Posts by Slicker

1) Message boards : Number crunching : Optimizing the apps (Message 2730)
Posted 26 May 2020 by Slicker
Post:
I have no idea what that means and haven't tried using a one instead.


The 980 config you posted says "verbose=0".

Was it a mistake or is there a reason to set verbose to 0?

I'd guess mistake, so maybe it should be corrected to 1?


It is a copy from the old website before Collatz switched to the current one, so no it wasn't a typo or mistake on my part.
I run 2 GTX980's and they are doing just fine with the zero.

My point was I have no clue what "verbose" even means in this context so a 1 or a zero wasn't tested by me any more than the settings for my GTX1080Ti gpu's were tested by me. Jon posted the original post and then people said changing this or that setting to this or that was faster so I copied the new settings and pasted them here.


Verbose just logs more information which helps with debugging if the apps aren't behaving as expected.
2) Message boards : Science : What algorithm does he use? Here is a fast algorithm... (Message 2609)
Posted 5 May 2020 by Slicker
Post:
Thanks for the heads up. The links have been corrected. Also, in addition to the lookup table, the sieve applications also use a variable sized sieve to further reduce the numbers that require testing making use of the modular restrictions listed on the wiki page.
3) Message boards : Science : Open source (Message 2603)
Posted 4 May 2020 by Slicker
Post:
I've never released the source code for this project and do not intend to ever do so. There were people who were cheating which required encrypting the data returned to the server. Releasing the source code would make it easier for people to cheat again.
4) Message boards : Number crunching : Help: Adjust CPU usage for GPU (Message 2445)
Posted 29 Mar 2020 by Slicker
Post:
Yes, the app_config is the place to make the changes.
The name field contains the app name.Note that the app name doesn't include the platform or plan class. At present, collatz_sieve is the only app name. Just change the cpu_usage to o.30. If you have already optimized the application for you GPU to get it close to 100%, then set the gpu_usage to 1.0.
<app_config>
<app>
<name>collatz_sieve</name>
<gpu_versions>
<gpu_usage>1.0</gpu_usage>
<cpu_usage>0.3</cpu_usage>
</gpu_versions>
</app>
</app_config>
5) Message boards : News : Choose Wisely (Message 2443)
Posted 26 Mar 2020 by Slicker
Post:
I would like to urge participants to switch to Rosetta@Home or Folding@Home (non-BOINC application but also distributed computing) to aid with Coronavirus research until a vaccine has been developed and tested. While it would be nice to solve a decades old mathematical conjecture, it would be better to do research to save lives. Please consider it. Please note that Folding@Home has both CPU and GPU applications.
6) Questions and Answers : Web site : How far has Collatz conjecture been computationally verified? (Message 1946)
Posted 25 Sep 2019 by Slicker
Post:
The WU in progress with the highest number is 6,415,879,838,212,923,850,752. However, that does not mean all numbers up to that have been checked. Each WU checks approximately 53 trillion numbers, but they aren't necessarily completed in order. When people abort them abandon the project, or they timeout or error out, they get resent to another computer. That sometimes happens half a dozen times before a WU gets completed. With 271K WUs in progress, it is impossible to give an exact answer.
7) Message boards : Number crunching : Three new problems this week. (Message 1833)
Posted 24 Jun 2019 by Slicker
Post:
I was up in Wisconsin for a wedding this past weekend and was told to not bring any electronic devices (other than cameras). So, while you were having troubles with Collatz, my King of Avalon shields ended during a kill event and I lost all my troops which will now take months to heal and/or create new. So, at least we all had a bad weekend (even worse if you include the post wedding hangover, LOL)

Anyway.... the file_upload_handler daemon was set to log the maximum info in order to try and track possible missing file (corrupt file?) issues and mismatched database vs. file system. Unfortunately, the volunteers process so many WUs that it filled the drive. I reset the daemon to the minimum log settings, cleared the log files, and restarted all the daemons. Let me know if you still have trouble uploading work. It appears to have fixed the issue but I hate to jump to conclusions.
8) Questions and Answers : Unix/Linux : GPU in Linux Mint (Message 1803)
Posted 12 Jun 2019 by Slicker
Post:
In general, the more power a GPU requires, the faster it will crunch. If you are a real glutton for punishment, an nVidia 8400GS, which you might be able to find for $10-15, will work but very, very, very slowly.
9) Message boards : Number crunching : New record in the "Recent High Steps" category [2766 steps by ian] (Message 1802)
Posted 12 Jun 2019 by Slicker
Post:
I added a top 25 of all time high steps to the daily high steps page. So unlike the results to date page, this one doesn't care if someone else found higher steps, it only cares that it is in the top 25 regardless of the date found.
10) Message boards : Number crunching : New record in the "Recent High Steps" category [2766 steps by ian] (Message 1800)
Posted 12 Jun 2019 by Slicker
Post:
While it beat the previous high of 2760, there have actually been several others that were even higher with the best ever result of 2830 steps. I've just been lax in getting the highest_steps page updated. I triple check each new record to make sure the result is correct.

2017-08-17 09:55:48
3,572,582,862,843,521,162,235
Steps: 2830


There have been 5 others recently that beat the 2760 mark, including the one from ian, but they all occurred after the one above so they aren't listed in the highest_steps page.

2017-11-13 12:10:42 | 3768165902922815297383 | 2768
2017-12-03 20:18:20 | 3815057969257217178619 | 2812
2018-02-27 19:33:57 | 4073990126350358383615 | 2794
2018-03-25 12:56:28 | 4157472480784741832731 | 2825
2019-04-23 22:19:09 | 5652248854384222946075 | 2766
11) Message boards : Number crunching : Any update on the science? (Message 1459)
Posted 23 Jan 2019 by Slicker
Post:
Now that hunting season is over, I am taking a few days off to go somewhere warm, but I'll take a look when I get back to see why the daily results and overall results/highest steps pages aren't working.
12) Message boards : Number crunching : New MacVidCards' NVIDIA GTX-1070 8GB Card. (Message 1380)
Posted 7 Jan 2019 by Slicker
Post:
In a nutshell, BOINC currently uses a 32-bit integer to store to total memory regardless of whether compiling for a 32 or a 64 bit platform and the max value of a 32-bit integer is 4GB. The database may also limit the values to 4GB although I can't say I've checked to see if that's the case, but as memory serves, the GPU data is stored as XML in a text field (a.k.a. domain) which was meant for other things and later re-purposed, along with several other data points (in other words, not the greatest database design). The XML is then parsed to output the data for internal or web use. If the database design used separate fields, all it would take is changing to a 64 bit integer for both the data and the coproc class. Since it isn't, there's more work involved than if it was a single variable store in a single field. But, as far as I know, the only place where it would cause problems is when a project requires more than 4GB of GPU RAM to run an app and I don't think that any projects have that requirement.
13) Message boards : News : Use at your own risk (Message 804)
Posted 18 Jul 2018 by Slicker
Post:
The validate inconclusive result issue should be fixed now. For whatever reason, some windows apps are not reporting the elapsed time correctly and one of the validator rules is to make sure that the WU runtime is reasonable. For example, if a WU finishes in only 3 seconds, there's no way it is valid. So rather than get the elapsed time from the WU, it now counts on the run time reported by the BOINC client for the non-GPU windows apps. There were 87 such results that have now been revalidated and granted credit.
14) Message boards : Number crunching : Validation Pending? (Message 803)
Posted 18 Jul 2018 by Slicker
Post:
The validate inconclusive result issue should be fixed now. For whatever reason, some windows apps are not reporting the elapsed time correctly and one of the validator rules is to make sure that the WU runtime is reasonable. For example, if a WU finishes in only 3 seconds, there's no way it is valid. So rather than get the elapsed time from the WU, it now counts on the run time reported by the BOINC client for the non-GPU windows apps. There were 87 such results that have now been revalidated and granted credit.
15) Message boards : News : Use at your own risk (Message 735)
Posted 12 Jul 2018 by Slicker
Post:
4 more validator versions later and results are now being granted credit.

Now to figure out how to resubmit all the WUs to be re-validated so they can get credit. MySQL runs out of memory if I submit all but submitting 1k and a time could take me weeks and in the mean time, others may turn in a valid result negating the original even thought it was valid.
16) Message boards : News : Use at your own risk (Message 723)
Posted 11 Jul 2018 by Slicker
Post:
I'm still trying to figure out why the WUs are getting marked as invalid when they are not.... I'm tempted to go back to the old validator source code and add in the new encryption as code since that worked OK. In the mean time, feel free to suspend or switch to other projects until I get this figured out.
17) Message boards : News : Use at your own risk (Message 600)
Posted 5 Jul 2018 by Slicker
Post:
The issue is that someone is trying to hack the output and their crap is causing the validator to crash. I'm trying to add code to filter their invalid results so it will stop crashing on every bad WU returned.
18) Message boards : News : Use at your own risk (Message 560)
Posted 26 Jun 2018 by Slicker
Post:
There's no join in Mudville!

The previous fix for the CPU credits wasn't working. Or rather, it didn't fix (and by fix I mean remove the stupid BOINC code that assumes all projects use floating point arithmetic). So, I commented out several hundred more lines of creditnew madness in both the validator.cpp and credit.cpp BOINC source code and then recompiled the server daemons.

I also manually changed the credits for the 204 WUs that were valid but not granted any credit.

If you run into a a problem, please provide the host id and result id as it makes it a lot easier to track down the problems. (Thanks, Conan for providing that info which led to this latest fix.)
19) Message boards : News : Use at your own risk (Message 559)
Posted 26 Jun 2018 by Slicker
Post:
> CUDA: NVIDIA GPU 0: GeForce GTX 1050 (driver version 390.48, CUDA version 9.1, compute capability 6.1, 1997MB, 1698MB available, 1862 GFLOPS peak)

Collatz only has OpenCL apps, not CUDA and from the description above, there's no OpenCL installed. Check out https://wiki.tiker.net/OpenCLHowTo
20) Message boards : Number crunching : Constantly no credits on Sieve v1.40 work units (Message 544)
Posted 24 Jun 2018 by Slicker
Post:
Fixed? Not fixed?


Next 20


©2020 Jon Sonntag; All rights reserved