Message boards :
Number crunching :
Seti Enhanced Credit Fair?
Message board moderation
Previous · 1 . . . 18 · 19 · 20 · 21 · 22 · 23 · Next
Author | Message |
---|---|
W-K 666 Send message Joined: 18 May 99 Posts: 19401 Credit: 40,757,560 RAC: 67 |
At the moment on Eistein you are able to take advantage of optimised applications. So bearing that in mind, Einstein, if a host is using the opptimised app, should be taken out of the comparison. All other projects will give approx the same credits/time. If you can identify were the developers started from it sometimes gives a hint as to which cpu/OS combination works best. For instance the CPDN app is an adaption of an older program written for mainframes etc using Fortran. So it has to be recognised that for some projects their apps are more suited to some cpu/os combinations than others. Some projects like cpu speed, some a large L2 cache (Seti) and some give better performance with large amounts of fast memory. So if you want to chase for the maximum credits, do your research and put your computer(s) where they will gain most credits. Andy |
Idefix Send message Joined: 7 Sep 99 Posts: 154 Credit: 482,193 RAC: 0 |
Hi, Hello all, Ok, one last try ... First off all, it's absolute irrelevant what a single user gets or what a single team gets. The stats for the entire project are important. That's the reason why I asked to explain the stats of the project and not the stats of user A or team B. The RAC: *Everybody* has observed a drop of the RAC for now more than 2 weeks. If the RAC really reflects the credit production in terms of "credit per day" or "credit per week" why in God's sake doesn't this reduced credit production show up in the credit graph of the project? Why in God's sake is there still a straight line in the credit graph although many people have been stating for two weeks now that their RAC is in free fall? Everybody observed the drop of the RAC. If the RAC really is as important as some of you want that it is why don't the total credits of the project show any evidence? Regards, Carsten |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13854 Credit: 208,696,464 RAC: 304 |
Everybody observed the drop of the RAC. If the RAC really is as important as some of you want that it is why don't the total credits of the project show any evidence? Because RAC is a meaningless, useless, irrelevant number. But still they get all worked up over it. Grant Darwin NT |
SargeD@SETI.USA Send message Joined: 24 Nov 02 Posts: 957 Credit: 3,848,754 RAC: 0 |
Maybe my question is unanswerable, I don't know, but everybody is just dancing all around it. If there are still MAJOR differences between each projects credit granting, then what has the change here accomplished? If the intent was to equalize the projects it appears to have failed. Otherwise there would not be projects with higher granted credit for me to move to. |
SargeD@SETI.USA Send message Joined: 24 Nov 02 Posts: 957 Credit: 3,848,754 RAC: 0 |
Everybody observed the drop of the RAC. If the RAC really is as important as some of you want that it is why don't the total credits of the project show any evidence? RAC is only meaningless if it means nothing to the person using it. Since RAC is meaningless to you (though not to me) let's pick another average to look at. My daily average for each week has always been within 10-15% of my RAC so let's look at it. Oh my goodness, my daily average for the week is dropping at about the same rate as my RAC. Okay, let's look at my daily production numbers. Hmmm, daily production has dropped much faster than my RAC or my daily average. Either way you look at it my claim is the same. |
Saimek Send message Joined: 25 Jan 00 Posts: 121 Credit: 454,423 RAC: 0 |
This new Enhanced is strange... i had trux calibrating client + optimized seti client before... for a 33 mins of work i was claiming about 32 credits so about 1 CREDIT per minute... it was calibrating client so i was claiming about as much as i deserve... now for 2:20 so about 140 minutes i am claiming 64 credits... so not even 0.5 credit / minute.... 2 times less than before.... where the problem? calibrating client was made on the fair calculations... |
Jord Send message Joined: 9 Jun 99 Posts: 15184 Credit: 4,362,181 RAC: 3 |
where the problem? Go read how Enhanced is calculating the credits here. It's no longer based on benchmarks times run time of result, but on the actual floating point operations per second that your CPU does. And thus, a calibrating client is of no use, as the benchmarks are no longer used. |
Keck_Komputers Send message Joined: 4 Jul 99 Posts: 1575 Credit: 4,152,111 RAC: 1 |
This new Enhanced is strange... i had trux calibrating client + optimized seti client before... There is no problem. The calibrating client and all the rest of the "optimised" clients are mostly designed to claim unfair amounts of credit per amount of time, they might claim an appropriate amount per task though. BOINC WIKI BOINCing since 2002/12/8 |
Idefix Send message Joined: 7 Sep 99 Posts: 154 Credit: 482,193 RAC: 0 |
Hi, My daily average for each week has always been within 10-15% of my RAC so let's look at it. Oh my goodness, my daily average for the week is dropping at about the same rate as my RAC. Okay, let's look at my daily production numbers. Hmmm, daily production has dropped much faster than my RAC or my daily average. Either way you look at it my claim is the same. My daily average, my RAC, my daily production numbers, my claim, my, my, my, ... Again: It's absolutely irrelevant what you get. You are a tiny little piece of the whole picture. Why do you think this tiny little piece represents the whole project? If you look at the whole picture the credits per day haven't changed. Only this is relevant. Regards, Carsten |
Xaak Send message Joined: 22 May 99 Posts: 32 Credit: 22,636,357 RAC: 0 |
The only valid comparison of granted credits vs time spent crunching is using unoptimized clients on all projects. If that's roughly even, averaged across different processors, then the playing field is as level as possible. If optimizing allows you to increase your credits:time ratio and the project allows this, then that's a bonus. Einstein allows the optimizations and so does seti enhanced. So you're still free to take advantage of the bonus with both. However, as the stock applications improve, the advantage of optimizing becomes less, so your bonus goes down. If the optimized apps get better and we gain more speed improvements, the the bonus for using them will go up, and if the stock application gets faster while the optimized clients stay roughly the same, the bonus goes down. XaaK |
Ingleside Send message Joined: 4 Feb 03 Posts: 1546 Credit: 15,832,022 RAC: 13 |
That does not answer my question. If you get 30 credits for 45 minutes work on Einstein and way less than that here, then how has enhanced made the credit system of Seti equal to that of other projects? Weren't we told that was the purpose of changing the way credits were granted? Sounds to me like they should have gone up and not down. The change by SETI@Home has made the credit-system roughly equal for the 94% that is not running optimized Einstein@home-application... There does the 94% come from? According to BoincStats last stats-update, Einstein@home is responsible for 11.71% of daily production, if you expect half of this is from users that is not running optimized application, it means 94.145% of BOINC's total daily production is not from someone running optimized Einstein@home. Anyway, isn't optimized Einstein@home at the moment giving an even bigger speed-up than optimized "old" v4.18 SETI@Home was giving? If so, the moment Einstein@home releases "official" optimized applications, or switches to "flops-counting", the drop in granted credit/hour for the optimizers will likely be even bigger in Einstein@home than the drop here in SETI@Home... So bottom line is, short-term Einstein@home will give an advantage over other projects, but "soon" this advantage will be removed, meaning users will get roughly the same credit/hour regardless of BOINC-project. |
SargeD@SETI.USA Send message Joined: 24 Nov 02 Posts: 957 Credit: 3,848,754 RAC: 0 |
Hi, Why can no one here respond without saying how meaningless it is or how meaningless I am. Each person who takes the time, equipment and money to support the project is important to the project. Yet there are many on this board who seem to think otherwise. You guys are NOT the project, nor do you speak officially for the project, so you and your rhetoric are meaningless to me. |
Geek@Play Send message Joined: 31 Jul 01 Posts: 2467 Credit: 86,146,931 RAC: 0 |
Hi, Sarge.........Just looking at your stats here I see that you have moved up 50 places in the last month, 11 in the last week and 2 in the last day. I don't see the justification of your claim! Boinc....Boinc....Boinc....Boinc.... |
jamin Send message Joined: 21 Mar 06 Posts: 65 Credit: 100,008 RAC: 0 |
IMHO it is unfair :) 1. Taking into account the pre-enhanced scores - the rules (ex. RAC) have changed but previous scores were not cancelled. 2. The credits amount per day has changed making the whole competition (for those who care) not transparent. But who cares - as long as we doing it for science sake :> |
ML1 Send message Joined: 25 Nov 01 Posts: 21225 Credit: 7,508,002 RAC: 20 |
[...] Great care has been taken to attempt to keep the s@h-enhanced scoring the same as was awarded for the standard s@h application. Ofcourse this will be less than was achieved for those that followed and used the improvements offered by all the optimised clients. Those optimisations are now incorporated into s@h-enhanced so that everyone takes advantage of those optimisations for generating more science. To keep everything fair, the credits have been rationalised to what they were before the x6 optimised clients inflated the credits rates by x6 of what most other participants (unoptimised) were being awarded. I consider this to be very fair for everyone. But who cares - as long as we doing it for science sake :> Well, a lot of people do seem to care. My priority is very much for the Science and for the fun of doing this. Happy crunchin', Martin See new freedom: Mageia Linux Take a look for yourself: Linux Format The Future is what We all make IT (GPLv3) |
SargeD@SETI.USA Send message Joined: 24 Nov 02 Posts: 957 Credit: 3,848,754 RAC: 0 |
That is based on the old system. I just changed my machines over to enhanced on Friday. My daily credit has dropped from almost 10,000 to just a little under 7000 in 3 days. That is a 30% drop no matter how you look at it. (Notice I did not use RAC since it is meaningless). The drop would probably have been more except I still have some old WUs working on a few machines and several in pending status. At this rate I expect to see those numbers you mention going the other way in about a week. |
1mp0£173 Send message Joined: 3 Apr 99 Posts: 8423 Credit: 356,897 RAC: 0 |
Everybody observed the drop of the RAC. If the RAC really is as important as some of you want that it is why don't the total credits of the project show any evidence? Remember that you were running optimized apps, and now you are not. Because the optimized apps were something like six times faster, then yes, your production credits have dropped. That will be true until such time as there are new, faster optimized Enhanced applications. No one denies that the advantage you received from optimized applications (and I did not say "cheat" because it was not cheating) is gone. It may come back when the optimizers have had time to study and adjust the Enhanced application. |
Julian Send message Joined: 19 Jul 99 Posts: 75 Credit: 2,470,428 RAC: 0 |
i am probably being stupid but how much difference are you all finding the enhanced is making. for example we have some new P4 2.8 machines. Old version crunch time with crunch 3R about 40 minutes, credits claimed about 8. Enhanced (but not yet optimized) crunch time about 40 hours credits claimed 40. We are a school, not in it for the credits but for the science (SETI is on the curriculum) However as I am interested in the science it means I am interested in getting units crunched. So I want as many SETI units crunched as possible, how many people in it for the glory will just dump seti and move to another project. Not that this is necessarily a bad thing all the projects have great science in them. I think this will rumble on and the solution is not obvious |
SargeD@SETI.USA Send message Joined: 24 Nov 02 Posts: 957 Credit: 3,848,754 RAC: 0 |
The unoptimized folks who were in a quorum got the same credits as everyone else in the quorum. If they were awarded low credits, then even the optimized folks in that quorum got low credit. If the credits awarded were higher then the unoptimized folks got the higher credits as well. That was the equalizer. So whether you were optimized or not is irrelevant. And please tell me where you came up with X6? I never saw that kind of increase in the entire time I ran optimized. Most of my awarded credits were in the neighborhood of 23-30. Are you insinuating that granted credits should have been only 4-5? Even before optimizing I was getting 12-14 on my fast machines not 4-5. |
Geek@Play Send message Joined: 31 Jul 01 Posts: 2467 Credit: 86,146,931 RAC: 0 |
Sarge........... I'm sure you remember the happy day you discovered "optimized" seti clients and installed them. Do you not remember that the number of work units you were able to crunch each day was increased by a factor of 5 or 6 because of the reduced time required to crunch each work unit? THAT is the advantage you received by moving to the optimized app. The credits received were nearly the same as you have pointed out for each work unit but you suddenly were crunching 5 to 6 times more work units. That advantage you previously enjoyed no longer exists, well not to the same degree anyway! Now the standard Seti app is itself optimized and the available optimized apps do not have the 5 to 6 times advantage as before. [edit]I also got caught in this and have seem my RAC drop from 6500 to it's present number. Now it's a level playing field and that is good! Also a deeper search is now done on the data, also good![/edit] Boinc....Boinc....Boinc....Boinc.... |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.