You can get the total memory using 'amdcccle', and gpu load using 'aticonfig --odgc'; I can't find a way to determine memory used/available short of writing a bit of code to do it.
Clearly BOINC contains the code to determine memory usage, since I see this at start-up:
19-Jun-2013 13:33:40 [---] CAL: ATI GPU 0: ATI Radeon HD 5800 series (Cypress) (CAL version 1.4.1741, 1024MB, 981MB available, 3584 GFLOPS peak)
Did some testing this weekend, with my new videocard, which I'll have to return tomorrow, because it's definitely broken.
BRP5 measurements.
Sapphire HD6850 2GB, Windows 7 64bit Ultimate SP1, Catalyst 13.4
HD6850 on average over 15 tasks, 17,546.13 seconds.
Sapphire HD7870 2GB, Windows 7 64bit Ultimate SP1, Catalyst 13.6-Beta2.
HD7870 on average over 9 tasks, 10147.07 seconds.
Why only 9 tasks? Well, this afternoon I removed the HD7870 and tomorrow I'll return it to the store, due to it having problems. Since I put it in I have seen wavy text whenever the card hit 59C. In Crysis3, it runs at 83C, even though it has a dual-fan. The HD6850 runs with the same setup at 70C (and it has a single fan). I've also seen definite text corruption. So the card is defective, something wrong with its RAMDAC or memory. It's going back, but I hope to get a new one as going from 17K to 10K is definitely a big thumbs up.
Hi Ageless,
It would be most helpful if you stated what CPU these tests were run on, if there is any overclocking involved (cpu & gpu) and what was running on the CPU during the test.
Gord
Since the time my online AMD cruncher downloaded the Linux AMD BRP5 version 1.36 application, I have seen a difference in run time compared to the version 1.34 application.
I averaged out 10 tasks for each case. Since I have cards installed in both x16 and x8 slots, I included run times for each.
v1.36
x16 3.0: 11,953.05
x8 3.0: 13,365.33
v1.34
x16 3.0: 12,316.96
x8 3.0: 14,486.49
Unfortunately, I had to turn off my other AMD cruncher due to extreme record breaking heat that we have here and I have not been able to try version 1.36 on this system yet. It has been a challenge keeping the temperatures at a reasonable level during the day with one system crunching.
Hi Ageless,
It would be most helpful if you stated what CPU these tests were run on, if there is any overclocking involved (cpu & gpu) and what was running on the CPU during the test.
i5-2500K, BOINC allowed to use 3 cores for Seti v7, last core free for GPU. No overclocking. Only one task per core and GPU.
You can get the total memory
)
You can get the total memory using 'amdcccle', and gpu load using 'aticonfig --odgc'; I can't find a way to determine memory used/available short of writing a bit of code to do it.
Clearly BOINC contains the code to determine memory usage, since I see this at start-up:
19-Jun-2013 13:33:40 [---] CAL: ATI GPU 0: ATI Radeon HD 5800 series (Cypress) (CAL version 1.4.1741, 1024MB, 981MB available, 3584 GFLOPS peak)HD7870 pitcairn at 900MHz GPU
)
HD7870 pitcairn at 900MHz GPU and 900MHz VRAM finishes 5 tasks simultaneously nearly 10 hours.
Tasks are all "Binary Radio Pulsar Search(Perseus Arm Survey) 1.34(opencl-ati)"
Opencl Enthusiast
GeForce 320.49 Beta Drivers
)
GeForce 320.49 Beta Drivers Released
A couple days ago but I just got the email from them.
http://www.geforce.com/whats-new/articles/nvidia-geforce-320-49-beta-drivers-released
As usual I just downloaded the previous version last week.
Did some testing this
)
Did some testing this weekend, with my new videocard, which I'll have to return tomorrow, because it's definitely broken.
BRP5 measurements.
Sapphire HD6850 2GB, Windows 7 64bit Ultimate SP1, Catalyst 13.4
HD6850 on average over 15 tasks, 17,546.13 seconds.
Sapphire HD7870 2GB, Windows 7 64bit Ultimate SP1, Catalyst 13.6-Beta2.
HD7870 on average over 9 tasks, 10147.07 seconds.
Why only 9 tasks? Well, this afternoon I removed the HD7870 and tomorrow I'll return it to the store, due to it having problems. Since I put it in I have seen wavy text whenever the card hit 59C. In Crysis3, it runs at 83C, even though it has a dual-fan. The HD6850 runs with the same setup at 70C (and it has a single fan). I've also seen definite text corruption. So the card is defective, something wrong with its RAMDAC or memory. It's going back, but I hope to get a new one as going from 17K to 10K is definitely a big thumbs up.
Leaving measurements for comparison for others.
Hi Ageless, It would be most
)
Hi Ageless,
It would be most helpful if you stated what CPU these tests were run on, if there is any overclocking involved (cpu & gpu) and what was running on the CPU during the test.
Gord
Since the time my online AMD
)
Since the time my online AMD cruncher downloaded the Linux AMD BRP5 version 1.36 application, I have seen a difference in run time compared to the version 1.34 application.
I averaged out 10 tasks for each case. Since I have cards installed in both x16 and x8 slots, I included run times for each.
v1.36
x16 3.0: 11,953.05
x8 3.0: 13,365.33
v1.34
x16 3.0: 12,316.96
x8 3.0: 14,486.49
Unfortunately, I had to turn off my other AMD cruncher due to extreme record breaking heat that we have here and I have not been able to try version 1.36 on this system yet. It has been a challenge keeping the temperatures at a reasonable level during the day with one system crunching.
RE: Hi Ageless, It would be
)
i5-2500K, BOINC allowed to use 3 cores for Seti v7, last core free for GPU. No overclocking. Only one task per core and GPU.
Thank you!
)
Thank you!
where did you get v1.36?
)
where did you get v1.36?
RE: where did you get
)
The application downloaded automatically. However, the application was reverted back to version 1.34 recently.