Message boards : Number crunching : Some task run longer than usual
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · Next

AuthorMessage
EG

Send message
Joined: 15 Dec 15
Posts: 7
Credit: 253,481,367
RAC: 0
Message 3736 - Posted: 25 Aug 2019, 23:37:44 UTC - in response to Message 3733.  
Last modified: 25 Aug 2019, 23:42:58 UTC

I think, we give too much credits recently. Seriously, at the time when I set credit it takes 6 hours to finish on single core on i7 CPU.
Recently, some tasks finish in 18 minutes on same machine.

I really don't like to be a person, who make "inflation" on BOINC projects...


Well, unfortunately, what your doing is credit deflation.

Your project, you can run it any way you like....
Obviously you don't like volunteer participation either......

To each his own I suppose....
ID: 3736 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Dingo
Avatar

Send message
Joined: 4 Feb 15
Posts: 17
Credit: 133,036,975
RAC: 0
Message 3737 - Posted: 26 Aug 2019, 5:23:45 UTC
Last modified: 26 Aug 2019, 5:26:25 UTC

Just a comparison from the same Raspbian V3 machine from one of the Team members that was crunching Universe and have since moved on.

Looking at a V3 that has validated tasks on both work units:

Universe New work unit and Credit

(24900.29/3600) / 666.7 credits = 6.916747222222222 ‬credits per hour


Universe old credit on same machine

(6711.44/3600) / 666.67 = 357.6001573432825‬ credits per hour

I think the extent of the drop is what is the issue, but if you can run your project without people volunteering their computers, at their expense, to do the work, carry on.
ID: 3737 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
additude

Send message
Joined: 6 Feb 19
Posts: 41
Credit: 131,325,319
RAC: 0
Message 3738 - Posted: 26 Aug 2019, 7:42:26 UTC

My Crunch times have gone from around 8,000 seconds to around 21,000 seconds with the 283 batch.
RAC is dropping accordingly.
I'm not whining and I'm not leaving.
Universe still provides a good credit compensation for work completed ratio.
Don't change anything, leave it just like it is.
I'm fine with it.
ID: 3738 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
davidBAM

Send message
Joined: 18 Oct 18
Posts: 6
Credit: 1,324,610,333
RAC: 0
Message 3740 - Posted: 26 Aug 2019, 9:24:47 UTC - in response to Message 3733.  

"I think, we give too much credits recently"

Well you are certainly going to see that decrease DRASTICALLY as the competitive crunchers leave the project.

You have also basically invalidated the Formula Boinc marathon for 2019 by the way. This may not matter to you but it matters greatly to many others.
ID: 3740 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Cruncher Pete

Send message
Joined: 23 Feb 15
Posts: 11
Credit: 183,472,470
RAC: 0
Message 3741 - Posted: 26 Aug 2019, 10:50:40 UTC - in response to Message 3738.  

I am whining and I am leaving. Why on earth did you change the credit system in the middle of the game. What chance has another user who now just joined have to catch up to another user who crunched the same amount of time with the same machine and now have no chance of catching up since he is getting 67% less credit. This is not fair and is not for me. Good luck retaining your users.
ID: 3741 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Hal Bregg
Avatar

Send message
Joined: 22 Oct 18
Posts: 19
Credit: 3,186,503
RAC: 0
Message 3742 - Posted: 26 Aug 2019, 11:55:00 UTC

This thread is going off-topic now.
I would suggest moving the discussion about credit to another thread.
ID: 3742 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
mmonnin

Send message
Joined: 2 Jun 16
Posts: 169
Credit: 317,253,046
RAC: 6
Message 3743 - Posted: 26 Aug 2019, 13:17:18 UTC - in response to Message 3742.  

Your run time question was answered. With a static credit, run time and RAC are intertwined.

Anyways, I agree, credit at U@H was pretty high compared to other projects. I don't agree that making tasks 2-3 times as along in one abrupt change was the best way to get back a normal CreditNew credit level. (This is not a request for CreditNew. You can check with NumberFields to see what happened with they moved to it.)
ID: 3743 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Christopher

Send message
Joined: 18 May 17
Posts: 16
Credit: 3,215,924
RAC: 0
Message 3744 - Posted: 26 Aug 2019, 13:20:54 UTC - in response to Message 3733.  

Everyone here has made some good points, Krzysztof (pun intended). The reality is that the silicon keeps getting faster. It always was and it always will. New tech is gonna grind out points faster. It will always appear that you are "giving out too many points" unless you constantly deflate.

However, deflation has it's risk. Competitive crunchers will move to projects that offer better return. People who rely on gridcoin to offset their power costs will also move on (unless practically everyone leaves, then the magnitude will go back up). And then, there's the issue of new users who won't be able to "catch up" and get a nice badge unless they use cutting edge hardware.

Yes, ultimately, it's all about the science. But credits help people see at a glace how much compute was accomplished (and, yes, that means that new and expensive machines win the arms race adjusting for compute time). If rules keep changing then the credit system won't reflect anything anymore.

I would propose that credit remains tied to the amount of work accomplished. If you would like to reward senior users who have crunched out work over longer periods of time on slower machines, you could consider a revamped badge system (ex. badges for years working on the project). In the worst case scenario, if you really want to deflate credit per amount of work, then I would suggest doing that when new projects come out (ex. ultraviolet or whatever).

I do agree with you; however, that work units will need to be enlarged as the silicon gets more powerful. Logistically and computationally it is more efficient. But I don't agree with credit deflation at least when the end user isn't really upgrading their silicon. They just feel penalized that they aren't on 7nm.

Just my two cents. As for me personally, I will be diversifying my compute operation. Not just because of this, but because of the reality of projects going dark after some time (ex. denis or enigma). I hope that you can get enough people to keep crunching this in order to keep the papers flowing to the publishers and keep the enthusiasm alive.
ID: 3744 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Topper

Send message
Joined: 12 Feb 18
Posts: 5
Credit: 3,012,667
RAC: 0
Message 3745 - Posted: 26 Aug 2019, 13:58:54 UTC

Holy Shit it now 7hours and before it was like 2,5... Holy Shit... And the same credits lol im out
ID: 3745 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
EG

Send message
Joined: 15 Dec 15
Posts: 7
Credit: 253,481,367
RAC: 0
Message 3746 - Posted: 26 Aug 2019, 16:28:21 UTC - in response to Message 3741.  
Last modified: 26 Aug 2019, 16:35:00 UTC

I am whining and I am leaving. Why on earth did you change the credit system in the middle of the game. What chance has another user who now just joined have to catch up to another user who crunched the same amount of time with the same machine and now have no chance of catching up since he is getting 67% less credit. This is not fair and is not for me. Good luck retaining your users.


I'm with Pete, it becomes an exercise in pointlessness.....
ID: 3746 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
davidBAM

Send message
Joined: 18 Oct 18
Posts: 6
Credit: 1,324,610,333
RAC: 0
Message 3747 - Posted: 27 Aug 2019, 15:13:51 UTC - in response to Message 3744.  
Last modified: 27 Aug 2019, 15:18:45 UTC

A wonderful post. Take a bow, Christopher.

I will certainly be off once the next couple of Formula Boinc sprints are announced. In going, can I just make the point that a completed WU advances the science of the project by precisely the same amount whether it be crunched by a competitive cruncher, by a Gridcoin cruncher or by a dyed-in-the-wool astrophysicist who crunches nothing else.
ID: 3747 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Jim1348

Send message
Joined: 28 Feb 15
Posts: 253
Credit: 200,562,581
RAC: 0
Message 3748 - Posted: 27 Aug 2019, 15:39:25 UTC - in response to Message 3747.  

In going, can I just make the point that a completed WU advances the science of the project by precisely the same amount whether it be crunched by a competitive cruncher, by a Gridcoin cruncher or by a dyed-in-the-wool astrophysicist who crunches nothing else.

In staying, I can point out that the number of crunchers has increased from 1848 to 1871 in the last day. Maybe the ones leaving are making room for the ones who want to do it.
ID: 3748 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
gambatesa

Send message
Joined: 30 May 18
Posts: 1
Credit: 1,009,518,041
RAC: 0
Message 3749 - Posted: 27 Aug 2019, 18:04:42 UTC - in response to Message 3748.  

Regarding to Gridcoin.. the RAC will decrease for everyone and the Magnitude will adjust.. the only problem should be users that cherrypick the faster Work Units.. but in a couple of weeks probably the fast workunits will expire soon.. no problem for me..i believe in the project and i'll not switch my hosts..
ID: 3749 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Pete Broad

Send message
Joined: 7 Jan 16
Posts: 20
Credit: 208,458,667
RAC: 22
Message 3750 - Posted: 27 Aug 2019, 19:56:00 UTC

I'm sticking with this project but I'm a bit concerned about the number of units I have awaiting validation. Its been hovering around the 900 mark for quite a time now while its usually 200-250!

Pete
ID: 3750 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Christopher

Send message
Joined: 18 May 17
Posts: 16
Credit: 3,215,924
RAC: 0
Message 3751 - Posted: 29 Aug 2019, 17:22:11 UTC

You know, quite frankly I would agree to a certain compromise considering that a bit of deflation might be necessary when compared to certain projects on BOINC. I would propose increasing the points per WU to an even 1000 or 1500. Nice, round, easy, an increase, but still some deflation gets accomplished. idk just my two cents.
ID: 3751 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Krzysztof Piszczek - wspieram ...
Project administrator
Project developer
Project tester
Avatar

Send message
Joined: 4 Feb 15
Posts: 841
Credit: 144,180,465
RAC: 2
Message 3752 - Posted: 30 Aug 2019, 17:01:58 UTC - in response to Message 3751.  

Ok.
I see that current points for WU's are problem for quite large group of us. Also, I see that small group of WU's now calculate even longer that most of them,
So, from next batch I will increase amount of credits by factor of 1.5 or 2 (I need bit more observation of current computation time)...
Krzysztof 'krzyszp' Piszczek

Member of Radioactive@Home team
My Patreon profile
Universe@Home on YT
ID: 3752 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
EG

Send message
Joined: 15 Dec 15
Posts: 7
Credit: 253,481,367
RAC: 0
Message 3753 - Posted: 30 Aug 2019, 20:36:22 UTC - in response to Message 3752.  
Last modified: 30 Aug 2019, 20:38:11 UTC

Ok.
I see that current points for WU's are problem for quite large group of us. Also, I see that small group of WU's now calculate even longer that most of them,
So, from next batch I will increase amount of credits by factor of 1.5 or 2 (I need bit more observation of current computation time)...


In practical reality, you should be increasing the points proportional to the increase in work done. since the increase was 3x (or thereabouts) the point increase should reflect that.

The only point in doing such an increase is to relieve server overwork from too many network requests.... but to penalize your clients/users for the amount of computer time and power they donate to you?

Nonsensical even from a scientific point of view. Yes you have a number of users who exclaim the science as their motivation, you will find that those who claim such only represent about 10% of the clients/users who contribute. The rest are motivated by the leaderboard..... Eliminate their ability to climb the leaderboard, you eliminate their desire to contribute to your project, just a fact of the Distributed Computing world......

Ultimately, you get to decide which direction you want your project to go......

But, many of your users, like me, have been crunching since before Boinc was even thought of. And if it wasn't for the leaderboard, there wouldn't be a Boinc.....
The leaderboard is what keeps your project viable from an operations standpoint... I know that isn't something most scientists want to recognize, but it is the facts of life in the DC world....

I sincerely hope you make a wise decision, and wish you well in your future endeavors with the project...
ID: 3753 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Dataman
Avatar

Send message
Joined: 24 Feb 15
Posts: 32
Credit: 609,507,165
RAC: 0
Message 3754 - Posted: 31 Aug 2019, 3:17:33 UTC - in response to Message 3752.  

Ok.
I see that current points for WU's are problem for quite large group of us. Also, I see that small group of WU's now calculate even longer that most of them,
So, from next batch I will increase amount of credits by factor of 1.5 or 2 (I need bit more observation of current computation time)...

Good decision. I will be back once my queue drains.
Cheers.
ID: 3754 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
JugNut

Send message
Joined: 11 Mar 15
Posts: 37
Credit: 271,242,973
RAC: 0
Message 3755 - Posted: 31 Aug 2019, 14:49:15 UTC - in response to Message 3754.  

Yes good news indeed :)
ID: 3755 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Penguin

Send message
Joined: 28 Mar 18
Posts: 24
Credit: 157,069,182
RAC: 1,634
Message 3758 - Posted: 2 Sep 2019, 15:30:30 UTC - in response to Message 3752.  

Ok.
I see that current points for WU's are problem for quite large group of us. Also, I see that small group of WU's now calculate even longer that most of them,
So, from next batch I will increase amount of credits by factor of 1.5 or 2 (I need bit more observation of current computation time)...


Yes that is something. Thank you for at least some kind of increase to compensate for the increased computation times. Credit reward should scale with amount of the calculations.
ID: 3758 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Previous · 1 · 2 · 3 · Next

Message boards : Number crunching : Some task run longer than usual




Copyright © 2024 Copernicus Astronomical Centre of the Polish Academy of Sciences
Project server and website managed by Krzysztof 'krzyszp' Piszczek