Message boards :
Number crunching :
Observation of CreditNew Impact
Message board moderation
Previous · 1 . . . 3 · 4 · 5 · 6 · 7 · 8 · 9 . . . 15 · Next
Author | Message |
---|---|
kittyman Send message Joined: 9 Jul 00 Posts: 51477 Credit: 1,018,363,574 RAC: 1,004 |
Look, folks. Here's why I have quit bitching about the current status of credits. As I suggest you all do too. The only impact of the current situation is when comparing Seti against other projects. Not very much within the ranks of Seti. Take a look at the top participants rankings. Not much has changed. I am now at number 3 instead of number 2. And that is due to a newcomer named jmi. Don't know where he came from or what he is running, but it is either a monster or two....or a gaggle of school computers such as mikeej42 is running. Anyways..... The point is, my position against those below me has not changed much. Deltoid is still about 50k below me in the number 4 spot. Our RACs have dropped almost in lockstep since the rollout. So, position-wise within the project, I have not lost much at all. Now, if you are worried about your credits or RAC against your buddy who has 10 times the RAC running a certain candy bar named project, I cannot help you there. But, what is crunched in Seti stays in Seti. And within this project, all things are really pretty much the same. That's why I locked my 'What has your RAC done since' thread. It was pointless to continue to discuss the obvious. I don't happen to like it, for various reasons previously stated, but it's all just eyewash when it comes right down to it. "Time is simply the mechanism that keeps everything from happening all at once." |
W-K 666 Send message Joined: 18 May 99 Posts: 19354 Credit: 40,757,560 RAC: 67 |
The only way I know of working out the number of tasks processed in a time period, is the status page line. Results received in last hour 38,542 4,404 I don't think you can just multiply this by 24 as it is probably at a low just now, as the largest number of crunchers is in the USA, and all the daytime and evening only computers over there, will now be off. @mark. As we know there can be nothing wrong with "creditnew", so if credits are falling then there has to be a reason why, along with Seti's position falling in BOINCworld. What would an outsider think? |
kittyman Send message Joined: 9 Jul 00 Posts: 51477 Credit: 1,018,363,574 RAC: 1,004 |
The only way I know of working out the number of tasks processed in a time period, is the status page line. I did not say there was nothing wrong with CreditScrewed. I just said I am not fretting about it anymore. And I also did post at one point that my only fear about the current situation is what other participants and potential participants would consider when choosing a project. Yes, I DO consider it to be a problem as far as Seti's position in Boinc compared to other projects. It may sway some people's decisions whether to join this project, or to continue with this project. I just know it's no longer a personal concern of mine. "Time is simply the mechanism that keeps everything from happening all at once." |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13841 Credit: 208,696,464 RAC: 304 |
The only impact of the current situation is when comparing Seti against other projects. And that is the whole point of BOINC & granting credit- not just so people get an idea of the amount of work they have done, but so that between projects the recogition of work done is comparable. Because while people say how much they're in it for the science & credit isn't important, if you have a look at the rest of the world you'll find that small & meaningless things are extremely important to many people. And by having credit parity between projects, should distributed computing become trendy, it will stop people that are obsesed by small & meaningless things from going from one project to another in search of greater returns- and often bring those projects down in the process just from the demand on their even more limited than Seti's resourses. So while i have no intention of crunching for other projects, i do consider it important that the awarding of credit is sorted out so that it is fair for all, not just some. Grant Darwin NT |
W-K 666 Send message Joined: 18 May 99 Posts: 19354 Credit: 40,757,560 RAC: 67 |
@mark. As we know there can be nothing wrong with "creditnew", Didn't you notice the sarcasm, I thought my position on a lot of things within BOINC was known and it is not favourable, and CreditNew is one thing that needs to be replaced ASAP. |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13841 Credit: 208,696,464 RAC: 304 |
and CreditNew is one thing that needs to be replaced ASAP. Or at least fixed so that it works as it should. Grant Darwin NT |
juan BFP Send message Joined: 16 Mar 07 Posts: 9786 Credit: 572,710,851 RAC: 3,799 |
The only way I know of working out the number of tasks processed in a time period, is the status page line. I belive in some place this number are avaiable, that could realy give us a way to look the real project production. Or maybe some one knows where to find the number of TFlops, anything that shows not the credit, the real production would be usefull. On other hand i agree with Mark, spend more time on creditnew issues is realy a waste of time if Dr. A does not change his position, nothing will change. That post a question, what is more likely to happens, find an ET signal or change DR.A position about Creditnew? I strongly belive the answer is the first one. I realy belive we could loose a lot of volunteers to other projects because the credit issue, as we loose few months ago with the bandwidth probems, but we can´t do nothing to avoid that. |
Sleepy Send message Joined: 21 May 99 Posts: 219 Credit: 98,947,784 RAC: 28,360 |
Look, folks. Dear Mike, sorry to repeat myself again also here. But I am afraid I cannot agree with you. It is not only about comparing different projects (which is however important. CreditNew was started right for this and is failing right in this. Funny, isn't it?). To be clearer, let's make a personal example. I upgraded (moderately, no monster cards) my rigs. I was thinking to reach some positions, which were before unreachable to me. It was not because my RAC was not enough, but only because at this rate it would become longer than my life. Upgrading those time got at least partially compatible for an average lifetime ;-). Now, with all RACs being roughly halfed, I am back at square one. You are at rank 2-3, therefore one of the major beneficiaries of this situation. If number 4 ever thought of catching you up, now I think he lost any will to do that. While this is not really one of the most important things in life, like any game of sort is, I cannot help saying that what is happening is not correct and that it is not true that we are in all the same boat. People at the top of the ranks are helped by the situation. People below trying to catchup and overpass some of the people above are hurt by this situation. And this is limited to SETI. This is the reason why this time I do not agree with you (I silently do almost always). If you transferred to real society what has happened, this would have tremendous impacts on social stability. You would not talk about that everybody is compared in the same way (try to cut everybody's wage in half... All the same... But all really correct and with no consequences?) Then it is all a game and back to work. But here we try to make it seem serious staff. ;-) :-) Greetings to the kitties! ;-) ;-) :-) :-) Sleepy |
Rolf Send message Joined: 16 Jun 09 Posts: 114 Credit: 7,817,146 RAC: 0 |
So while i have no intention of crunching for other projects, i do consider it important that the awarding of credit is sorted out so that it is fair for all, not just some. And calculation of credit should be very well documented and trustworthy! |
kittyman Send message Joined: 9 Jul 00 Posts: 51477 Credit: 1,018,363,574 RAC: 1,004 |
Look, folks. This is a science project. It's success as such has at times made it a poster child for being a social project as well. There are no wages here, we are all volunteers. I am, you are, all others here are as well. I have said enough about the credit situation. Either things will right themselves or they shall stay as they are. Dr. Anderson's algorithms for credit granting have fallen flat on their arse. I doubt that he shall rewrite them now, too much egg on face to do so. Best to stand pat, eh? When an honest man trips and falls, the appropriate thing to do is to stand back up, admit the failure, and then don't do it again. I'm afraid some folks never learn that lesson. "Time is simply the mechanism that keeps everything from happening all at once." |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14676 Credit: 200,643,578 RAC: 874 |
The only way I know of working out the number of tasks processed in a time period, is the status page line. You can see graphs of the 'Results received per hour', and some averages, at http://setistats.haveland.com/sah_results.html. There was definitely a drop when SETI v7 was rolled out, but that was expected (even desired): the new app does more work and takes longer - and there was also the gap before the new optimised apps were generally available. And judging by the download numbers, there are probably a (large) number of users of the old v6 Lunatics apps, who haven't noticed that they need to update yet - they will be doing no measurable work at all. The TFlops question is an interesting one. So far as I know, projects have no reliable, absolute, way of aggregating the power of their volunteers' computers. As has been said in this thread, we don't record the flopcounter any more - and that was a pretty arbitrary surrogate, anyway. You will see TFlops - sometimes even PetaFlops - on most projects' server status pages, and on the statistics sites. But take it with a very large pinch of salt. From what I understand, the published TFlops figures are all reverse-engineered from the credit awarded - and calculated at the nominal defined rate of credit awarded, unchanged (except for a bug-fix) since BOINC started. So, according to the public record, the number of TFlops contributed by SETI volunteers will have reduced over the last three weeks, because of the reduction of the credit awards. Conversely, my least-favourite work of fiction is Top 100 multi-project BOINC participants. Could somebody with a thorough working knowledge of modern silicon fabrication technology cast an engineer's eye over those individual GFlop claims, please, and see what you make of them? |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13841 Credit: 208,696,464 RAC: 304 |
So Flops is the new MIPS... Grant Darwin NT |
Sirius B Send message Joined: 26 Dec 00 Posts: 24909 Credit: 3,081,182 RAC: 7 |
Jeez, a year or two ago, I had all this on a thread titled "Boinc's Death Knell" & immediately, some individuals attacked me for attacking Dr A. They failed to read, which for types like that is nothing new. It's been mentioned here already that "others" chase other meaningless projects to score larger credits. WTF? are they on, Ganga or heavier stuff? I'm a WWII buff so Enigma is a worthwhile project for me. Malaria is a killer disease, so is cancer & other medical projects are also worthwhile. So putting those into context then, Seti is an absolutely worthless project! The "race" is on to find E.T with many like Mark racing ahead - kudos to them as they have the opportunity to do so. The rest of us poodle along in the slow lane & the difference between them all is? Quit the bitching & raise valid issues like getting the coding correct (as near to 100% as humanly possible that is) & roll the dice! |
MikeN Send message Joined: 24 Jan 11 Posts: 319 Credit: 64,719,409 RAC: 85 |
I'm a WWII buff so Enigma is a worthwhile project for me. Malaria is a killer disease, so is cancer & other medical projects are also worthwhile. So putting those into context then, Seti is an absolutely worthless project! Except if we find ET and they are more advanced than us, then they might be able to tell us how to cure cancer,malaria and other known diseases and save us billions in research every year! |
Ingleside Send message Joined: 4 Feb 03 Posts: 1546 Credit: 15,832,022 RAC: 13 |
If this is a permanent impact to RAC, so be it. One way to think about this is users has gotten a bonus for continuing to crunch during all the years with maxed-out bandwidth and very frequent server-problems, and this bonus is now removed since the problems are (hopefully) fixed... As for CreditNew, where's multiple possibilities: 1: Things haven't stabilized yet, it just goes very slowly. 2: Things has already stabilized, this is the intended behaviour. 3: Where's a bug what needs fixing. Due to the complexity of CreditNew, I've no idea which is the current status. "I make so many mistakes. But then just think of all the mistakes I don't make, although I might." |
Sirius B Send message Joined: 26 Dec 00 Posts: 24909 Credit: 3,081,182 RAC: 7 |
I'm a WWII buff so Enigma is a worthwhile project for me. Malaria is a killer disease, so is cancer & other medical projects are also worthwhile. So putting those into context then, Seti is an absolutely worthless project! Exactly! But until that happens, other projects are just as worthwhile - the odds are that those projects will find solutions before we find E.T. or he finds us. All I've seen since Boinc took over from classic is a lot of bitching because changes are affecting positions on a worthless "leaderboard". |
William Send message Joined: 14 Feb 13 Posts: 2037 Credit: 17,689,662 RAC: 0 |
If this is a permanent impact to RAC, so be it. One way to think about this is users has gotten a bonus for continuing to crunch during all the years with maxed-out bandwidth and very frequent server-problems, and this bonus is now removed since the problems are (hopefully) fixed... Doesn't look like RAC has stabilised yet. Maybe in a month or two. We need all the small CPUs reporting in quite a few tasks... 2: Things has already stabilized, this is the intended behaviour. Less credit wasn't intended - more like a collateral. 3: Where's a bug what needs fixing. CreditNew in itself is a bug ;) Due to the complexity of CreditNew, I've no idea which is the current status. As complex as a large ball of wool that has been chased by a kitty across the house. You keep tripping and God help you if you're trying to find where the thread is worn. Personally, I don't care about credit. Looking at BoincStats, the rig used to pull around 7.2k with V6 - it now sits around 2k. That's less than a third. It runs GPU only for SETI. Anybody care to dig out a CPU only machine and compare? A person who won't read has no advantage over one who can't read. (Mark Twain) |
QSilver Send message Joined: 26 May 99 Posts: 232 Credit: 6,452,764 RAC: 0 |
I'm a WWII buff so Enigma is a worthwhile project for me. Malaria is a killer disease, so is cancer & other medical projects are also worthwhile. So putting those into context then, Seti is an absolutely worthless project! But maybe ET will do all that for us only because they want To Serve Man. |
juan BFP Send message Joined: 16 Mar 07 Posts: 9786 Credit: 572,710,851 RAC: 3,799 |
You can see graphs of the 'Results received per hour', and some averages, at http://setistats.haveland.com/sah_results.html. There was definitely a drop when SETI v7 was rolled out, but that was expected (even desired): the new app does more work and takes longer - and there was also the gap before the new optimised apps were generally available. And judging by the download numbers, there are probably a (large) number of users of the old v6 Lunatics apps, who haven't noticed that they need to update yet - they will be doing no measurable work at all. I agree this graphs show a lot less returned WU per day something realy not expected since the new optimized apps are expected to be faster than the old stock apps, even with correlation, since a lot users are now running optimized apps than old stock, nothing related to credit of course. Or a large number of hosts are not updated to V7 or we loose a lot of crunchers. Both hypotesis are not good The TFlops question is an interesting one. So far as I know, projects have no reliable, absolute, way of aggregating the power of their volunteers' computers. As has been said in this thread, we don't record the flopcounter any more - and that was a pretty arbitrary surrogate, anyway. Sorry this time i could´t agree, the most realible way to realy measure a computer power as the Flop mark, Mega, Tera, Phenta, whatever. Even the supercomputers have a way to measure it. You allways see. That SC have the capacity of xxx TFlops, etc. That number could not be come from a credit reverse engenhering, it´s comes form the hardware itself, the sum of the capacities of all conected clusters at any given time. Compared against a reference. Maybe by count all flopscount on each WU returned on a given period of time, i realy don´t have the knowledge on that area, but certainly we have someone in the comunity who have the skills to show us how. . ...Conversely, my least-favourite work of fiction is Top 100 multi-project BOINC participants. Could somebody with a thorough working knowledge of modern silicon fabrication technology cast an engineer's eye over those individual GFlop claims, please, and see what you make of them? As you could see no one of the 100 participants run´s SETI, that can´t be a coincidence... everyone knows "if you want a BOINC high RAC do not run SETI"... and now with the creditnew mess that afirmative is more stong than ever. We all run SETI for a dream, help to find our little green friend, not for credit. As Mark post before we all need to stop complain about creditnew, is a lost of time. In the begining i belive we all was trying to find why the RAC was falling and if was something at our side or related to the new builds, now we all know that´s not thecase and know can´t do nothing about. We just need to adapt to live with this "small credit paid per WU" nothing impossible since we all are in the same sinking boat. |
Dorphas Send message Joined: 16 May 99 Posts: 118 Credit: 8,007,247 RAC: 0 |
i wish the powers to be would work half as hard on showing us science results of our work than spending all their time tweaking damn credits. it's about time they start showing us the results of our work. then maybe people can talk about what we found in our work instead of credits. "it's all about the science".....not until they start showing us this "science". |
©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.