Message boards :
Number crunching :
Seti Enhanced Credit Fair?
Message board moderation
Previous · 1 . . . 11 · 12 · 13 · 14 · 15 · 16 · 17 . . . 23 · Next
Author | Message |
---|---|
SargeD@SETI.USA Send message Joined: 24 Nov 02 Posts: 957 Credit: 3,848,754 RAC: 0 |
YOU cannot see other options. But YOU are not the deciding official and maybe they will see and understand where we are coming from. Most of us WANT to crunch for Seti and not some other project. However, we are being pushed into a corner where many of us may decide to either switch or just quit crunching entirely. I have been with Seti since the early days, though I lost the password to my original account and had to make a new one. I crunched during the time when my machine might turn out 1 WU in 24 hours. I have no real desire to leave Seti for another project but the way things are headed now I may have to make that decision. |
1mp0£173 Send message Joined: 3 Apr 99 Posts: 8423 Credit: 356,897 RAC: 0 |
Just out of interest why do you need the return immediately option? I have the same need, but the servers are mission-critical, so I do not run anything on them that isn't strictly needed. If they break, I don't eat. I use Servers Alive for monitoring, it doesn't depend on the speed of a work unit. I've also found UltraVNC to be quite handy. |
W-K 666 Send message Joined: 18 May 99 Posts: 19313 Credit: 40,757,560 RAC: 67 |
If you continue crunching for Seti you will on average get the same credits/time as any other project, using enhanced. If people are not happy with this situation, then it would probably be better for them to leave the BOINC community. I don't think anybody should feel they are being pushed into a corner, why should they feel that way, just because ever since BOINC/Seti started the biggest, in capitals, has been the difference in claimed credits for different cpu's, OS's etc. etc. The main reason for this was the reliance on the benchmark figure for nearly everything important to us the users of the software. JM7 has fixed a lot of the promlems with his scheduler, you now get the correct amount of work in your cache, it ensures units don't miss deadlines, as long as the user leaves it alone. The most important thing to be fixed, in the users mind's was the claimed credits, the Fpops calculation does that. Because the number of Fpops depends on the AR and that also decides what analysis is done or not done it makes the curve like a badly designed multiple ski jump. It has been noticed that the time taken and the credits claimed may not be as accurate as would be liked, but a least some of the developers know about it and I hope they have enough influence to get it adjusted. There is no way, that Seti can go alone and offically base the credits on what users could claim using optimised apps and clients. They have to be approx equal to those granted on all the other projects. And on average I think they have achieved that. And your comment on 24 hrs to crunch a unit is not needed try this computer result on Beta Beta resultid=779892if you want time to crunch a unit. To all the nay sayers, next time there is an invitation to do a trail, come and join us there, your concerns will be listened to and if relevant will be acted upon before release into the wild. On the enhanced trial if credits is your thing you could have had a period of getting over 40/hour on a Pent M 1.86, without any apps or client being optimised. Andy |
Geek@Play Send message Joined: 31 Jul 01 Posts: 2467 Credit: 86,146,931 RAC: 0 |
Well Said WinterKnight!! You get a + from me! Boinc....Boinc....Boinc....Boinc.... |
nairb Send message Joined: 18 Mar 03 Posts: 201 Credit: 5,447,501 RAC: 5 |
If E.T had been found by a cruncher using an optimised application. Would he have still been called a cheat. |
eaglescouter Send message Joined: 28 Dec 02 Posts: 162 Credit: 42,012,553 RAC: 0 |
It is sad to see people leaving the SETI Project over this issue of credits, yet I fully understand and support their decision. I began crunching simply to test the results of my overclocking project, then I got involved with a team race and bought more boxes. As I sit in my computer room (90 degrees F with the air conditioner running at full speed) I wonder if anybody at Berkeley understands those of us who have a competitive nature, those of us who spend (significant) amounts of money to acquire these worthless points. My quandry: Should I spend more money to upgrade the air conditiner capacity on my home due to the heat output of the SETI farm, or should I turn off the farm, quit SETI and use the money to take a cruise with my wife? I like SETI, I enjoy the quest for increasing my RAC, I think team races are awesome. I don't mind the utility bill, nor the high room temperatures. Unfortunately it looks like Berkeley does not care about any of the things that I like in this project. If Berkeley is going to make a credit change in the SETI project, then I have one simple request; all I want is a uniform point system that gives the same points per hour regardless of which BOINC project I crunch on a specific machine. Don't make another project more valuable than another. One hour on my Dual Opteron should be worth the same number of points on every BOINC project. If we are doing a wholesale revision of the credit system, then we should probably have a reset just like when we switched from Classic. Thus making points into an apples to apples comparison. As for me, I have never crunched for any other project, just SETI and only SETI. This credit change is demoralizing and is doing great damage to our teams, something must be done to fix this damage. I'm not certain what I will do yet. The cruise sounds fun, but I've crunched for SETI for so long, it would be hard to leave. I guess the decision is in the hands of Berkeley, they can give me my worthless points which cost them nothing, or I can save a ton of money. It's up to you Berkeley, call it extortion if you wish, but my worthless points have far more value to me than you appear to understand. It's not too many computers, it's a lack of circuit breakers for this room. But we can fix it :) |
eaglescouter Send message Joined: 28 Dec 02 Posts: 162 Credit: 42,012,553 RAC: 0 |
When I looked at the beta there were no credits being offered, and since I'm another credit monger I walked away. No credit = no participation If they wanted participation they should have offered me credit equal to my rac for that machine during the beta period. I would have gladly participated. I don't expect much for my contributions, just these silly worthless points. It's not too many computers, it's a lack of circuit breakers for this room. But we can fix it :) |
Jack Gulley Send message Joined: 4 Mar 03 Posts: 423 Credit: 526,566 RAC: 0 |
I have not see the following posted here yet so a valid comparison can be made. Number of FLOP counts for: The Reference Unit ran on the stock 4.18 application that earned 32 credits. A typical Work Unit ran on the stock 4.18 application that earned 24 credits. The same two WU's ran on an optimized application. (Should be the same FLOP.) The same two WU's ran on the Enhanced 5.12 application. (If possible.) Several typical Enhanced WU's ran on the 5.12 Enhanced application earning 62. Then we can compare the actual work performed in each case with the credit granted. |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13835 Credit: 208,696,464 RAC: 304 |
If Berkeley is going to make a credit change in the SETI project, then I have one simple request; all I want is a uniform point system that gives the same points per hour regardless of which BOINC project I crunch on a specific machine. Don't make another project more valuable than another. One hour on my Dual Opteron should be worth the same number of points on every BOINC project. That is exactly what they have just done, and those used to getting huge amounts of credits per hour are now finding that the playing field has been leveled & their optimised applications don't give them the advantage they had before (at least not yet anyway). If we are doing a wholesale revision of the credit system, then we should probably have a reset just like when we switched from Classic. Thus making points into an apples to apples comparison. But it's not a wholesale revision as such, just fixing up something that people have been carrying on about for the last couple of years or so. Of course now it gives others something else to carry on about. This credit change is demoralizing and is doing great damage to our teams, something must be done to fix this damage. Damaging teams? Everyone is doing the same work & so will get the same credit. Those that have faster machines will get more credit per hour. Tose that have optimised clients will get more credit per hour than those that don't. Just not as much as they are used to. It's up to you Berkeley, call it extortion if you wish, but my worthless points have far more value to me than you appear to understand. Which would make me think you really need that cruise. Take a break & get some perspective on life. Grant Darwin NT |
W-K 666 Send message Joined: 18 May 99 Posts: 19313 Credit: 40,757,560 RAC: 67 |
It is sad to see people leaving the SETI Project over this issue of credits, yet I fully understand and support their decision. Isn't this para, at odds with everything else you said, The Fpops method has been calibrated to claim and grant approximately equal credits with all other projects. The other projects it must be pointed out set their credit calculations to be approximately equal to Seti as Seti was the first. If Seti was to grant more credits/time as most of the nay sayers seem to want then the other projects would be totally free to grant any credits they wanted, probably leapfrogging Seti and in the end we would be demanding 1 * 10^n in advance before we would even download a unit. |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13835 Credit: 208,696,464 RAC: 304 |
This issue is not going to just blow over, regardless of how much anyone wishes it would. Yes it will. People will stay or people will leave, life will go on. Grant Darwin NT |
Grant (SSSF) Send message Joined: 19 Aug 99 Posts: 13835 Credit: 208,696,464 RAC: 304 |
Most of us WANT to crunch for Seti and not some other project. Then it's simple- keep crunching. I have no real desire to leave Seti for another project but the way things are headed now I may have to make that decision. If you don't want to leave then don't. If you want to, then do so. It's up to you. Grant Darwin NT |
Jord Send message Joined: 9 Jun 99 Posts: 15184 Credit: 4,362,181 RAC: 3 |
This issue is not going to just blow over, regardless of how much anyone wishes it would. Someone want a link to some threads where people demanded that Classic was kept on running? Or in the least that the "New Seti" followed the "crediting" as per Classic? |
thinksnow Send message Joined: 26 Dec 02 Posts: 41 Credit: 505,798 RAC: 0 |
Maybe I'm missing something, but aren't people using the optimized clients still getting through the WU's faster? And if they get through the units faster, then that means they crunch more per day. If tey crunch more per day, then they get more credit per day, yeah? If you're still getting more credit per day using an optimized client than you would be using the standard core, then what is all the gnashing of teeth and cry-babying about? As soon as everyones queues are depleted, there will be only enhanced units to work on and those who crunch them faster will still get more credit than those who are slower and they'll *still* get to brag about their massive crunching ability (and teh science will still get done). ...or am I missing something? Main rig: AMD Opteron 165 Dual Core, 2Gb PC3200, 450Gb onboard + 1Tb RAID-5 NAS HTPC1: P4 2.8E, 1.5Gb PC3200, 3x80Gb SATA, Fusion5Lite + PVR-250, Sony 50A10 50" HD-LCD HTPC2: P4 2.8E, 1GB PC3200, 200Gb PATA, PVR-250, JVC 32" tube |
Jord Send message Joined: 9 Jun 99 Posts: 15184 Credit: 4,362,181 RAC: 3 |
Maybe I'm missing something, but aren't people using the optimized clients still getting through the WU's faster? And if they get through the units faster, then that means they crunch more per day. If tey crunch more per day, then they get more credit per day, yeah? Yes, but even with an SE result, on an optimized client, it can still vary between 45 minutes and 17 hours. Depending on AR of the result. And that is something the credit mongers don't like. :) They ran their hundreds of results per day in an almost standard time of less than 1 to 2 hours. Yet SE, even if optimized, doesn't do that anymore. And it doesn't credit as they "are used to". So they leave here and go to other places where they can still crunch 75 results a day, not just 2. I wonder why they aren't all crunching HashClash. 10,000 results/day quota. All running in 2 minutes or less. Must be fun. |
Crunch3r Send message Joined: 15 Apr 99 Posts: 1546 Credit: 3,438,823 RAC: 0 |
Thats BS.. it's not that WU's take longer to crunch nor that it's now based on fpop counting... it's about credit has been cut down. Join BOINC United now! |
1mp0£173 Send message Joined: 3 Apr 99 Posts: 8423 Credit: 356,897 RAC: 0 |
|
Jord Send message Joined: 9 Jun 99 Posts: 15184 Credit: 4,362,181 RAC: 3 |
Read the threads, dear Crunch3r. Most of the people using the optimized clients say they don't want to run long results. Then there's the couple who say that the credits aren't right. And there are only a minimum amount of people who say that the results run for too long and they don't get the credit they are used to. I know you were asked why you'd released an optimized client at the start of SE being released. I know you said because you promised it. Maybe that for a next time you can promise to release a client when all the bugs have been taken out. Yet even your client can't reduce an SE VLAR running time to 2 hours. You might also want to explain why this is to people. |
Crunch3r Send message Joined: 15 Apr 99 Posts: 1546 Credit: 3,438,823 RAC: 0 |
Well the enhanced app. should not have been released in the first place because its to buggy... read the threads i stated that not only once ;) Anyway regarding beeing a stable app. well the most stable is still the optimized app... Join BOINC United now! |
DrBob Send message Joined: 27 Sep 99 Posts: 37 Credit: 10,712,463 RAC: 0 |
Agreed |
©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.