Message boards :
Number crunching :
SETI@home v8.12 Windows GPU applications support thread
Message board moderation
Previous · 1 . . . 6 · 7 · 8 · 9 · 10 · 11 · 12 . . . 17 · Next
Author | Message |
---|---|
robertmiles Send message Joined: 16 Jan 12 Posts: 213 Credit: 4,117,756 RAC: 6 ![]() |
I ran the setup procedure again, and switched to the Nvidia OpenCL version. A suggestion for the setup procedure, to help those running it for the first time: Change the first page to show the four GROUPS of methods, with the appropriate hardware types needed for each group, and buttons to select the desired group. Then have four more pages that the buttons on the first page can select, to choose methods within each group, A minor problem with the installation procedure: The copying N items window does not close automatically after reaching 0 seconds remaining, and its Cancel button is EXTREMELY slow to do anything, if it ever does. The X at top right has the same problem. How are users supposed to close this window so they can shut down the setup and start using what was set up? I finally used the Task Manager to end the process for the setup procedure. Will this interfere with using what was set up? Of course, if Raistmer is trying to restrict the testing to only users who can read his mind, to prevent the r3500 version from ever being accepted as stock, he is coming close. |
robertmiles Send message Joined: 16 Jan 12 Posts: 213 Credit: 4,117,756 RAC: 6 ![]() |
The two 64-bit versions of the installation procedure have Beta3 and Beta4 in the hidden parts of their names, and are slightly different sizes. A definite confusion factor for testing the r3500 application. I now have the Beta4 version installed on both of my desktops. I saw something in the Beta4 release notes, but am unable to find those release notes again, about a minimum driver version. Was it 363.*? If so, expect problems under Windows Vista, since at least two BOINC projects are known to have have problems with the two released drivers since 362.00 but before Nvidia stopped supporting Windows Vista for new drivers. If so, I'll have to limit the GPU boards I test, since my Windows 10 desktop does not have an adequate PSU for most of the graphics boards I have available, and will not even boot with those past the GeForce 500 generation. |
![]() ![]() Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121 ![]() ![]() |
OMG... you describe experience with Lunatics installer What relation I have to it at all besides finally you get installed one of my apps (or one of not mine)?? SETI apps news We're not gonna fight them. We're gonna transcend them. |
![]() ![]() Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121 ![]() ![]() |
Minimum driver requiremets for MB OpenCL NV comes from this table: https://drive.google.com/open?id=1gegAsoKPGiUnwidZUx7UFb9SaKXv-uNr-grB8QmVC4k (in short, 350.xx and up) SETI apps news We're not gonna fight them. We're gonna transcend them. |
Stephen "Heretic" ![]() ![]() ![]() ![]() Send message Joined: 20 Sep 12 Posts: 5557 Credit: 192,787,363 RAC: 628 ![]() ![]() |
. . Actually that is one for the Lunatics 0.45 installer thread. It is a little confusing that only part of the download file names are visible forcing a user to either guess or d/l both to get the right one. Sorry Richard but it would be nice to see the whole file name if that is possible. |
Richard Haselgrove ![]() Send message Joined: 4 Jul 99 Posts: 14658 Credit: 200,643,578 RAC: 874 ![]() ![]() |
Please continue discussions about installer usability in the Open Beta test: SoG for NVidia, Lunatics v0.45 - Beta4 update thread. |
![]() ![]() ![]() Send message Joined: 17 Feb 01 Posts: 34290 Credit: 79,922,639 RAC: 80 ![]() ![]() |
Raistmer, check Lunatics. With each crime and every kindness we birth our future. |
![]() ![]() Send message Joined: 16 Jun 01 Posts: 6325 Credit: 106,370,077 RAC: 121 ![]() ![]() |
Raistmer, check Lunatics. Thanks! SETI apps news We're not gonna fight them. We're gonna transcend them. |
robertmiles Send message Joined: 16 Jan 12 Posts: 213 Credit: 4,117,756 RAC: 6 ![]() |
My desktops have finished over 100 r3500 tasks so far. Most of them are waiting for a wingmate to finish so they can be validated. |
![]() ![]() ![]() Send message Joined: 17 Feb 01 Posts: 34290 Credit: 79,922,639 RAC: 80 ![]() ![]() |
My desktops have finished over 100 r3500 tasks so far. Most of them are waiting for a wingmate to finish so they can be validated. Much more interesting is your error rate with guppi tasks. Your card produces false overflows but surprisingly only guppies. Arecibo tasks seems to work fine. With each crime and every kindness we birth our future. |
robertmiles Send message Joined: 16 Jan 12 Posts: 213 Credit: 4,117,756 RAC: 6 ![]() |
My desktops have finished over 100 r3500 tasks so far. Most of them are waiting for a wingmate to finish so they can be validated. Which card? Different ones in each desktop. |
robertmiles Send message Joined: 16 Jan 12 Posts: 213 Credit: 4,117,756 RAC: 6 ![]() |
The stderr files do not show much about whether the task was successful or not. However, for many of my tasks marked inconclusive, a result overflow was shown: SETI@Home Informational message -9 result_overflow NOTE: The number of results detected equals the storage space allocated. I did not find any of my tasks marked as valid that showed this result overflow. Does this mean that r3500 needs to be updated to reserve space for more results? Or if the amount of space varies for the different options, should stderr show how much space was allocated? You may need to check whether the versions for the various hardware platforms make the same choice for which results to send back if there is not room for all of them, and whether the ones sent back are the ones you want most. Also, how well does the validator handle cases where some versions of the application send back all of the results, and some send back only as many as there is room for? Also, how do I tell if a task has produced a false overflow? And how do I tell which tasks are Arecibo? |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13778 Credit: 208,696,464 RAC: 304 ![]() ![]() |
Does this mean that r3500 needs to be updated to reserve space for more results? Nope. In this case, too many results means that the application has detected too many spikes, pulses, triplets etc. This generally occurs if there is noise in the WU (eg from a RADAR), or there is a problem with the processing hardware- overclocked, memory problems, over heating, or just not too healthy. Also, how do I tell if a task has produced a false overflow? It won't be validated when it is checked against other systems results. If it's a noisy WU, then the overflow was correct & you will get credit for the work done. And how do I tell which tasks are Arecibo? Green Bank telescope work generally starts off with blc_ and has _guppie_ not far from the start. Hence they are often referred to as Guppies. All the other types of Work Unit are Arecibo. Grant Darwin NT |
robertmiles Send message Joined: 16 Jan 12 Posts: 213 Credit: 4,117,756 RAC: 6 ![]() |
I got a message twice on one of my desktops, the one with the GTX 560: SETI@home: Notice from BOINC Task postponed: Suspicious spike results, host needs reboot or maintenance 8/19/2016 4:20:55 PM I restarted Windows after the first one; no difference. Note that there's no indication of which task, so I'm unable to check if it finished properly. This GPU appears to be working properly for all other BOINC projects sending it GPU tasks. |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13778 Credit: 208,696,464 RAC: 304 ![]() ![]() |
SETI@home: Notice from BOINC If it's what I think it is, it relates to the SoG application- it was a bit aggressive in it's settings for determining if something was noisy or not. I don't know when the stock application was last updated, but the current SoG application available through the Lunatics installer doesn't have that issue. Grant Darwin NT |
![]() ![]() ![]() Send message Joined: 27 May 99 Posts: 5517 Credit: 528,817,460 RAC: 242 ![]() ![]() |
Isn't that the sanity check? I might be wrong but I thought that is what they determined it was. |
robertmiles Send message Joined: 16 Jan 12 Posts: 213 Credit: 4,117,756 RAC: 6 ![]() |
SETI@home: Notice from BOINC This was AFTER I switched from the stock SoG application to the Lunatics Beta4 r3500 version. I see only one of my tasks marked as invalid, after two wingmates, both running SoG versions (probably stock), gave outputs that passed the validator, but did not agree with my r3500 outputs. No output overflow shown for either wingmate. I'm also running beta tests for another project (DENIS@home) that is not yet always recovering from checkpoints, so I'd prefer to minimize the number of Windows reboots and restarts to limit the need for recovering from checkpoints. |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13778 Credit: 208,696,464 RAC: 304 ![]() ![]() |
Isn't that the sanity check? I might be wrong but I thought that is what they determined it was. Yeah, the sanity check wasn't quite sane. Grant Darwin NT |
Stephen "Heretic" ![]() ![]() ![]() ![]() Send message Joined: 20 Sep 12 Posts: 5557 Credit: 192,787,363 RAC: 628 ![]() ![]() |
Isn't that the sanity check? I might be wrong but I thought that is what they determined it was. . . I am pretty sure that is what Raistmer attributed it to. |
![]() ![]() Send message Joined: 24 Nov 06 Posts: 7489 Credit: 91,093,184 RAC: 0 ![]() |
Isn't that the sanity check? I might be wrong but I thought that is what they determined it was. There are problems with sanity checks in general, and IIRC with this one specifically, because the author has dissapeared, and the rationale isn't documented. "Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions. |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.