KiB/s
Completed:    2012-02-05T15:56:57
Submitted:    2012-02-05T15:57:01
Upload size:    256.29KiB
Upload speed:    64.07KiB/s
ServerResponce:    WORK_ACK (400)
Credit:       511
PPD:       321.6
Avg TPF:    00:22:27
Min TPF:       00:20:33
Max TPF:    02:18:31
WorkServer:    171.64.65.102

PRCG:       Project:5732 run:2 clone:574 gen:1163
unit:       0x74c78eec4f2a8c71048b023e00021664
client:       MARVIN-PC
Slot:       00
Hardware:    RV710 [Radeon HD 4350]
Core:       0x11
Corestatus:    FINISHED_UNIT (100 = 0x64)
Compiler:       Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 14.00.50727.762 for 80x86
BoardType:    AMD
Downloaded:    2012-02-02T13:15:22
Download size:   96.94KiB
Download speed:   82.18KiB/s
Completed:    2012-02-04T02:10:45
Submitted:    2012-02-04T02:10:48
Upload size:    246.53KiB
Upload speed:    82.18KiB/s
ServerResponce:    WORK_ACK (400)
Credit:       511
PPD:       332.14
Avg TPF:    00:21:43
Min TPF:       00:20:27
Max TPF:    00:56:18
WorkServer:    171.64.65.102


I was mainly running it because I wanted to have my setup as complicated as possible, I knew I was loosing production but I hoped the increased log complexity would help with getting my log parser as good as possible.

I can get enough logs from other people to make that a useless addition to my information, and I decided it was enough. My uniprocessor clients are making about the same ppd, so even if I place the card back in the htpc it probably isn't a good idea to fold on it ( though not sure about wattage, e2160 isn't a powerhog either ).

I didn't want to run smp:3 because some units didn't like it afaik.
MtM
 
Posts: 3054
Joined: Fri Jun 27, 2008 2:20 pm
Location: The Netherlands

Re: HD 4xxx gpu's still getting wu's? {So far, yes}

Postby bruce » Wed Feb 29, 2012 5:03 pm

MtM wrote:I didn't want to run smp:3 because some units didn't like it afaik.


Tell that to AMD who created the X3. ;)

Actually, Kasson has consistently said "large primes" and I wouldn't call 3 large. In fact, the lowest value for which I've seen problem reports is 7, and it's not consistent. Some projects don't like it; others run fine; but since you don't have any way to manage them (and without trial-and-error, nobody knows which might fail) 7 is not recommended. I don't remember any reports of any kind for 5.

For example, if you have an X6, Gromacs will partition the space into a 3x2 grid, so the factor 3 is still there.
bruce
 
Posts: 22853
Joined: Thu Nov 29, 2007 10:13 pm
Location: So. Cal.

Re: HD 4xxx gpu's still getting wu's? {So far, yes}

Postby MtM » Wed Feb 29, 2012 7:57 pm

7 and 11 are the number which came to mind as most problematic, 3 probably just sounded odd ( yes, I don't think I'd ever buy or had bought a broken quad.. uhm native triple core :lol: ).
MtM
 
Posts: 3054
Joined: Fri Jun 27, 2008 2:20 pm
Location: The Netherlands

Previous

Return to ATI specific issues

Who is online

Users browsing this forum: No registered users and 0 guests

cron