Posts by valterc
log in
1) Message boards : Windows : Near permanent calc error (Message 23764)
Posted 361 days ago by Profile valterc
So, the problem here is when boinc suspends/kills the application (maybe we could ask the developers to wait a little longer before doing this). Anyway, what I noticed is that I get problems only in the boxes that have also some share with the Moo project. Even if the share is low (say 100% collatz, 10% Moo) boinc downloads a lot of Moo workunits and after some time it continues to stop collatz (high priority I guess). Maybe is the way the Moo application leaves the gpu memory after using it, I don't know. No problems at all if the share is with Milkyway.
2) Message boards : Windows : Near permanent calc error (Message 23743)
Posted 368 days ago by Profile valterc
I also did a lot of checking around after getting this kind of errors. The error appears (sometimes) when the task is suspended and restarted (for boinc related or personal reasons). The common (good) behavior (stderr) is the following

...
Resuming at 3093177794165886418944
actual threads 64
Suspending...
Collatz Config Settings:
...

But sometimes there is not the "Suspending..." line and the task errors out afterwards, see this one https://boinc.thesonntags.com/collatz/result.php?resultid=124541366
3) Message boards : Number crunching : Out of Memory errors (Message 23675)
Posted 380 days ago by Profile valterc
It could be two things. You could need to repair/install the C runtime again. You should also run a quick repair on your hard drive (SCAN - Check Disk). Also what is your hardware GPU or CPU details will help direct a better solution.

Yep, just a week ago I started noticing random hard disk failures. I was able to do a sector by sector copy of the faulty hard disk to another one, skipping a lot of bad sectors (the original disk is no more able to pass the DST short test, it's definitely broken). Anyways, windows went up, after the copy, sfc /scannow was able to fix a lot of errors expect one that I had to manually fix. Now things are getting better, no errors until now, so probably your hint was the right one. (Just to be sure I will reinstall the C runtime and some other software too). The host is a I7-4930k with two r9-290X gpus.
4) Message boards : Number crunching : Out of Memory errors (Message 23671)
Posted 382 days ago by Profile valterc
I'm also started to get memory errors (see here: https://boinc.thesonntags.com/collatz/results.php?hostid=145423&offset=0&show_names=0&state=6&appid=). Does anyone know what may be the problem?
5) Message boards : Number crunching : Badges (Message 23482)
Posted 412 days ago by Profile valterc
I made the new ones for the TN-Grid project using inkscape which is an open source graphic editor (powerful but relatively easy to use) which works with the svg format (it's a vectorial format so no scaling problems). It's also really easy to make png images from that.
6) Message boards : News : Collatz Sieve v1.21 released for Windows (Message 21245)
Posted 862 days ago by Profile valterc
@Crystal: I had the same error
<message>
(unknown error) - exit code -1073741819 (0xc0000005)
</message>


Just solved upgrading to the latest Catalyst driver (had 14.4 before)
7) Message boards : Number crunching : Benchmarking the collatz application (Message 20949)
Posted 897 days ago by Profile valterc
Given the large set of the collatz configuration parameters and the even more large set of available 'crunching units' (cpu, gpu, etc.) it would be very nice to have something like a benchmark suite.

I'm thinking about a program that tests the whole set (or a subset) of the available configuration parameters and outputs the most efficient one for the given hardware. This could also probably be implemented using a script calling the micro app.

Any thoughts, ideas?
8) Message boards : Number crunching : Invalid task (Message 20921)
Posted 903 days ago by Profile valterc
I also have some invalid results (stderr seems fine to me) please look at http://boinc.thesonntags.com/collatz/results.php?userid=2089&offset=0&show_names=0&state=5&appid=
9) Message boards : News : New Windows CUDA and OpenCL Versions Released (Message 20791)
Posted 922 days ago by Profile valterc
I started to get validation errors on two different computers, just two, nothing to be worried about, statistically speaking, but I never got them before. I checked the stderr log and it seems fine.

For your reference: http://boinc.thesonntags.com/collatz/results.php?userid=2089&offset=0&show_names=0&state=5&appid=

Regarding performance, I have some statistics on a couple of R280-X, config parameters are:

items_per_kernel=20
kernels_per_reduction=9
threads=8
sleep=1
(lut_size=16,18 for the 6.08, doesn't make any noticeable difference between the two)

6.04: Large 18,730.05 sec. ave, 45,866.05 credit/hour
6.08: Solo 1,336.77 sec. ave, 31,791.19 credit/hour
10) Message boards : News : New Windows CUDA and OpenCL Versions Released (Message 20750)
Posted 924 days ago by Profile valterc
Any hints about the right value for the "lut_size" configuration option for a high end gpu (r290-x r280-x)?
11) Message boards : Number crunching : Computation Errors (Message 20701)
Posted 930 days ago by Profile valterc
The number of workunits erroring out is increasing. My guess is that we reached a point in computation (numbers) which triggered a small application's bug.


We have a winner! There was an error with the lookup tablebut it only appears when certain numbers are checked. The new sieve app uses different code to generate the lookup table. When I use the lookup table generated by the sieve app, it works ok. So, I moved that sieve code into the v6 code base and have updated the CUDA apps to 6.05.

That's great! These kind of bugs are usually very hard to find. Please check the opencl applications as well.
12) Message boards : Number crunching : Computation Errors (Message 20688)
Posted 932 days ago by Profile valterc
I do not see a relationship between GPU type and error rate, all of my AMD cards are producing errors, sometimes. Furthermore some workunits will completely fail because of this high error rate, see this one for example:
http://boinc.thesonntags.com/collatz/workunit.php?wuid=15991927.
13) Message boards : Number crunching : Computation Errors (Message 20660)
Posted 938 days ago by Profile valterc
The number of workunits erroring out is increasing. My guess is that we reached a point in computation (numbers) which triggered a small application's bug.
14) Message boards : Number crunching : Computation Errors (Message 20618)
Posted 944 days ago by Profile valterc
I noticed the same behavior on two different machines (all with R280-X gpus), (unknown error) - exit code -16777217 (0xfeffffff)

In some cases many different computers fail the same way, see this http://boinc.thesonntags.com/collatz/workunit.php?wuid=15790536
15) Message boards : Number crunching : 2x290-x (problems) (Message 19551)
Posted 1332 days ago by Profile valterc
I did not try to run two wus at a time, nor I raised the items_per_kernel to more then 21. That's because
1- I have to fight with temperatures, the card at the top of the other is running a little bit hot... as expected...
2- I also may have some hardware problems... memtest86 gave me errors while using the higher (1866Mhz) XMP profile, using the other profile (1600Mhz) seems okay but I have to do some further testing...
16) Message boards : Number crunching : 2x290-x (problems) (Message 19534)
Posted 1335 days ago by Profile valterc
just 1 wu on each gpu, with the following config:
verbose=1
items_per_kernel=20
kernels_per_reduction=9
threads=8
sleep=1

This is a snapshot of the gpu utilization:
17) Message boards : Number crunching : 2x290-x (problems) (Message 19529)
Posted 1336 days ago by Profile valterc
After some testing I think that I may have some kind of hardware problem. It seems that if I crunch Collatz gpu only everything is going ok, as if I just crunch cpu workunits only, or test the system with prime95. I start to get some windows freezes if I BOTH use cpu and gpu (the last freeze messed up, someway, my network switches...) I have to dig deeper... BTW there are two Sapphire r290-x tri-x, the cpu is a 4930k and the PSU is a Corsair RM1000.
18) Message boards : Number crunching : 2x290-x (problems) (Message 19501)
Posted 1338 days ago by Profile valterc
Hi all,
I just got this new PC (http://boinc.thesonntags.com/collatz/show_host_detail.php?hostid=145423), installed Win7, Catalyst 14.4 and started crunching Collatz workunits. What I got are a lot of computation errors (say around 50%) so I started investigating the reasons, also playing with the config file. What I noticed is that, regardless of the items_per_kernel value, the gpu usage (monitored using both Afterburner and GPU-Z) stays fixed at 100% for a while, then drops to another value (which depends on items_per kernel) and back to 100%. This is a strange behavior, on any other pc the gpu usage, after proper configuration, stays almost fixed around 97-98%. I don't know if this may be related to my error rate...

Any hints or suggestions?
Thanks a lot in advance.
19) Message boards : Number crunching : Where's the top 1% RAC point currently? (Message 18996)
Posted 1384 days ago by Profile valterc
Gold, Silver and Bronze Badge minimums are:
thresholds for user badges: 844312.88696495 176265.09555125 11119.552035708
thresholds for team badges: 6061718.8836996 1734115.3718501 71889.900445213

There was a bug in the php (does anyone ever run this stuff before it gets checked in to version control?) which was keeping it from working.


Tell that to Dr. A so he can fix it on SETI as well.

Could you please explain this bug, and how to fix it? (thanks)
20) Message boards : News : Collatz Back Online with Solo Collatz Only (Message 18325)
Posted 1441 days ago by Profile valterc
with my hd7970, two weeks ago, I had the following (gpu load at ~95%)
verbose=1
items_per_kernel=20
kernels_per_reduction=9
threads=8
sleep=1

Now the gpu load is ~82-83%, I also tried to increase items_per_kernel to 22 but nothing seems to change....


Next 20

Main page · Your account · Message boards


Copyright © 2018 Jon Sonntag; All rights reserved.