I don't have anything to lose with credit being dropped. <smile>

The problem with jerking granted credit around is that new users can't be compared to older users. Instead of focusing so much on BOINC-wide values, David Anderson and the BOINC projects as a whole should totally give up on any and all ideas about making Alberts = ETs = ClimateCycles = ABCs = Cosmos = MilkyWays and just freeze BOINC-wide standings as they are today and let the stat sites come up with their own metrics, much like BOINCStats has the "World Cup" points... The cobblestone system is so hopelessly flawed in implementation that with all of the ups and downs over the years, current rankings are really meaningless from a historical standpoint, only having importance for the here and now...until the next change in the NPV of the "cobblestone".

Credit history comparisons can really only be compared between similar types of work units. Multiple trendlines can be created (cropped between credit/application/workunit changes) which shows progress clearer than examining or comparing histories of two users or projects.

Are cobblestones flawed? Maybe, but anything else that is substituted in will be just as flawed when you compare projects. FLOP comparison will never work.

There does not exist a universal currency between projects. Perhaps we should setup a credit exchange program?

If cobblestones are flawed then what is the use of milestones?

---

A BLAST FROM YOUR PAST

I don't have anything to lose with credit being dropped. <smile>

The problem with jerking granted credit around is that new users can't be compared to older users. Instead of focusing so much on BOINC-wide values, David Anderson and the BOINC projects as a whole should totally give up on any and all ideas about making Alberts = ETs = ClimateCycles = ABCs = Cosmos = MilkyWays and just freeze BOINC-wide standings as they are today and let the stat sites come up with their own metrics, much like BOINCStats has the "World Cup" points... The cobblestone system is so hopelessly flawed in implementation that with all of the ups and downs over the years, current rankings are really meaningless from a historical standpoint, only having importance for the here and now...until the next change in the NPV of the "cobblestone".

Agreed completely :)
With any credit per flop change (no matter up or down! ) and with change from credit per hour to credit per flop any total credit value become useless. Only RAC matters a litte from performance measurement point of view. But again, almost only inside single project, cross-project measurements still impossible.

Milkyway should add a counter for counting done Workunits - to compare users

Milkyway should add a counter for counting done Workunits - to compare users

Already passed stage. SETI had task counting when SETI@home classic was active.
Since that time tasks bacome very different in workload they have so just counting their number became useless.
Here in MW same situation AFAIK, different tasks types have different run time on very same software/hardware combo.

Milkyway should add a counter for counting done Workunits - to compare users

Do you think the Top Participant, Host, or Team List would be any different than it is now by counting WU's done ???

Ahhh -

This thread explains why I have seen such a huge drop in credit.. I was just about ready to start looking for crashed boxes. But it now appears that the project staff decided to cave in and go back on their word about not lowering the credit.
Go figure.

---

.

The following is what I posted on seti@home.. I hope they like it...




Message 932967 - Posted 13 Sep 2009 4:00:18 UTC


Last modified: 13 Sep 2009 4:01:50 UTC

Ok now that the politics has landed on the Milky way project to make them bring their project more in line with projects like SETI. Where is the ATI application on SETI so that I can get a copy to use? I mean come on I want to use it to compare with Milky Way project.

Maybe I am mistaken and maybe SETI does not have one for this project. If not why would you bully other projects to conform to something that you don't have here. Just think about that one..

Heck you can't even keep this project running for more that two days.

SM..
____________

Nice !!!

---

.

Milkyway should add a counter for counting done Workunits - to compare users

Yah, ok - and we could then apply a factor of say 4.5 or 6.2 or some other arbitrary number in a lame attempt to keep all BOINC projects on so called even playing field, and we could call them...cobblestones.

Great idea !!!!!!

---

.

If cobblestones are flawed then what is the use of milestones?

I wish they'd use headstones on the idea of cross-project comparisons and let the rankings in each project be totally separate from one another. If this is done, "milestones" in individual projects would actually mean something. As it is now, if I did 100,000 tasks to get 1,000,000 credits (10 credits each), with credit lowering always in play, someone will have to have done MORE work for the same apparent ranking / work done metric.

What I've been in favor of for several years now is just letting each project set their own credit with BOINC no longer supporting any kind of cross-project rankings at all. If 3rd party statistics sites wish to come up with their own scheme, that's fine. It would be 3rd party and not BOINC. We could basically just go down to 1 credit for 1 completed task. If a different project decided it wanted to award 1 trillion credits for 1 completed task, that would be fine as well, since it is confined to that project and that project alone. Projects could say that there is not any intention of comparing the work done in their project vs. work done in another project and that users can take up their complaints about any cross-project comparisons with the 3rd party stat sites. This means the project staff is freed up and the BOINC staff can concentrate solely on improving the software in tangible terms, not intangible scoring metrics.

In other words, it is the attempt to equate the work done that is the root problem. If no effort is made to equate work between projects, this "problem" that the BOINC administration created would instantly be solved.

I basically agree that the whole comparison is flawed because someone decides that my I7 has to be the same as some 386 way back when (still ignores the 3 other cores and 4 hyperthreads). If you are going to do that then you have to counter the increased performance due mainly to CPU changes but also code or compiler changes. Personally I think GPU credits should be counted as a different from CPU partly do with to the technology but also the projects that use them.

My goals are more within a project than across them. I choose projects I am interested or in the case of GPU's something that can use it. I would do more GPUGRID but I really not appreciate the long run times with short deadlines.

Boincstats has it cup points but it is also flawed especially compared to projects that do not have continuous work. Also you have to allow those computers that are not working 24/7.

If cobblestones are flawed then what is the use of milestones?

I wish they'd use headstones on the idea of cross-project comparisons and let the rankings in each project be totally separate from one another. If this is done, "milestones" in individual projects would actually mean something. As it is now, if I did 100,000 tasks to get 1,000,000 credits (10 credits each), with credit lowering always in play, someone will have to have done MORE work for the same apparent ranking / work done metric.

What I've been in favor of for several years now is just letting each project set their own credit with BOINC no longer supporting any kind of cross-project rankings at all. If 3rd party statistics sites wish to come up with their own scheme, that's fine. It would be 3rd party and not BOINC. We could basically just go down to 1 credit for 1 completed task. If a different project decided it wanted to award 1 trillion credits for 1 completed task, that would be fine as well, since it is confined to that project and that project alone. Projects could say that there is not any intention of comparing the work done in their project vs. work done in another project and that users can take up their complaints about any cross-project comparisons with the 3rd party stat sites. This means the project staff is freed up and the BOINC staff can concentrate solely on improving the software in tangible terms, not intangible scoring metrics.

In other words, it is the attempt to equate the work done that is the root problem. If no effort is made to equate work between projects, this "problem" that the BOINC administration created would instantly be solved.

Yeah!
And this will solve another problem I mentioned earlier.
Different projects have VERY DIFFERENT OPTIMIZATION LEVELS
Moreover, they have different algorithms and different scientific goals.
How all this can be compared at all???
If some project use bad approach to solve its scientific goal - will helping to such project cost equal ?
In short, each project is self-contained separate entity and all this egalitarianism just deciving and unneeded...
If each project will have its own metric no "fairplay rule" would be at all.
PArticipants will chose projects for their scientific merits, not how much they pay.

But competition inside each project will be still possible, so all fun stay with us, all negative gone :)

And last but not least, no one will then blame SETI for credit issues !!!!!

Looks as though Milkyway@Home has installed the "Restrictor Plate"(NASCAR Racing Reference) again in order to slow all of us down again. I'll be back when the pastures are greener again. Now if they would pay part of my electric bill each month then lowering the credit wouldn't be a problem, but since they don't I look to get the most bang for the "Proverbial" buck ($$$). You can't compare one project with another since the science, goals, and scientific formulas are vastly different. It's like pricing apples and peanuts, they are completely different.

---

I have since put some machines on the seti@home project. It is my hopes that they find a smarter group of scientists out in space who will understand common sense.

And also common curtsy about using other folks equipment.

Like informing the host of credit changes far in advance before doing it. Thus giving the owner of the computer the option of pulling his machine off of that project sooner if they choose to do so.

I do remember a while back when Travis made the statement that he was following DA's formula for credits compared to cpu one, and now he has lied and gone and changed his world.

Go figure. Do you learn that sort of tactics at his university of DA?

SM...

---

OK people, lets just please simmer down here a little bit. I really do see no point in getting all uptight about it. If you think that the credits awarded here are not to your liking, then you obviously have the option to go and crunch for another Boinc project instead.

If you feel as I do, that the project here is worth while supporting, then the credits awarded are secondary. Boinc has always had credit hounds, they come, they make a noise, and they go. Yadda yadda yadda .....

@ Starman - Hope you stay with Sicituradastra :-)

---

Don't drink water, that's the stuff that rusts pipes

This concerns me. I am trying hard to get into thye BOINC top 100. I stand at 106 currently. Credits are not secondary. They may be for you, but they may not be for others. Thwarting achievements by manipulating credits - well, you may as well thwart continuing interest to participate in this or any project.

With respect, I join those sounding concern about the credit reductions and hope to hear what the project has to say about it soon.

---

Others lost credit too so your place in top 100 will not change much.
If only your credit was lwered it would be different situation ;)

Others lost credit too so your place in top 100 will not change much.
If only your credit was lwered it would be different situation ;)

Ah, but not all others are crunching MW, so I get cut while they don't ;)

---

Others lost credit too so your place in top 100 will not change much.
If only your credit was lwered it would be different situation ;)

??????


What????

That is ONLY assuming that those in front of him are crunching MW. AFAIK, the other projects have not gone and done the stupid thing like reducing credit, Unannounced, or, after it was reduced.

Like the CS project admins here.

---

.