21)
Message boards :
Number crunching :
GPU Requirements
(Message 52496)
Posted 14 Jan 2012 by Beyond Post: Hi Peter, ideally set your preferences to: Run only the selected applications: MilkyWay@Home: no MilkyWay@Home N-Body Simulation: yes You can also set the following if you like: Use CPU, Enforced by version 6.10+: yes Use AMD (ATI) GPU, Enforced by version 6.10+: no Use NVIDIA GPU, Enforced by version 6.10+: no That way you'll only be doing the WUs that GPUs are unable to run. Your GPU will work fine at Collatz, just set your preferences there to use only the GPU. |
22)
Message boards :
Number crunching :
79XX Dont Run
(Message 52495)
Posted 14 Jan 2012 by Beyond Post: None - sorry - its a generalised utility that covers essentials, a throw away vender bribe rofl - so its kind of as is. Cant think of any workaround it might have for that. Unfortunately makes it pretty useless for multiple GPU setups, even setting a fixed fan speed on one card resets the other to automatic and the temp skyrockets. Maybe either I'm missing something obvious or it'll get fixed someday. Afterburner works pretty good for the most part and controls NV cards as well. |
23)
Message boards :
Number crunching :
Help finalizing video card selection
(Message 52491)
Posted 14 Jan 2012 by Beyond Post: My son wanted me to piece together a desktop for him since he can't stand gaming on his laptop. The only thing I am going to have to buy is the video card, because his current choice of games is not very demanding I am trying to keep the card in the $30-50 range. I have it narrowed down to three, NVIDIA GT430 and GT520 as well as AMD HD6450. I understand the one from AMD is not a double precision and therefore will not run MW@H Here's an alternative idea: find an HD 4770 card on eBay or elsewhere (here's one for instance): http://www.ebay.com/itm/NEW-ATI-RADEON-HD-4770-512MB-PCI-E-DUAL-DVI-VIDEO-CARD-/200645441435?pt=PCC_Video_TV_Cards&hash=item2eb766779b MUCH faster than the cards you listed above and works great for MW, Collatz & Moo! |
24)
Message boards :
Number crunching :
79XX Dont Run
(Message 52490)
Posted 14 Jan 2012 by Beyond Post: Try Sapphire Trixx - works on reference AMD cards Zy, thanks for the link. Question: just tried this (4.1.0) on a box with 2 5850 cards (different brands) and it mostly seems to work except that whenever I set the custom fan profile for one card the other resets to automatic. Tried everything I can think of, any ideas? Afterburner 2.10 does not exhibit this problem. |
25)
Message boards :
Number crunching :
79XX Dont Run
(Message 52458)
Posted 13 Jan 2012 by Beyond Post: I ran the cards overnight @1115/300 to see if low memory worked (I run my 5970s @770/175). The invalids came in, whilst still a very small number, they were still there. Sometimes it takes a while for the new invalids to show up. All my machines are getting a 2-3% invalid rate. When I start MW on a new box there are no invalids for a while, then they start dribbling in. They're all of the sderr bug type and I think a lot of them get held in "pending" until they're finally marked as invalid. Even Matt's machines show the same problem and it's one of the 3 reasons I dropped MW for a while (along with the tiny queue and unreliable servers). Since the queue size is up I'm back to give it another try but the false invalids are unique to MW and quite irritating. |
26)
Message boards :
Number crunching :
79XX Dont Run
(Message 52424)
Posted 12 Jan 2012 by Beyond Post: Run Time is 52 secs with GPU set to 1125/1575, the max speed inside the driver, power slider is +5 It's interesting that even at this point the HD 7970 is running the NVidia optimized app at 3.5x faster than the top GTX580 card that I could find in the database. Pretty impressive and the HD 7970 also uses less power. BTW, the page below needs to be corrected and updated: http://milkyway.cs.rpi.edu/milkyway/gpu_list.php The HD 69xx series isn't even listed (unless the top AMD listing of HD5800 is a typo) and no the GTX480 is not faster than the GTX580 according to the WU completion times in "Top Computers". |
27)
Message boards :
Number crunching :
HD7970 on the horizon ..
(Message 52406)
Posted 12 Jan 2012 by Beyond Post: How do they benchmark on any BOINC projects vs the other AMD/ATI cards? Ah, found it. Looks like device 0 is doing ~60 sec/WU & device 1 is at ~96 sec/WU: Found 1 platform(s) Platform 0 information: Name: AMD Accelerated Parallel Processing Version: OpenCL 1.1 AMD-APP (851.4) Vendor: Advanced Micro Devices, Inc. Extensions: cl_khr_icd cl_amd_event_callback cl_amd_offline_devices cl_khr_d3d10_sharing Profile: FULL_PROFILE Didn't find preferred platform Using device 0 on platform 0 Found 3 CL device(s) Trying to show unknown MWDoubleExts 6 Device Tahiti (Advanced Micro Devices, Inc.:0x1002) (CL_DEVICE_TYPE_GPU) Driver version: CAL 1.4.1658 (VM) Version: OpenCL 1.1 AMD-APP (851.4) Compute capability: 0.0 Image support: CL_TRUE Max compute units: 32 Clock frequency: 925 Mhz Global mem size: 3221225472 Local mem size: 32768 Max const buf size: 65536 Double extension: Unhandled MWDoubleExts |
28)
Message boards :
Number crunching :
HD7970 on the horizon ..
(Message 52401)
Posted 12 Jan 2012 by Beyond Post: How do they benchmark on any BOINC projects vs the other AMD/ATI cards? The 7970 looks very promising but are there any projects that it works with now? I notice that MW, Collatz and Moo! are not working so far. |
29)
Message boards :
Number crunching :
GPU reference - Time to complete a WU
(Message 52329)
Posted 8 Jan 2012 by Beyond Post: That are CPU runtimes you are given, not GPU runtimes, that card can't be that fast. I was very surprised aswell with the 4770 performance Carlos, I can confirm at least your results. I have a box with 2 x 4770 cards at 830MHz in non-crossfire mode. It's running 2 MW WUs on each GPU and averages 2:47/WU which figures to ~165K credits/day if running MW exclusively. Right now they're split evenly between MW, Collatz & Moo! For most projects it seems that two 4770 cards = one 5850. |
30)
Message boards :
Number crunching :
Dual GPUs - failed to calculate likelihood
(Message 51753)
Posted 21 Nov 2011 by Beyond Post: Thanks for the responses. The validation errors are unfortunately normal and it's been said they're due to the handing of the stderr.out in BOINC/MW. They did not happen in the old ATI app. Heat has nothing to do with it. |
31)
Message boards :
Number crunching :
MilkyWay CUDA takes up a whole core
(Message 51729)
Posted 19 Nov 2011 by Beyond Post: It's possible to achieve a small benefit by running 2 WUs per GPU because there is a small pause in GPU usage each time a WU nears completion. If 2 WUs are run in tandem, then the second one will use all available GPU during this period. Running more than 2 at a time will yeid no additional benefit and has the potential to cause additional problems. |
32)
Message boards :
Number crunching :
Radeon 6790 - no double precision - impossible!
(Message 51698)
Posted 15 Nov 2011 by Beyond Post: Settled on the 6950.... problem is, the 650 W PSU wont drive the 6950 plus even ONE of the 4770's What 650w are you using? Quality PSUs are essential. I have no problem running X6 machines with 2 5850 cards or 1 5870/5850 & 1 460GTX on either XFX or Antec 650w PSUs. Had one machine running for a long time with a 4770 and a 460 on an Antec 500 EA PSU. It drew 400w according to the killawatt and never had a problem with it. The 4770 (have had 10 of them) uses <100w at 99% utilization, the 5850 about double, the 460GTX a bit more than the 5850. As I understand it the 4770 had the additional capabilities such as DP because it was the test bed for the 40nm process at ATI/AMD. It still offers good performance/watt along with the 58xx and 69xx series. |
33)
Message boards :
Number crunching :
Suddenly most results invalid
(Message 51525)
Posted 27 Oct 2011 by Beyond Post: I wonder if something as simple as flushing the stderr (or adding some extra newlines) before the end of the program could fix this. I bet it would. It looks like most of the ATI projects are using the Stderr output and this is the only one that's giving missing or truncated Stderr results. |
34)
Message boards :
Number crunching :
Suddenly most results invalid
(Message 51524)
Posted 27 Oct 2011 by Beyond Post: Not all machines are giving invalid results. So far have i been having a perfect run. That's another matter that i cant get enough WU's. You probably won't see much of the invalids with this low turnover rate: Number of tasks completed 10 Max tasks per day 10010 Number of tasks today 4 Consecutive valid tasks 1 Average processing rate 2.9387240425833 Average turnaround time 0.31 days Only 1 consecutive valid task... |
35)
Message boards :
Number crunching :
Suddenly most results invalid
(Message 51515)
Posted 26 Oct 2011 by Beyond Post: We get results back from the stderr for quite a while (this wasn't true about a year ago), and I don't many other projects do this. Since it's not working so well, can this be changed? |
36)
Message boards :
Number crunching :
Suddenly most results invalid
(Message 51455)
Posted 21 Oct 2011 by Beyond Post: I really do apologize for being a bit of noob - But I was wondering, how do you know if your WUs get validated or not? I don't know if there is any connection, but I noticed I'm getting considerably less credit than I used to for doing what [I think] is as much work. Look here, notice the choices at the top of the page: http://milkyway.cs.rpi.edu/milkyway/results.php?hostid=329377&offset=0&show_names=0&state=4&appid= |
37)
Message boards :
Number crunching :
Suddenly most results invalid
(Message 51447)
Posted 19 Oct 2011 by Beyond Post: If you look at the top computers you may notice that almost all of them regularly post invalid tasks. This is VERY unusual compared to other projects and may indicate either a problem with 0.82 or a validator problem. The former CP apps did not exhibit this behavior. |
38)
Message boards :
Number crunching :
Frustration!
(Message 51443)
Posted 18 Oct 2011 by Beyond Post: Hi Pete, the cc_config.xml has to be in the base BOINC DATA directory, not the program directory. The DATA directory has the subfolders named 'notices, projects, slots and symbols'. Then restart BOINC. I would also recommend BOINC 6.12.34 (which you're using). Versions later than that so far have some serious problems. You're welcome. MW goes down a lot. I would suggest a backup project such as Collatz which you can set at a project priority of 0. Then it will DL 1 Collatz WU/core at a time until MW becomes available again. |
39)
Message boards :
Number crunching :
Frustration!
(Message 51435)
Posted 17 Oct 2011 by Beyond Post: <cc_config> Hi Pete, the cc_config.xml has to be in the base BOINC DATA directory, not the program directory. The DATA directory has the subfolders named 'notices, projects, slots and symbols'. Then restart BOINC. I would also recommend BOINC 6.12.34 (which you're using). Versions later than that so far have some serious problems. |
40)
Message boards :
News :
Make-Up Credits to Run Wednesday and Thursday
(Message 51257)
Posted 29 Sep 2011 by Beyond Post: The server room that houses Milkyway@home is currently running on a backup AC system. As it is a server room, AC is required even when outside temperatures are not very high. Any news, it's been 7 weeks... |
©2024 Astroinformatics Group