Posts by Speedy51
log in
21) Message boards : News : upcoming S66 batch change (Message 9288)
Posted 19 Nov 2023 by Speedy51

Preparing for the new batch there will be some changes:

- grace_period reduced from 12h to 6h
- deadline would be 6h (in total you have a deadline of 12h)

Message 9260 - Posted: 15 Nov 2023, 22:38:27 UTC - in response to Message 9259.
The server still has running the grace_period so you have 12h + 12h

This is a shame, hopefully not much work will have to be reissued because of tasks missing deadlines, for people that need to turn their machines off overnight
22) Message boards : News : upcoming S66 batch change (Message 9284)
Posted 19 Nov 2023 by Speedy51
When work is available I will happily help process it. Thanks for all of the work you put in behind-the-scenes so work can get processed
23) Message boards : News : upcoming S66 batch change (Message 9281)
Posted 18 Nov 2023 by Speedy51
Asking out of curiosity, do you have any idea when more work will be available?


After all results are back. All primes which are found need to be removed from the sievefile. Already decreased deadline from the current batch today. I will increase work for the new batch to get it done in time. The current speed is awful.

Thanks for the update. I wondered if Auck all results needed to be returned before new one can go out. Hope the remaining result return quickly for you
24) Message boards : News : upcoming S66 batch change (Message 9279)
Posted 18 Nov 2023 by Speedy51
Asking out of curiosity, do you have any idea when more work will be available?
25) Message boards : Number crunching : Progress ? (Message 9264)
Posted 16 Nov 2023 by Speedy51
Is there a page we can go to to see how Sierpinski / Riesel Base - short is progressing?
26) Message boards : News : upcoming S66 batch change (Message 9261)
Posted 16 Nov 2023 by Speedy51
Yes I agree deadlines can certainly be shorter. Could I suggest instead of a 0.5 (12 hour) moving to a .75 (18 hour) deadline? Reason being not everybody runs the computer is 24 hours a day.

I feel potentially with a 0.5 day deadline you may get some hosts missing the deadline


The server still has running the grace_period so you have 12h + 12h

Fantastic nice touch. When processing I will do my best to get the work returned before deadline.
27) Message boards : News : upcoming S66 batch change (Message 9259)
Posted 15 Nov 2023 by Speedy51
Yes I agree deadlines can certainly be shorter. Could I suggest instead of a 0.5 (12 hour) moving to a .75 (18 hour) deadline? Reason being not everybody runs the computer is 24 hours a day.

I feel potentially with a 0.5 day deadline you may get some hosts missing the deadline
28) Message boards : Number crunching : TF credit change (Message 9244)
Posted 8 Nov 2023 by Speedy51
Has been some time since I have given an update on what credit as being given for tasks. We are currently processing tasks in the following range. TF_74-75_366-376M On a high-end RTX 3080 tasks are running between 2 minutes and 53 seconds and 3 minutes and 4 seconds and being given 5,700 credits.

Looks like in the last day or so we had switched to a higher range number? "74-75_262-405M" On my high-end card these are running for 3 minutes 52 seconds and we are receiving 5400 credits.

In the post I have quoted I was receiving 300 more credits for roughly 60 seconds less runtime. Not criticising just pointing out
29) Message boards : Number crunching : TF credit change (Message 9193)
Posted 26 Oct 2023 by Speedy51
Has been some time since I have given an update on what credit as being given for tasks. We are currently processing tasks in the following range. TF_74-75_366-376M On a high-end RTX 3080 tasks are running between 2 minutes and 53 seconds and 3 minutes and 4 seconds and being given 5,700 credits.
30) Message boards : News : Formula BOINC 2023 24.08.2023 23:00 (UTC) - 27.08.2023 22:59 (UTC) (Message 9014)
Posted 31 Aug 2023 by Speedy51
Thanks for details. Hope the user gets their host sorted
31) Message boards : News : Formula BOINC 2023 24.08.2023 23:00 (UTC) - 27.08.2023 22:59 (UTC) (Message 9012)
Posted 31 Aug 2023 by Speedy51
Out of curiosity do you know/have an idea how many "TF" tasks there are left to recycle? Could also be possible that I just had a patch of 80 units that had missed their "start deadline"


I cant see any missed deadline WUs. 45k WUs are left but new WUs will be added every time.

Thanks for checking, appreciate it.

I just received another 80 tasks and they all appear to have _1 at the end. Seems I have got tasks that the 1st copy got "error while computing" I will clean these up :-)
[url]https://srbase.my-firewall.org/sr5/result.php?resultid=111025684
[/url] here is a task I was referring to in my previous message
32) Message boards : News : Formula BOINC 2023 24.08.2023 23:00 (UTC) - 27.08.2023 22:59 (UTC) (Message 9010)
Posted 31 Aug 2023 by Speedy51
Out of curiosity do you know/have an idea how many "TF" tasks there are left to recycle? Could also be possible that I just had a patch of 80 units that had missed their "start deadline"
33) Message boards : Number crunching : TF credit change (Message 8832)
Posted 1 May 2023 by Speedy51
Yes I understand in regards to purging
I don't recall seeing any tasks from 196 – 202M maybe this took place overnight while my computer was off. Are you able to let me know what the end work unit number is in the batch like "62,361" I am aware the Workunit numbers do not have a, in then
34) Message boards : Number crunching : TF credit change (Message 8830)
Posted 1 May 2023 by Speedy51
I have just completed TF_74-75_202-207M_wu_62361_0 Runtime of 301 seconds i think this is the same range as I completed this morning 202 – 207M and not sure how you mean "you refer the runtime from the last batch" unfortunately all of my old results have been deleted,
35) Message boards : Number crunching : TF credit change (Message 8827)
Posted 1 May 2023 by Speedy51
Interesting you mention about the credits because I was thinking it was a "large" drop in credit for around 15 seconds of runtime
36) Message boards : Number crunching : TF credit change (Message 8825)
Posted 1 May 2023 by Speedy51
Tasks with the following 74-75_202-207M in their names are giving 10,000 credits. For me they are running for just over 5 minutes however I have a high-end card RTX 3080.
Unfortunately I don't have any tasks so I am not able to get the name credit drop seems to be 2,500 tasks were giving 12,500 credits from memory runtime was around 5 minutes 15 seconds


Looks different, my RX5500XT test got around 57min while we are now in the next range. Always testing the last number in batch (207M instead of 202M)

Thanks for the time comparison. May I ask what you mean by "looks different"
37) Message boards : Number crunching : TF credit change (Message 8822)
Posted 30 Apr 2023 by Speedy51
Tasks with the following 74-75_202-207M in their names are giving 10,000 credits. For me they are running for just over 5 minutes however I have a high-end card RTX 3080.
Unfortunately I don't have any tasks so I am not able to get the name credit drop seems to be 2,500 tasks were giving 12,500 credits from memory runtime was around 5 minutes 15 seconds
38) Message boards : News : Formula BOINC 2023 27.04.2023 20:00 (UTC) - 30.04.2023 19:59 (UTC) (Message 8821)
Posted 30 Apr 2023 by Speedy51
As I write project currently is at 1.221 P flops https://grafana.kiska.pw/d/boinc/boinc?orgId=1&var-project=srBase
39) Message boards : Number crunching : Trial Factoring Goal(s) (Message 8765)
Posted 27 Mar 2023 by Speedy51
OK, as the reason for the credit drop because we have finished/completed another bit level/range? If no why credits dropped from 13,500 to 12,500?
40) Message boards : Number crunching : Trial Factoring Goal(s) (Message 8763)
Posted 27 Mar 2023 by Speedy51
Have we reached the next bitlvl? I see credit is now 12,500 a drop of 1000, for what appears to be about a 8 second drop in run time. We got to the 1st credit drop quicker than I expected. Fantastic!


Previous 20 · Next 20

Main page · Your account · Message boards


Copyright © 2014-2024 BOINC Confederation / rebirther