ecm_op WU's very low credit
ecm_op WU's very low credit
The ecm_op (not the P2 ones) WU's take a long time for very little credit, for example:
https://www.rechenkraft.net/yoyo//worku ... d=52225912
The credit is 220 for almost 13 hours crunching. Compared to other types of WU on this computer, that's very low and seems wrong to me. Can the reward for ecm_op WU's be re-evaluated?
https://www.rechenkraft.net/yoyo//worku ... d=52225912
The credit is 220 for almost 13 hours crunching. Compared to other types of WU on this computer, that's very low and seems wrong to me. Can the reward for ecm_op WU's be re-evaluated?
- Michael H.W. Weber
- Vereinsvorstand
- Beiträge: 22435
- Registriert: 07.01.2002 01:00
- Wohnort: Marpurk
- Kontaktdaten:
Re: ecm_op WU's very low credit
May I ask what other types of WUs you are talking about, exactly?
Michael.
Michael.
Fördern, kooperieren und konstruieren statt fordern, konkurrieren und konsumieren.
http://signature.statseb.fr I: Kaputte Seite A
http://signature.statseb.fr II: Kaputte Seite B
http://signature.statseb.fr I: Kaputte Seite A
http://signature.statseb.fr II: Kaputte Seite B
Re: ecm_op WU's very low credit
Siever:
https://www.rechenkraft.net/yoyo//worku ... d=52182837
https://www.rechenkraft.net/yoyo//worku ... d=52182602
ecm_op P2:
https://www.rechenkraft.net/yoyo//worku ... d=52241846
https://www.rechenkraft.net/yoyo//worku ... d=52241313
ecm_ru:
https://www.rechenkraft.net/yoyo//worku ... d=52201777
https://www.rechenkraft.net/yoyo//worku ... d=52201763
ecm_hc:
https://www.rechenkraft.net/yoyo//worku ... d=52188120
https://www.rechenkraft.net/yoyo//worku ... d=52187556
to name a few.
As far as I can tell, the P1 ecm_op is the worst rewarding WU by a big margin.
https://www.rechenkraft.net/yoyo//worku ... d=52182837
https://www.rechenkraft.net/yoyo//worku ... d=52182602
ecm_op P2:
https://www.rechenkraft.net/yoyo//worku ... d=52241846
https://www.rechenkraft.net/yoyo//worku ... d=52241313
ecm_ru:
https://www.rechenkraft.net/yoyo//worku ... d=52201777
https://www.rechenkraft.net/yoyo//worku ... d=52201763
ecm_hc:
https://www.rechenkraft.net/yoyo//worku ... d=52188120
https://www.rechenkraft.net/yoyo//worku ... d=52187556
to name a few.
As far as I can tell, the P1 ecm_op is the worst rewarding WU by a big margin.
Re: ecm_op WU's very low credit
I forgot to mention: I hit a boinc-client bug when creating an app_config.xml, and downloaded weeks of work. That's why you'll see a lot of aborted tasks when checking my computers.
According to the release notes for version 7.16, this bug is fixed:
https://boinc.berkeley.edu/wiki/Release ... BOINC_7.16
but apparently it isn't, as I am running 7.16.6.
According to the release notes for version 7.16, this bug is fixed:
https://boinc.berkeley.edu/wiki/Release ... BOINC_7.16
but apparently it isn't, as I am running 7.16.6.
Re: ecm_op WU's very low credit
How many credits would be appropriate?
Re: ecm_op WU's very low credit
Also many sieve tasks generate rather low credits - those tasks granted me 8.5 credits for 1 hour CPU time:
https://www.rechenkraft.net/yoyo/result ... d=72908730
https://www.rechenkraft.net/yoyo/result ... d=72963202
Currently I have 12.000 credits for sieve tasks. Continuing like this would mean to spend (100.000-12.000) / 8.5 = 10.300 CPU hours = 1.2 CPU years to obtain the silver badge
I am crunching for helping progress mathematics and not primarily for badges, however some sort of "virtual reward" after some months of crunching would be nice to see.
https://www.rechenkraft.net/yoyo/result ... d=72908730
https://www.rechenkraft.net/yoyo/result ... d=72963202
Currently I have 12.000 credits for sieve tasks. Continuing like this would mean to spend (100.000-12.000) / 8.5 = 10.300 CPU hours = 1.2 CPU years to obtain the silver badge
I am crunching for helping progress mathematics and not primarily for badges, however some sort of "virtual reward" after some months of crunching would be nice to see.
Re: ecm_op WU's very low credit
Switch to Linux and just do siever work units.
They get more than double the credit the rest of us get.
Or switch to the latest MacOS and get nothing at all...
Interesting that you mentioned the app_config causing a problem. Now I've removed mine from the two problematic hosts they are downloading the expected amount of work.
Is this BOINC bug documented? Both hosts (one Mac, one Windows) were using 7.16.11 -- so it can't have been fixed.
They get more than double the credit the rest of us get.
Or switch to the latest MacOS and get nothing at all...
Interesting that you mentioned the app_config causing a problem. Now I've removed mine from the two problematic hosts they are downloading the expected amount of work.
Is this BOINC bug documented? Both hosts (one Mac, one Windows) were using 7.16.11 -- so it can't have been fixed.