WARNING: This website is obsolete! Please follow this link to get to the new Albert@Home website!
Running on ATI |
Message boards :
Problems and Bug Reports :
Running on ATI
Message board moderation
Previous · 1 · 2 · 3 · 4 · 5 · Next
Author | Message |
---|---|
Oliver Behnke Volunteer moderator Project administrator Project developer Send message Joined: 4 Sep 07 Posts: 130 Credit: 8,545,955 RAC: 0 |
Well, we haven't clearly identified a 4xxx specific problem so far. We're still investigating the overall numerical accuracy/correctness of our OpenCL apps in general though. Given AMD's driver quality that's no trivial, straight-forward task unfortunately. Please be patient with us. Oliver |
Millenium Send message Joined: 22 Jan 12 Posts: 2 Credit: 14,500 RAC: 0 |
Q9550 and 6950 here, driver 11.11, average GPU use 60%-70% |
Alex Send message Joined: 1 Mar 05 Posts: 88 Credit: 398,734 RAC: 0 |
Hi, I'm getting the following message: 22.01.2012 14:54:19 | Albert@Home | Nachricht vom Server: Neuere BOINC Version benötigt. Bitte die neueste Software installieren. My system is: 22.01.2012 09:31:02 | | Starting BOINC client version 7.0.8 for windows_x86_64 22.01.2012 09:31:02 | | log flags: file_xfer, sched_ops, task 22.01.2012 09:31:02 | | Libraries: libcurl/7.21.6 OpenSSL/1.0.0d zlib/1.2.5 22.01.2012 09:31:02 | | Data directory: C:\ProgramData\BOINC 22.01.2012 09:31:02 | | Running under account Service 22.01.2012 09:31:02 | | Processor: 5 AuthenticAMD AMD Phenom(tm) II X4 840 Processor [Family 16 Model 5 Stepping 3] 22.01.2012 09:31:02 | | Processor: 512.00 KB cache 22.01.2012 09:31:02 | | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 htt pni cx16 syscall nx lm svm sse4a osvw ibs skinit wdt page1gb rdtscp 3dnowext 3dnow 22.01.2012 09:31:02 | | OS: Microsoft Windows 7: Home Premium x64 Edition, Service Pack 1, (06.01.7601.00) 22.01.2012 09:31:02 | | Memory: 3.75 GB physical, 7.50 GB virtual 22.01.2012 09:31:02 | | Disk: 55.80 GB total, 22.67 GB free 22.01.2012 09:31:02 | | Local time is UTC +1 hours 22.01.2012 09:31:02 | | VirtualBox version: 4.1.8 22.01.2012 09:31:02 | | NVIDIA GPU 0: GeForce GT 430 (driver version 285.62, CUDA version 4.10, compute capability 2.1, 1024MB, 955MB available, 269 GFLOPS peak) 22.01.2012 09:31:02 | | ATI GPU 0: ATI Radeon HD 2300/2400/3200 (RV610) (CAL version 1.4.1016, 368MB, 336MB available, 112 GFLOPS peak) 22.01.2012 09:31:02 | | ATI GPU 1: ATI Radeon HD 2300/2400/3200 (RV610) (CAL version 1.4.1016, 368MB, 336MB available, 112 GFLOPS peak) Any ideas? Alexander |
Alex Send message Joined: 1 Mar 05 Posts: 88 Credit: 398,734 RAC: 0 |
OK, I found a hint in the HD7900 thread. BM 7.0.11 is latest version, but the boinc page lists only up to version 7.0.8 Has anyone a link to the newest version? |
TRuEQ & TuVaLu Send message Joined: 11 Sep 06 Posts: 75 Credit: 615,315 RAC: 0 |
Well, we haven't clearly identified a 4xxx specific problem so far. We're still investigating the overall numerical accuracy/correctness of our OpenCL apps in general though. Given AMD's driver quality that's no trivial, straight-forward task unfortunately. I posted earlier that my 4850 card had 512MB of memory so you said it would be lowered to 490MB But ageless showed me how to chech for free mem in Boincmanager. My 512MB card only has 480MB free Is there any chance that you can lower again to 450MB? |
Alex Send message Joined: 1 Mar 05 Posts: 88 Credit: 398,734 RAC: 0 |
Found all I need to know. Need a better AMD-card. |
TRuEQ & TuVaLu Send message Joined: 11 Sep 06 Posts: 75 Credit: 615,315 RAC: 0 |
Found all I need to know. I think you could have use of ATI/AMD Catalyst driver 11.12 or 12.1 for your 69xx card. |
Alex Send message Joined: 1 Mar 05 Posts: 88 Credit: 398,734 RAC: 0 |
Hi, not really, there are big problems coming up with the new drivers. On my AMD-System I've upgraded to newest Catalyst driver and BM 7.0.11 with the effect, that collatz is no longer running. Seti has rarely Astropulse wu's for AMD and the memory on my AMD-card is too small for Albert ATI open-cl apps. On my other system I run older drivers and older BM to be able to run MW, this project needs drivers < 11.10 An intresting fact is, that with the new HD7970 one can run the nVidia OpenCL-app, faster than the CAL-app on HD6970, without the need of latest BM. 6.12.34 is good for that ... So, after upgrading to latest available versions I have a GPU running dry. It's really a mess with these versions and requirements. |
TRuEQ & TuVaLu Send message Joined: 11 Sep 06 Posts: 75 Credit: 615,315 RAC: 0 |
Yes, it is a mess with the drivers, I agree on that. And Collatz is thinking of do an Open-Cl app too, but it will be slower then the old app they say. |
pragmatic prancing periodic problem child, left Send message Joined: 26 Jan 05 Posts: 1639 Credit: 70,000 RAC: 0 |
OK, I found a hint in the HD7900 thread. The Change Log Thread has links. Jord. BOINC FAQ Service They say most of your brain shuts down in cryo-sleep. All but the primitive side, the animal side. No wonder I'm still awake. |
EselTreiber Send message Joined: 29 Apr 08 Posts: 2 Credit: 48,003 RAC: 0 |
Today, AMD released a new beta driver, Catalyst 12.2. With this driver, Workunits are using 81% of the GPU (HD 6950@6970 full Shader). It could be even more, because the CPU limits the GPU-load. My Core i7-2600k ( 4.3GHz, only one Albert@home WU is running) is not able to deliver enough performance to bring GPU-load to 100%. If the CPU starts crunching other WUs (one core is exclusivly for Albert@home GPU-Workunit), the GPU-load drops to 75% and lower. Is there a chance, that CPU-load for GPU-WUs could be lowered in the future? |
Oliver Behnke Volunteer moderator Project administrator Project developer Send message Joined: 4 Sep 07 Posts: 130 Credit: 8,545,955 RAC: 0 |
OK, I found a hint in the HD7900 thread. FWIW, you'll find available BOINC packages here: http://boinc.berkeley.edu/dl Oliver PS: 7.0.12 is out |
Oliver Behnke Volunteer moderator Project administrator Project developer Send message Joined: 4 Sep 07 Posts: 130 Credit: 8,545,955 RAC: 0 |
If you ask whether our app's memory requirements might be reduced to 450 MB at some point, then probably since we always strive to support as many cards as possible by keeping our requirements as low as possible. Right now we can't due to other technical constraints. Oliver |
Oliver Behnke Volunteer moderator Project administrator Project developer Send message Joined: 4 Sep 07 Posts: 130 Credit: 8,545,955 RAC: 0 |
Yes, we have some ideas for that in the pipeline, it's just a matter of time to implement and test them. However, keep in mind that not all algorithms can be ported 100% to the GPU. This really depends on the problem and our current implementation is already not that bad after all. Right now we still need to focus on getting a first stable and numerically correct release out of the door -- and apart from our own code we largely depend on BOINC's readiness WRT OpenCL and proper AMD drivers, both are not easy to come by. Best Oliver |
pragmatic prancing periodic problem child, left Send message Joined: 26 Jan 05 Posts: 1639 Credit: 70,000 RAC: 0 |
Flagging a weird one. I caught this one running at zero GPU load for more than an hour, before I aborted it. http://albert.phys.uwm.edu/result.php?resultid=107592 It had been running for well over 4 hours and had an estimated time of another 4 hours, was at 41% done. Its progress was increasing. But still, I didn't trust it. Jord. BOINC FAQ Service They say most of your brain shuts down in cryo-sleep. All but the primitive side, the animal side. No wonder I'm still awake. |
pragmatic prancing periodic problem child, left Send message Joined: 26 Jan 05 Posts: 1639 Credit: 70,000 RAC: 0 |
OK, I ended up aborting all my work in cache, after I found a couple of things. a) For some unknown reason running an Albert task on my ATI GPU would make my internet connection stop working, leaving me with loads upon loads of page loading faults, even between same forum pages. Disable the Albert task and the internet would fly. b) Checking in GPU-Z I found that all these tasks would start with some GPU load before it tapering off to no load at all, or only a small bump every 5 minutes or so. I've seen this happen before with Albert GPU tasks. A reboot of the system didn't help in this matter. I'll test again after I found some I/O recording and debugging programs, to see if I can show you what I see. Jord. BOINC FAQ Service They say most of your brain shuts down in cryo-sleep. All but the primitive side, the animal side. No wonder I'm still awake. |
Alex Send message Joined: 1 Mar 05 Posts: 88 Credit: 398,734 RAC: 0 |
Hi, yesterday I installed my new system with two ATI-GPU's, latest BM 7.0.11 , latest AMD drivers and an outdated nVidia GPU (9500). I've seen that the GPU usage is very very low, so it makes sense to run more than one wu on every GPU. Is there an app_info available for that purpose? Alexander |
Oliver Behnke Volunteer moderator Project administrator Project developer Send message Joined: 4 Sep 07 Posts: 130 Credit: 8,545,955 RAC: 0 |
Hi Alex, Hi, We don't provide an app_info.xml but I'm sure people will help you prepare one yourself. Also, BOINC 7.0.12 is the latest and the GPU usage is usually between 60-80%. However, this is only the case when there is enough CPU power left to feed the GPU (see previous postings). Best, Oliver |
Alex Send message Joined: 1 Mar 05 Posts: 88 Credit: 398,734 RAC: 0 |
Hi Oliver, Thx for your reply. I checked the gpu usage over a longer period now; there are longer gaps with a usage between 0 and 3% (which caused me to say 'very very low'), shorter periods with ~35% and long times with a usage between 65 and 75%. This is the on-chip GPU of my FM1 Cpu. I got the whole system (Cpu, Mainboard and Ram) for ~250.- Euro. A good GPU costs about the same price and I don't want to run that one at ~30% as on my HD5830, this is a wast of resources and its expensive. Even for my on-chip gpu I calculate: 1 wu in ~ 7 hrs @ 65% average 2 wu's in ~ 11 hrs @ 95% average --> better performance. :-)) So if anyone can compile an app_info , PLEASE HELP! Alexander PS: intresting, the next wu is after 1:30:06 at 56%, quite faster than the first ones! BTW, the first two completed and validated, thats much better than the CPU wu's ... |
Oliver Behnke Volunteer moderator Project administrator Project developer Send message Joined: 4 Sep 07 Posts: 130 Credit: 8,545,955 RAC: 0 |
Now that's interesting. Can you post a link to that box? I'd like to take a look at the specs! Is this using an AMD APU or similar?
Good news :-) Oliver PS: we identified and fixed an issue that was causing a number of validation issues. v1.20 will be released short and should improve things. |