Posts by TJM

61) Message boards : News : Enigma Cuda 1.07 (Message 5801) Posted 8 Jul 2017 by TJM Post: CUDA app version changed to 1.07 for both Windows x86 and x86_64. I have removed the plan_classes "cuda" and "cuda23" because the app required compute capability 2.1+ cards & I haven't seen even a single task completed with success on "cuda23" plan_class. The new version was recompiled to allow compute capability 2.0 cards: GeForce GTX 590, GeForce GTX 580, GeForce GTX 570, GeForce GTX 480, GeForce GTX 470, GeForce GTX 465, GeForce GTX 480M (source: https://en.wikipedia.org/wiki/CUDA) so in theory anything from Fermi up should work. The CUDA app requires Visual Studio 2015 c++ redistributable package installed to run.
62) Message boards : Number crunching : A GPU version of Enigma software (Message 5780) Posted 7 Jul 2017 by TJM Post: Yep, but since it is not that critical, I'll do it later today or tomorrow morning. At this moment I added single walk of ALQFI (M3/Heer settings, ukw B) to check if it'll pass through the server. If it does I'll switch it to full batch. I have had some of these complete. Pretty quick. I used short workunits in two batches for testing purposes, I'll avoid that in the future because it slows down processing a bit and also causes unnecessary stress on the server.
63) Message boards : Number crunching : Enigma@Home server log (Message 5766) Posted 6 Jul 2017 by TJM Post: I have cleaned the log (started a fresh one) because due to changes in the server code it posted way too much info. I'll make the old log visible again after cleaning up. The new log will show only new top scores, I'll make all the unique top results available for review via separate script.
64) Message boards : Number crunching : A GPU version of Enigma software (Message 5531) Posted 6 Jul 2017 by TJM Post: Do you have Visual Studio 2015 c++ redistributable package installed ?
65) Message boards : Number crunching : A GPU version of Enigma software (Message 5379) Posted 6 Jul 2017 by TJM Post: You have to allow beta work in project preferences on the website.
66) Message boards : Number crunching : A GPU version of Enigma software (Message 5376) Posted 5 Jul 2017 by TJM Post: Yep, but since it is not that critical, I'll do it later today or tomorrow morning. At this moment I added single walk of ALQFI (M3/Heer settings, ukw B) to check if it'll pass through the server. If it does I'll switch it to full batch.
67) Message boards : Number crunching : g3-test Tasks All Fail (Message 5372) Posted 5 Jul 2017 by TJM Post: The problem is fixed.
68) Message boards : Number crunching : A GPU version of Enigma software (Message 5370) Posted 4 Jul 2017 by TJM Post: It was restarted at some point. The app saves it's state on exit but since it does not report it to the BOINC client, it thinks that it was a fresh start and reports only the fraction of runtime, that's required to finish the task.
69) Message boards : Number crunching : g3-test Tasks All Fail (Message 5367) Posted 4 Jul 2017 by TJM Post: It is a known bug and will be fixed in next server upgrade. Just ignore them for now, these tasks will download and fail on start without wasting any resources.
70) Message boards : News : GPU app (Message 5366) Posted 4 Jul 2017 by TJM Post: I'll ask the author to add support for "--device #" passed via command line. Also, I have removed x86 and x86_64 plan classes "cuda" as the app requires compute capability 2.0+ anyway. Probably after next app upgrade I'll leave just the "cuda_fermi" plan class.
71) Message boards : Number crunching : A GPU version of Enigma software (Message 5360) Posted 3 Jul 2017 by TJM Post: I have 4 rigs 2 t3500 Xeon machines. & 2 dual cores, 1 amd & 1 intel, that will not run the app New install of win10 on the t3500 + VC2015 still no go. Pulled intel 2 core and install ( i5 4 core and it worked.) How do I make the Xeon work, Thanks. Run the executable from command line and check for any error messages. You can use the executable from project folder or just download this https://s3.eu-central-1.amazonaws.com/enigma4/test_runs/2.0_test.zip and run one of the bat files.
72) Message boards : Number crunching : A GPU version of Enigma software (Message 5359) Posted 3 Jul 2017 by TJM Post: Yesterday I have released test M3 batch which has failed. The app expects lower case UKW & left ring when running M4 task from resume (upper case won't work), but will fail with lower case UKW when running M3 task. I can easily work around that on the server side, since the input files are autogenerated it's just a minor tweak to use mixed case for M4 and all upper case for M3, but future app versions must be able to read resume files like this - once the batch is running, the input files have their checksums stored in the database and any change will result in download errors.
73) Message boards : Number crunching : A GPU version of Enigma software (Message 5351) Posted 2 Jul 2017 by TJM Post: http://www.enigmaathome.net/gpu_os.php?sort=1#table1
74) Message boards : Number crunching : A GPU version of Enigma software (Message 5350) Posted 2 Jul 2017 by TJM Post: I'd still like to give the 580's a try, but if it turned out as bad as you say then of course i'd put them to work at a different project where they still have value. I do have better cards so either way I wont miss out. I have prepared test suite for you to try out: https://s3.eu-central-1.amazonaws.com/enigma4/test_runs/2.0_test.zip Inside you'll find 3 bat files to run enigma-cuda with different settings, included executable is 64bit, replace it with 32bit version if you need. Check if it works and tell us what are the runtimes - the app reports runtime once it's done.
75) Message boards : Number crunching : A GPU version of Enigma software (Message 5347) Posted 1 Jul 2017 by TJM Post: IMO limiting the cards at 2.1 does not make sense, aren't all the cards listed as 2.0 actually much faster than majority of the compute capability 2.1 cards ? https://en.wikipedia.org/wiki/CUDA#GPUs_supported https://en.wikipedia.org/wiki/List_of_Nvidia_graphics_processing_units
76) Message boards : Number crunching : A GPU version of Enigma software (Message 5345) Posted 1 Jul 2017 by TJM Post: Crunching since 3 days on a not aced 1070 and i see mistakes. 18 of 140 WUs are exiting with error 195 Is this normal? They are only lasting 2 seconds so that's not a big deal. Must be a bug of some sort, but that's either a bug in the app or on the server side. 195 means that the app could not start so perhaps one of the workunit files is missing or corrupted. If it fails on one host, it will also fail for everyone running a resend of affected WU. Just ignore this for now, the server saves a lot of debug info for workunits which fail, so I should be able to find the reason and possibly fix the problem soon.
77) Message boards : Number crunching : A GPU version of Enigma software (Message 5333) Posted 28 Jun 2017 by TJM Post: What on this this God forsaken Earth does it take to get GPU work? Every time the server shows work available I hammer the thing and have yet to get even one task!!!! Win XP, GT 730, works fine on all other GPU projects. Prefs are correct (Nvidia GPU and even test apps checked). This is just a second day of testing so do not expect everything to work. There is a bug in the Win32 template files which will be fixed later today.
78) Message boards : Number crunching : A GPU version of Enigma software (Message 5313) Posted 27 Jun 2017 by TJM Post: It was a bug in the file generator, it added space at the end of the line which then caused md5 errors. I think it's fixed now.
79) Message boards : Number crunching : A GPU version of Enigma software (Message 5311) Posted 27 Jun 2017 by TJM Post: I'm moving the input file generator from main server to download mirror. Once it's done and confirmed working, there will be lots of WUs.
80) Message boards : Number crunching : A GPU version of Enigma software (Message 5308) Posted 26 Jun 2017 by TJM Post: There are some of the test tasks available every now and then. Large batch was scheduled for today but since I had some issues with the server, it has to wait until tomorrow morning (which is just a couple of hours from now, as it's already late here).

Previous 20 · Next 20