credits
log in

Advanced search

Message boards : Number crunching : credits

1 · 2 · Next
Author Message
Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 3606
Credit: 22,879,153
RAC: 0
Message 589 - Posted: 9 Jan 2015, 21:40:47 UTC

I want to go back to credit new for some reasons. There are pro and cons.

pro:
- dont need to split a base into 3-4 parts (less work)
- can load a complete base
- dont need to estimate or calculate higher n values (400+)

cons:
- for restart on a checkpoint the time will not be saved, only the rest time influence into the credits.
- AVX should not have any advantage in credit rate.

What do you think?

Profile Cruncher Pete
Send message
Joined: 3 Dec 14
Posts: 16
Credit: 65,930,895
RAC: 27,515
Message 590 - Posted: 9 Jan 2015, 21:58:56 UTC - in response to Message 589.

Rebirther if you do that, you will loose me (not a big deal) and lots of other users I am sure, that will make your job a lot easier.

Playing with users credits like this is not fair. Just think of any new comers. They will have no chance to catch up to the top users as they would have received more credits for the same amount of work.

It might just be me but I do not like Credit New for the same reason as stated. I purposely crunch those projects that do not use Credit New for it is not fair to anybody to move the goal post in the middle of the game, favouring one team or user.

Just think. In order to be fair to all, you would have to play with everybodies score yet once again and reduce their credits to what they should have received had you started with Credit New.

Keep on crunching and have a nice day...

Profile Michael Goetz
Avatar
Send message
Joined: 1 Jan 15
Posts: 18
Credit: 303,916
RAC: 0
Message 593 - Posted: 9 Jan 2015, 22:17:20 UTC - in response to Message 589.

- AVX should not have any advantage in credit rate.


That's understating it a bit. What actually happens is that AVX (and to a larger extent, FMA3) CPUs compute faster than non-AVX CPUs, but have the same BOINC benchmark numbers, if everything else is equal. Therefore, to BOINC, the AVX CPU did less work, and gets less credit for the same task than does the non-AVX CPU.

From a credit-hunter's perspective, this means SRBase will be a good place to put AMD CPUs since they'll get about the same credit per core-hour as AVX CPUs. Under a more fair credit system, AMD CPUs get fewer credits per core-hour.

People who care more about the conjectures and primes than about credits and badges won't change their habits.

As someone with a lot of experience in this area (I'm one of the people who run PrimeGrid, for those who don't know me), getting the credits right with LLR is a horribly complex problem. The built-in BOINC credit system works BACKWARDS with AVX-enabled apps, rewarding slower non-AVX computers with more credit. That's not ideal, but there's no easy fix. Rebirther's doing the best he can -- and he's got a lot of details to take care of just keeping SRBase running without worrying about credit all the time.

Please keep in mind this is a BETA project -- it's going to be a while until all the kinks are worked out. PrimeGrid was running for 8 years before we figured out a way to make the credit work reasonably well. Internally, PrimeGrid and SRBase are very different, so it's not easy to just take our credit system and migrate it over here. (If it was, Rebirther would have already done it.)

Joe
Send message
Joined: 20 Dec 14
Posts: 8
Credit: 16,735,286
RAC: 1,681
Message 595 - Posted: 9 Jan 2015, 23:36:46 UTC - in response to Message 589.

Credit new and I like many will not crunch your wu's.

The only fair system is giving a set amount of credit for each work unit.

"Credit New" bye bye

Profile Michael Goetz
Avatar
Send message
Joined: 1 Jan 15
Posts: 18
Credit: 303,916
RAC: 0
Message 596 - Posted: 10 Jan 2015, 1:00:04 UTC

Rebirther,

What you might want to try is to calculate credit dynamically on CONSTANT * (log10(k)+log10(b)*n)^2. I.e., credit is proportional to the size of the number squared. (Adjust constant as needed so credit is reasonable.) The credit will scale nicely with the size of the task.

That won't take FFT size variations into account, so it's not perfect, but it's significantly easier to do than a system that looks at FFT sizes, is MUCH better than BOINC's credit (new or old), and saves you from having to manually assign fixed credit to every batch of tasks.

To everyone who is saying "Use fixed credit or else!" -- fixed credit is not as good as you think it is. There's a lot of variation in task run lengths from similar looking numbers, and with fixed credit you get as much credit for the longer tasks as for the shorter tasks. Look at this another way -- if the other guy is systematically aborting all of his longer running tasks and only running the shorter ones, thus leaving the longer tasks for other people (like yourself), that means he's getting more credit per hour. It's still better than BOINC credit, but not nearly as good as it could be, and is VERY labor intensive for the admin -- prohibitively so. There may come a point where Rebirther could decide it's just too much work (and the crowd is too nasty) for it to be worth his trouble. You might want to try being a bit more supportive of all the work he's put into this site, and make the criticism a bit more constructive. He's doing it for you.

Profile Cruncher Pete
Send message
Joined: 3 Dec 14
Posts: 16
Credit: 65,930,895
RAC: 27,515
Message 597 - Posted: 10 Jan 2015, 9:37:39 UTC - in response to Message 593.



From a credit-hunter's perspective, this means SRBase will be a good place to put AMD CPUs since they'll get about the same credit per core-hour as AVX CPUs. Under a more fair credit system, AMD CPUs get fewer credits per core-hour.

People who care more about the conjectures and primes than about credits and badges won't change their habits.


I am one of those that do care about fair credit and badges for the priviledge of running this project. It should be the other way around for it seems that to a project credit is irrelevent and they think we will spend thousand of dollars on equipment and power to have the priviledge of working for them for nothing in return.

In reply to rebirther's statement that he wants to know what we think about
changing the credits to Credits New, I replied. It seems he has made up his mind for he has written that statement today and without waiting for anybody elses comment he has already changed the credit to Credit New. Furthermore, the credit system has been changed a number of times in the last few days, yet in the News Section of the project it was only mentioned once and no statement about the sudden reduction in credit.

The only way I can object to this kind of statement and management is to cease crunching for this project. Whilst I accept that this project is still in Beta and problems likely to occur Credit New sucks and is unfair to users who have purchased powerful and fast machines favouring less efficient ones. If you do not care about credits and think that users will stil follow you why don't you stop credit altogether and see if your statement is correct. Just make a statement in the News Section that from now on this project will not give any credit for it is designed for those who are interested in Prime Numbers only.

Keep on crunching and have a nice day...

Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 3606
Credit: 22,879,153
RAC: 0
Message 599 - Posted: 10 Jan 2015, 9:41:23 UTC - in response to Message 597.



From a credit-hunter's perspective, this means SRBase will be a good place to put AMD CPUs since they'll get about the same credit per core-hour as AVX CPUs. Under a more fair credit system, AMD CPUs get fewer credits per core-hour.

People who care more about the conjectures and primes than about credits and badges won't change their habits.


I am one of those that do care about fair credit and badges for the priviledge of running this project. It should be the other way around for it seems that to a project credit is irrelevent and they think we will spend thousand of dollars on equipment and power to have the priviledge of working for them for nothing in return.

In reply to rebirther's statement that he wants to know what we think about
changing the credits to Credits New, I replied. It seems he has made up his mind for he has written that statement today and without waiting for anybody elses comment he has already changed the credit to Credit New. Furthermore, the credit system has been changed a number of times in the last few days, yet in the News Section of the project it was only mentioned once and no statement about the sudden reduction in credit.

The only way I can object to this kind of statement and management is to cease crunching for this project. Whilst I accept that this project is still in Beta and problems likely to occur Credit New sucks and is unfair to users who have purchased powerful and fast machines favouring less efficient ones. If you do not care about credits and think that users will stil follow you why don't you stop credit altogether and see if your statement is correct. Just make a statement in the News Section that from now on this project will not give any credit for it is designed for those who are interested in Prime Numbers only.

Keep on crunching and have a nice day...


I have only changed the R3 to credit new because there are still some 900cr WUs around. I will changed it back if all results are returned.

Profile Cruncher Pete
Send message
Joined: 3 Dec 14
Posts: 16
Credit: 65,930,895
RAC: 27,515
Message 610 - Posted: 12 Jan 2015, 7:37:14 UTC - in response to Message 596.

You might want to try being a bit more supportive of all the work he's put into this site, and make the criticism a bit more constructive. He's doing it for you.


I do not understand Michael. Had you read my remarks in this and other threads to rebirther you would have found that I suggested that rebirther run a test server before he does any adjustment to the main server. I even offered a machine of mine to help him out. I suggested that he makes more use of the News Section of the project to keep users up to date. To date there is no mention about the 900 x error and its subsequent adjustment to scores or in deed that the project is now using Credit New. Are they not constructive criticisms?

As regards your statement that : "He is doing it for me." Please explain how is that so. He keeps changing the credits from day to day and does not explain to us why he is doing that in the News. Not everybody has time to search the Message Box where threads are often gets kind of hidden among a number of others. He reduces my credit and does not give credit at all for Work Units I submitted. (Divided by 900 but did not multiply by 5 to give me credit for my work) This has been proven by one user and his scores were adjusted. Why all of our other involved users score were not subjected to the same correction.) When questioned rebirther did offer to adjust mine as well if I give him my previous scores. I can not do that for my system do not keep a database. That sort of information is only available to project Admins.

All I am asking is to treat everybody equally and tell us what is going on in the News section. I have been around long enough to realize that it is not easy to run a New project and I accept that bugs, mistakes and errors will occur. What I will not accept is statement like yours that I should be more constructive in my criticism and remarks like He is doing it for you. I say again, How so? Is he going to pay for the Electricity costs that I have wasted by stuffing up my credits? This is an open project in Beta but is not a Test project. It is time Projects realize that we are not just cannon fodder for we are HELPING THEM and they are not helping us.

Profile Michael Goetz
Avatar
Send message
Joined: 1 Jan 15
Posts: 18
Credit: 303,916
RAC: 0
Message 611 - Posted: 12 Jan 2015, 8:19:08 UTC - in response to Message 610.

You might want to try being a bit more supportive of all the work he's put into this site, and make the criticism a bit more constructive. He's doing it for you.


I do not understand Michael.


Indeed, I don't believe you were one of the people my "try to be constructive" comments were aimed at. I didn't want to name individuals, so I apologize if you felt included in that wide net. There have been a few "Do this or I quit!" types of posts which don't seem to me to be particularly helpful other than to point out that they don't like credit new.

By the way, I've decided to try not to post here anymore. I'm not sure if I'm helping or hurting Rebirther, so I'm just going to keep my mouth shut from now on. I wish him luck with the site and I wish everyone else luck with knocking off k's from all of the CRUS conjectures.

Profile shiva
Avatar
Send message
Joined: 22 Dec 14
Posts: 1
Credit: 10,323,587
RAC: 0
Message 616 - Posted: 12 Jan 2015, 14:15:01 UTC
Last modified: 12 Jan 2015, 14:19:58 UTC

I do not pretend to know any thing about the problems of credit. I do know that I started doing this in 1999 and I have worked a few projects and read a lot about credit. I will say that I do this for fun, and credit, and badges. I will work your project till I get a million credits, then move to a new project if there is a new one. I will come back to your project more or less in the order of credit payment. I have a few projects with over 2.5 million credits, 22 to be exact. So I'm just a user that wants to help out and sees credit and badges as my payment. I will say it does upset me when my credit payment goes crazy with changes. I have no idea why I have gotten 900 credits for something I was getting 5. I think 900 is to high and I really don't think 5 is enough. I moved about 6 more machines to this project yesterday and came here trying to figure out why I only had 1933 credits yesterday. if that is all I'm getting I will move the 6 new boxes on to something else because that isn't worth my time and effort. By the way rebirther, thanks for the heads up the other day that I had a machine that was acting up and creating a bunch of errors. I really hate errors. :) Oh yea I really don't care for credit new.

Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 3606
Credit: 22,879,153
RAC: 0
Message 617 - Posted: 12 Jan 2015, 16:27:39 UTC - in response to Message 611.

You might want to try being a bit more supportive of all the work he's put into this site, and make the criticism a bit more constructive. He's doing it for you.


I do not understand Michael.


Indeed, I don't believe you were one of the people my "try to be constructive" comments were aimed at. I didn't want to name individuals, so I apologize if you felt included in that wide net. There have been a few "Do this or I quit!" types of posts which don't seem to me to be particularly helpful other than to point out that they don't like credit new.

By the way, I've decided to try not to post here anymore. I'm not sure if I'm helping or hurting Rebirther, so I'm just going to keep my mouth shut from now on. I wish him luck with the site and I wish everyone else luck with knocking off k's from all of the CRUS conjectures.


Dont worry Michael, Iam still tweaking and find the best average credit rate for all ranges.

Profile Coleslaw
Avatar
Send message
Joined: 1 Dec 14
Posts: 8
Credit: 3,228,910
RAC: 1,070
Message 624 - Posted: 13 Jan 2015, 15:51:02 UTC
Last modified: 13 Jan 2015, 15:56:34 UTC

I agree that I do not care for Credit New at all. Those projects do have an unfair bias that has turned quite a few of the bigger crunchers off to the projects that use it. I understand making things easier for the administrators, but I also feel that they need to keep their donors happy. Anyone can say a project was made for the donors, but as mentioned above there is no answer to how that really is. Prime numbers may be used in encryption, but lets face it, the mass majority of us are not suffering due to the lack of Prime numbers. So, I don't see a prime generating project to be for "us". A project that seems to be for "us" is WUProp as the data is very helpful to "us". A project like WCG that is putting out humanitarian research results could easily be argued as being for "us" even though that approaches a large grey area.

I think rebirther does a good job with the limited resources. However, as others have suggested, he needs to work on his organizational skills when it comes to information posted in the forums. I have a heck of a time at BU gathering the current information because it is so clustered. This may not be BU, but it seems to have a similar issue with information. This is very frustrating.

I have withheld contributing a lot here simply because it is new and going through the growing pains as many projects do. However, calling it a BETA project and justifying its failings in that way is just stupid IMO. I have argued this before as some project ride that "BETA" title for years. Technically all projects are always growing and refining. If you have live usable data, then you should be considered production. Again, that is my opinion. That doesn't mean they can't have BETA work. If a project is labeled ALPHA or BETA, to me that means they are ironing out their project and don't have production work going. Test servers should be used for that. A good example of a BETA project should be YAFU as they use the project for testing the latest BOINC server software on top of other work. It is used to test multi-threaded apps and how Credit New works.

Michael Goetz, there is no reason why you shouldn't continue posting. I think even if your comments are addressed to rebirther, others may learn from them.
____________

Profile Michael Goetz
Avatar
Send message
Joined: 1 Jan 15
Posts: 18
Credit: 303,916
RAC: 0
Message 629 - Posted: 13 Jan 2015, 17:19:43 UTC - in response to Message 624.

Michael Goetz, there is no reason why you shouldn't continue posting. I think even if your comments are addressed to rebirther, others may learn from them.


I think I should clarify what I meant when I said "the project is for the users".

There are many people who have put a lot of effort into searching for prime numbers, and more relevant to SRbase, to working on proving the various Sieprinski and Riesel conjectures. I'm one of those people. So is Rebirther. It's fair to say nearly everyone at the Mersenne forums and those involved in CRUS also fall into that category.

It's THOSE people for whom I see SRbase as being beneficial. It provides them with an easy way to organize their efforts. That's what I meant by "it's for the users."

I apologize for not being clearer or more inclusive.

brinktastee
Avatar
Send message
Joined: 29 Dec 14
Posts: 5
Credit: 340,916
RAC: 0
Message 701 - Posted: 18 Jan 2015, 17:42:14 UTC - in response to Message 624.

Michael Goetz, there is no reason why you shouldn't continue posting. I think even if your comments are addressed to rebirther, others may learn from them.

+1

Profile Steve Hawker*
Avatar
Send message
Joined: 8 Dec 14
Posts: 26
Credit: 610,307
RAC: 0
Message 739 - Posted: 21 Jan 2015, 22:29:24 UTC

Reb,

+1 for no CreditNew

On the topic of credits, can you comment on the variation in cr/hr? I'm seeing this on my MacBook Pro:

Riesel Base v0.01 35
Riesel Base - short v0.01 16
Sierpinski / Riesel Base v0.02 20
Sierpinski / Riesel Base - average v0.01 41
Sierpinski / Riesel Base - long v0.01 20
Sierpinski / Riesel Base - short v0.01 31
Sierpinski Base v0.01 73
Sierpinski Base - short v0.01 18

Profile rebirther
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 2 Jan 13
Posts: 3606
Credit: 22,879,153
RAC: 0
Message 740 - Posted: 22 Jan 2015, 6:04:04 UTC - in response to Message 739.

Reb,

+1 for no CreditNew

On the topic of credits, can you comment on the variation in cr/hr? I'm seeing this on my MacBook Pro:

Riesel Base v0.01 35
Riesel Base - short v0.01 16
Sierpinski / Riesel Base v0.02 20
Sierpinski / Riesel Base - average v0.01 41
Sierpinski / Riesel Base - long v0.01 20
Sierpinski / Riesel Base - short v0.01 31
Sierpinski Base v0.01 73
Sierpinski Base - short v0.01 18


Iam still tweaking. The longrunners are not easy.

Roadranner
Send message
Joined: 10 Dec 14
Posts: 7
Credit: 5,407,836
RAC: 0
Message 2336 - Posted: 8 Feb 2016, 15:15:52 UTC
Last modified: 8 Feb 2016, 15:25:15 UTC

Reb,

just a calculation:

R/B average - about 5 hours: 720 cr
R/B long - about 7 hours: 580 cr

Edit:

Please ignore the above - I mixed up different machines.

Crunchi
Avatar
Send message
Joined: 1 Jan 15
Posts: 3
Credit: 599
RAC: 0
Message 2390 - Posted: 20 Feb 2016, 21:50:51 UTC
Last modified: 20 Feb 2016, 21:51:37 UTC

There is two types of persons here: one looking for prime, and second looking for credits, badges , and so on.
If you are looking for prime, and to solve base, then you dont care how many credits you will get, main and only reward is your name associated to prime.
If you are hunting credits, and badges, then you will leave this place once you got your goal.
So if you think that credit system is unfair, then you can take your computers "with you" and go to many BOINC projects you have.
If you like to help solving CRUS bases, then you can sole it here and in bons earn some credits, or like my doing that at home, without any help from anyone.
Just make sieve and press START LLR and wait :)
When you find it, it is yours.

Whatever group of people belonging to, I wish you many primes to found!
And last, : noone is forcing you to be on this project. Noone holding your hand.
You can leave any moment you decide, without any problems
That is your right to do.
But in that case, dont say: credit system is unfair. It is irrelevant to you, you are left project.
Best regards to all

Profile Steve Hawker*
Avatar
Send message
Joined: 8 Dec 14
Posts: 26
Credit: 610,307
RAC: 0
Message 2607 - Posted: 29 May 2016, 16:29:47 UTC - in response to Message 2390.

There is two types of persons here: one looking for prime, and second looking for credits, badges , and so on.

I've given up on credits almost entirely. I crunch for credits until I get 25,000 then I crunch for hours. I've reached that milestone on nearly all my active projects. Badges will happen along the way. Finding a prime is a delicious bonus.

Profile vaughan
Send message
Joined: 9 Dec 14
Posts: 52
Credit: 139,444,321
RAC: 299,613
Message 2618 - Posted: 4 Jun 2016, 1:37:30 UTC - in response to Message 2607.

There is two types of persons here: one looking for prime, and second looking for credits, badges , and so on.

I've given up on credits almost entirely. I crunch for credits until I get 25,000 then I crunch for hours. I've reached that milestone on nearly all my active projects. Badges will happen along the way. Finding a prime is a delicious bonus.

I'm a third variation!

I initially crunched to reach a personal target 1M in the project. Then on reaching that it was to get 10M per sub-project - this is still a work in progress and is taking a very long time. To assist the process I have purchased several Xeon class server CPUs (such a shame the SuperMicro motherboards are so expensive when the CPUs are reasonably priced).

My preference is for shorter running tasks as I prefer to run my computers at idle speeds overnight so I can sleep. If a challenge is on however, I'll run my farm flat out 24/7 as I'm super competitive :)

I totally agree with Steve Hawker's observation that "finding a prime is a delicious bonus".
____________

1 · 2 · Next
Post to thread

Message boards : Number crunching : credits


Main page · Your account · Message boards


Copyright © 2014-2018 BOINC Confederation / rebirther