Use at your own risk

Message boards : News : Use at your own risk
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 6 · 7 · 8 · 9 · 10 · 11 · 12 . . . 19 · Next

AuthorMessage
Slicker
Project administrator

Send message
Joined: 11 Jun 09
Posts: 56
Credit: 936,681,243
RAC: 760,770
Message 389 - Posted: 17 May 2018, 18:30:30 UTC - in response to Message 385.  

None of the WUs should ever end up as inconclusive because they are either valid or not. The validation is done within the WU. e.g. the CPU WUs doulble check every new "high" using a separate algorithm and if they don't agree, it fails. If they do, it should validate. There shouldn't an "inconclusive". I'm going to turn off the file deleter so that once I figure out what is going on I can re-validate the tasks so you should get credit.


Thanks for this Slicker,

Just had another WU that had validated (like my 1st one), disappear from my account list (just like my 1st one). This time it was on a Linux machine and ran for about 420,000 seconds.
Both were awarded Zero credit.
Have one still there (an inconclusive) and another still running.

I am glad that you are letting me post as I still have Zero RAC, so thanks for that.

Thanks for all your hard work.

Conan


The issue was because BOINC, being stupid as usual, was rejecting the WUs because it didn't like the FLOPS count. LOL. There are no FLOPS in Collatz. Only integer calculations. So, I edited credit.cpp and commented out all the stupid code and re-validated all inconclusive WUs.
ID: 389 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Gero-T

Send message
Joined: 9 Oct 16
Posts: 9
Credit: 57,231,987,399
RAC: 0
Message 391 - Posted: 17 May 2018, 18:52:19 UTC - in response to Message 382.  
Last modified: 17 May 2018, 18:55:38 UTC

why do you use http://..... ?
use [/b]https://... !!
ID: 391 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
[AF>Le_Pommier] Jerome_C2005

Send message
Joined: 27 Jun 12
Posts: 13
Credit: 76,271,140
RAC: 11,316
Message 392 - Posted: 17 May 2018, 20:10:12 UTC

Hi Slicker

just to mention : https://boinc.thesonntags.com/collatz/forum_thread.php?id=25

In case you don't (have time to) check all the forum.

Thanks.
ID: 392 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Woodles

Send message
Joined: 29 Oct 09
Posts: 3
Credit: 10,827,978,045
RAC: 9,575
Message 394 - Posted: 17 May 2018, 21:36:51 UTC
Last modified: 17 May 2018, 21:41:51 UTC

Well after spending 24 hours asking for work and not getting a single workunit I decided to experiment.

I have my primary profile set to a resource share of 0 so that I can disable projects remotely just be changing the location of each host. Normally, the hosts run under location "school" with a resource share of 1000.

I noticed previously that when asking for work, the GPU had an apparent share of 0.00 so I changed the primary setting to have a resource share of 1000, set the default location to "---" (the primary one), removed the project then added it back and now the error messages are listing the share as 0.50 (I'm also running Moo with a resource share of 1000) ... and I'm getting work!

17/05/2018 22:15:31 | collatz | Sending scheduler request: Requested by user.
17/05/2018 22:15:31 | collatz | Requesting new tasks for NVIDIA GPU
17/05/2018 22:15:31 | collatz | [sched_op] CPU work request: 0.00 seconds; 0.00 devices
17/05/2018 22:15:31 | collatz | [sched_op] NVIDIA GPU work request: 86261.29 seconds; 0.50 devices
17/05/2018 22:15:34 | collatz | Scheduler request completed: got 101 new tasks
17/05/2018 22:15:34 | collatz | [sched_op] Server version 711
17/05/2018 22:15:34 | collatz | Project requested delay of 121 seconds

It would appear that for some reason the preferences no longer get updated when using profiles other than the primary one.
ID: 394 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Conan

Send message
Joined: 17 Jun 09
Posts: 17
Credit: 111,996,128
RAC: 0
Message 410 - Posted: 20 May 2018, 9:07:03 UTC - in response to Message 389.  
Last modified: 20 May 2018, 9:08:05 UTC

None of the WUs should ever end up as inconclusive because they are either valid or not. The validation is done within the WU. e.g. the CPU WUs doulble check every new "high" using a separate algorithm and if they don't agree, it fails. If they do, it should validate. There shouldn't an "inconclusive". I'm going to turn off the file deleter so that once I figure out what is going on I can re-validate the tasks so you should get credit.


Thanks for this Slicker,

Just had another WU that had validated (like my 1st one), disappear from my account list (just like my 1st one). This time it was on a Linux machine and ran for about 420,000 seconds.
Both were awarded Zero credit.
Have one still there (an inconclusive) and another still running.

I am glad that you are letting me post as I still have Zero RAC, so thanks for that.

Thanks for all your hard work.

Conan


The issue was because BOINC, being stupid as usual, was rejecting the WUs because it didn't like the FLOPS count. LOL. There are no FLOPS in Collatz. Only integer calculations. So, I edited credit.cpp and commented out all the stupid code and re-validated all inconclusive WUs.


Again thanks for your hard work here Slicker.

I had a Windows work unit that ran for 540,000 or so seconds, then had a Linux WU that ran for 420,000 or so seconds, they both disappeared without credit.
I think that I recieved credit that was for those two work units about 2 days ago, it all went to one computer but that is not a concern.
Since then I have another 2 WUs on different Linux machines that went from inconclusive to pending yesterday, but they also have disappeared without credit.
One ran for 420,700 seconds and the other ran for 786,630 seconds.

So I still have an RAC of 0.00 and I am still thankful you let me post.

Another WU has downloaded on my Windows computer, so the fun starts again.

Conan
ID: 410 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
rcthardcore

Send message
Joined: 15 May 10
Posts: 6
Credit: 54,774,630
RAC: 0
Message 422 - Posted: 23 May 2018, 3:42:06 UTC - in response to Message 215.  

Even though nobody has answered my original question about who the user was that was cheating, not sure what this will do but will give it a try. Thanks for the info and help.
ID: 422 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Padanian

Send message
Joined: 28 May 10
Posts: 8
Credit: 1,407,385,311
RAC: 114,027
Message 424 - Posted: 23 May 2018, 10:22:56 UTC

Strange behaviour for two specific wu:

https://boinc.thesonntags.com/collatz/workunit.php?wuid=912326

https://boinc.thesonntags.com/collatz/workunit.php?wuid=917872

CPU calculation for another user is validated without credit given, while my calculation is considered invalid.
ID: 424 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
[AF>Le_Pommier] Jerome_C2005

Send message
Joined: 27 Jun 12
Posts: 13
Credit: 76,271,140
RAC: 11,316
Message 433 - Posted: 24 May 2018, 20:28:19 UTC

Great news :

- Mac OS ATI OpenCL tasks now on air
- they are working !
- they take 5,5 hours to run on my old Radeon HD 4850 (...)

Not that great news : they use almost one CPU core (0,929), much more than the old app used to...

Do I have to exclude one core in boinc preferences so it will run more efficiently ?
ID: 433 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile mikey
Avatar

Send message
Joined: 11 Aug 09
Posts: 458
Credit: 11,736,565,187
RAC: 6,073,659
Message 436 - Posted: 25 May 2018, 11:09:57 UTC - in response to Message 433.  

Great news :

- Mac OS ATI OpenCL tasks now on air
- they are working !
- they take 5,5 hours to run on my old Radeon HD 4850 (...)

Not that great news : they use almost one CPU core (0,929), much more than the old app used to...

Do I have to exclude one core in boinc preferences so it will run more efficiently ?


Try it and see, exclude one cpu core and see if they run faster, if not go back to not excluding a cpu core.
ID: 436 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Anthony Ayiomamitis

Send message
Joined: 21 Jan 15
Posts: 13
Credit: 7,405,988,191
RAC: 13,590,979
Message 441 - Posted: 25 May 2018, 15:21:57 UTC - in response to Message 433.  
Last modified: 25 May 2018, 15:22:32 UTC

Do I have to exclude one core in boinc preferences so it will run more efficiently ?

I would not rely too much on what BOINC shows in this regard. For example, with my 1080 running Collatz, the CPU contribution is less than one second total time for each unit. Also, I have left one logical core (ie one thread) available for Collatz and which is at best 12.5% of what is available (ie one thread out of eight threads on my quad core i7 6700k). Of course, precisely the opposite with my 1080 and which constantly shows 99-100% utilization.
ID: 441 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
[AF>Le_Pommier] Jerome_C2005

Send message
Joined: 27 Jun 12
Posts: 13
Credit: 76,271,140
RAC: 11,316
Message 442 - Posted: 25 May 2018, 15:34:20 UTC
Last modified: 25 May 2018, 15:34:51 UTC

Thanks for your answers, i'll try the exclusion and see if it changes execution time.

Unfortunately on the Mac (at least on mine with my old GPU) there are no tool (that I know) that inform about the actual use of the GPU while the task is running, like with Windows of Linux. The only test I could do with PrimeGrid OpenCL ATI tasks is having 2 tasks running at the same time and their execution time has been doubled, so useless. But PG tasks require only 0,01% of CPU (officially)...

I'll also do this test with Collatz.
ID: 442 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
JMrkvicka

Send message
Joined: 25 Jul 17
Posts: 13
Credit: 794,955,732
RAC: 1,155,102
Message 448 - Posted: 28 May 2018, 21:19:20 UTC

Any reason to not install the new BOINC client version 7.10.2?
ID: 448 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Plomos

Send message
Joined: 27 Apr 11
Posts: 1
Credit: 20,963,430
RAC: 0
Message 459 - Posted: 2 Jun 2018, 21:08:47 UTC

I briefly installed boinc 7.10.2 and my AMD card was not detected in boinc so I went back to the old version

AMD 6870 radeon
ID: 459 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Alessandro Freda

Send message
Joined: 23 Oct 09
Posts: 4
Credit: 730,800,060
RAC: 0
Message 464 - Posted: 4 Jun 2018, 14:29:15 UTC - in response to Message 205.  

@Alessandro : There is no work available for all platforms / OS / GPUs at the moment, Slicker is working on priorities right now, be patient !

Thanks again for all the hard work Slicker, et ne baisses pas les bras ;)

Edit : I hadn't seen Slicker latest message just above, so mine is even more correct :)

Also be sure to use use "http://" and NOT "https://" when you sign on again.


Ok, after a month I can't get works yet.
I've detach and attach with new client 7.10.2 (https), and try to get works on 2 PC (first with a Radeon 7790, 2nd with intel GPU) that before works without problems. I think the main problem is that the collatz work dir stay empty, it can downdown even the executable, on both PCs, so I suspect the problem could be on my user entry on server.
ID: 464 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Conan

Send message
Joined: 17 Jun 09
Posts: 17
Credit: 111,996,128
RAC: 0
Message 465 - Posted: 4 Jun 2018, 23:21:53 UTC - in response to Message 464.  

@Alessandro : There is no work available for all platforms / OS / GPUs at the moment, Slicker is working on priorities right now, be patient !

Thanks again for all the hard work Slicker, et ne baisses pas les bras ;)

Edit : I hadn't seen Slicker latest message just above, so mine is even more correct :)

Also be sure to use use "http://" and NOT "https://" when you sign on again.


Ok, after a month I can't get works yet.
I've detach and attach with new client 7.10.2 (https), and try to get works on 2 PC (first with a Radeon 7790, 2nd with intel GPU) that before works without problems. I think the main problem is that the collatz work dir stay empty, it can downdown even the executable, on both PCs, so I suspect the problem could be on my user entry on server.


What happens if you add <http_1_0>1</http_1_0> to your cc_config.xml file?

I was given this to try when for the last month I have not been able to download or even contact goofyxgrid@home and get work, then with this it all worked. It changes the standard request http 1.1 to 1.0 I think.

Goes in the options section like this

<cc_config>
<options>
<http_1_0>1</http_1_0>
</options>
</cc_config>

It may alreadt be in your cc_config.xml file, if you have one, but will have a 0 instead of a 1 in the middle.

Conan
ID: 465 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Bill F

Send message
Joined: 27 May 18
Posts: 13
Credit: 20,335,708
RAC: 34,150
Message 467 - Posted: 5 Jun 2018, 0:38:03 UTC
Last modified: 5 Jun 2018, 1:35:01 UTC

For download purposes the Open CL version is 2.0 correct ?

Could that be preventing some devices from down loading ? I have heard that not all OS loads with the vanilla drivers support Open CL or the 2.0 level.

Bill F

P.S. If you could take a peek at Team creation to see if it could be easily turned back on it would be helpful for us new Teams
ID: 467 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Alessandro Freda

Send message
Joined: 23 Oct 09
Posts: 4
Credit: 730,800,060
RAC: 0
Message 468 - Posted: 5 Jun 2018, 16:47:16 UTC - in response to Message 465.  

[quote][quote]Ok, after a month I can't get works yet.
I've detach and attach with new client 7.10.2 (https), and try to get works on 2 PC (first with a Radeon 7790, 2nd with intel GPU) that before works without problems. I think the main problem is that the collatz work dir stay empty, it can downdown even the executable, on both PCs, so I suspect the problem could be on my user entry on server.


What happens if you add <http_1_0>1</http_1_0> to your cc_config.xml file?

I was given this to try when for the last month I have not been able to download or even contact goofyxgrid@home and get work, then with this it all worked. It changes the standard request http 1.1 to 1.0 I think.

Goes in the options section like this

<cc_config>
<options>
<http_1_0>1</http_1_0>
</options>
</cc_config>

It may alreadt be in your cc_config.xml file, if you have one, but will have a 0 instead of a 1 in the middle.

Conan


Yes I had a 0 in my cc_config file, after switch to 1, and restart BOINC got this log:

05/06/2018 18:42:33 | | Fetching configuration file from https://boinc.thesonntags.com/collatz/get_project_config.php
05/06/2018 18:42:47 | collatz | Master file download succeeded
05/06/2018 18:42:52 | collatz | Sending scheduler request: Project initialization.
05/06/2018 18:42:52 | collatz | Requesting new tasks for CPU and AMD/ATI GPU
05/06/2018 18:42:54 | collatz | Scheduler request completed: got 0 new tasks
05/06/2018 18:42:54 | collatz | No tasks sent
05/06/2018 18:42:54 | collatz | Tasks for CPU are available, but your preferences are set to not accept them
05/06/2018 18:42:54 | collatz | Tasks for NVIDIA GPU are available, but your preferences are set to not accept them
05/06/2018 18:42:54 | collatz | Tasks for Intel GPU are available, but your preferences are set to not accept them
05/06/2018 18:42:54 | collatz | New computer location: work

Notice the work location, is for PC with ATI 7790 GPU.
The projects\boinc.thesonntags.com_collatz dir continue to stay empty.
ID: 468 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Michael

Send message
Joined: 31 Oct 10
Posts: 19
Credit: 112,013,048
RAC: 38,937
Message 518 - Posted: 16 Jun 2018, 11:23:44 UTC - in response to Message 1.  

Still not fixed ?? Everytime they show up like this ZERO credits are issued. WHY???
2768965
2405700
44441
10 Jun 2018, 9:34:31 UTC
16 Jun 2018, 5:01:53 UTC
Completed, validation inconclusive
498,386.86
396,562.60
pending
Collatz Sieve v1.40
windows_x86_64
2768966
2405747
44441
10 Jun 2018, 9:34:31 UTC
16 Jun 2018, 2:30:08 UTC
Completed, validation inconclusive
489,283.83
387,497.40
pending
Collatz Sieve v1.40
windows_x86_64
2768973
2461808
44441
10 Jun 2018, 9:34:31 UTC
16 Jun 2018, 3:36:18 UTC
Completed, validation inconclusive
493,272.95
391,351.30
pending
Collatz Sieve v1.40
windows_x86_64
ID: 518 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Michael

Send message
Joined: 31 Oct 10
Posts: 19
Credit: 112,013,048
RAC: 38,937
Message 519 - Posted: 16 Jun 2018, 21:10:33 UTC - in response to Message 1.  

Yea but WTF constantly getting these

2768965
2405700
44441
10 Jun 2018, 9:34:31 UTC
16 Jun 2018, 5:01:53 UTC
Completed and validated
498,386.86
396,562.60
0.00
Collatz Sieve v1.40
windows_x86_64
2768966
2405747
44441
10 Jun 2018, 9:34:31 UTC
16 Jun 2018, 2:30:08 UTC
Completed and validated
489,283.83
387,497.40
0.00
Collatz Sieve v1.40
windows_x86_64
2768973
2461808
44441
10 Jun 2018, 9:34:31 UTC
16 Jun 2018, 3:36:18 UTC
Completed and validated
493,272.95
391,351.30
0.00
Collatz Sieve v1.40
windows_x86_64

So I guess you get free computer use.
ID: 519 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Michael

Send message
Joined: 31 Oct 10
Posts: 19
Credit: 112,013,048
RAC: 38,937
Message 522 - Posted: 17 Jun 2018, 9:43:21 UTC - in response to Message 1.  

And yet another

2851530
2539659
44441
11 Jun 2018, 16:53:58 UTC
17 Jun 2018, 9:39:53 UTC
Completed, validation inconclusive
492,179.40
390,542.40
pending
Collatz Sieve v1.40
windows_x86_64
ID: 522 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Previous · 1 . . . 6 · 7 · 8 · 9 · 10 · 11 · 12 . . . 19 · Next

Message boards : News : Use at your own risk


©2019 Jon Sonntag; All rights reserved