The classic in my eyes worked better and scored better than this "New and Improved" BOINC version.
I ran SAH classic for the science as well as the credit for doing the wu.
This new system with it's "claimed" vs "granted" credits sucks you know what!
I think that the rest of my team feels the same and After trying out the new version using BOINC I don't think we will continue once the classic comes to an end.
To all that continue on, Good luck,
Daemon

Have you really given it a chance?

Why do you think it 'sucks'?
The claimed - granted system means you get paid, in credits, the same as the others who did the unit.
Should you receive full payment for noisy units that finish in several seconds?
Should you receive more or less credits for doing exactly the same unit?
Why not try the optimized clients from Maverick - Here .
With these you can cut your proccessing time/unit and therefore do more units and on average and still get the average 28 approx credits granted.
The main problem is the benchmarking, which is being discussed and/or worked on, which governs the predicted time of processing, the number of units you can download, scheduling - if you do more than one project -, and credits.

Andy

WinterKnight - The main problem is the benchmarking, which is being discussed and/or worked on, which governs the predicted time of processing, the number of units you can download, scheduling - if you do more than one project -, and credits.

Your right. Screw the artificial benchmark. This isn't Future Mark and I don't need to know how fast my video card is with it's OPTIMIZED DRIVERS. I do a wu and get 1 wu credit. That's all there should be to it. Yes, sometimes there will be the bad or noisy packet but that is life. I finish a bad one in 5 minutes or a tough one in 3 hours. Call it luck of the draw but to insert some artificial benchmark and base my "granted credit" on it from other systems is what SUCKS. I did the wu now give me 1 credit like the classic did.
Daemon
BTW - I tried the optimized client.

Your right. Screw the artificial benchmark. This isn't Future Mark and I don't need to know how fast my video card is with it's OPTIMIZED DRIVERS. I do a wu and get 1 wu credit. That's all there should be to it. Yes, sometimes there will be the bad or noisy packet but that is life. I finish a bad one in 5 minutes or a tough one in 3 hours. Call it luck of the draw but to insert some artificial benchmark and base my "granted credit" on it from other systems is what SUCKS. I did the wu now give me 1 credit like the classic did.
Daemon
BTW - I tried the optimized client.

and get screwed by all the cheaters?

Andy

It may help to focus on some of the things that will be new for you now that you are running SETI on the BOINC platform. I remember that the stat situation knocked me for a loop at first. Once I found BOINCstats I never looked back.

I'll admit that it still bothers me that a work unit value varies on different machines and on the amount of time they spend processing. I try to remember that people are trying to create a system that will work on work units for other projects and for SETI combined.

---

I think that the rest of my team feels the same and After trying out the new version using BOINC I don't think we will continue once the classic comes to an end.

All two of you?

---

me@rescam.org

[quote]and get screwed by all the cheaters?

Andy

Cheaters are a fact of life. They're found in games to corporations. I'm not in competition with them, just my fellow enthusiats that are playing by the rules. (When the rules are fair)
This is not to be confused with how the benchmarks work and how there is obvious bias on the platform the client is used. This does not create a fair and even field right from the start. Let alone, I don't like getting a seemingly arbitrary credit for the wu. Good or Bad, it's the way I and my fellow team feel about this.

All two of you?

Maybe you should do your research a little better.

Thanks for the info StokeyBob.

I'm not after starting a flame session. It's just that the people running this project need to know... whatever the feedback is.
Daemon

We know what you are going through. BOINC is a big change from the old comfortable Classic SETI.

If you test out BOINCstats enter your name in the upper left. Then after your name is found click on the little graph looking icon way over to the right of your name. That will get you to the graphs of your history.

I just can't get enough graphs.

---

I just can't get enough graphs.

LOL:) If you only knew!!!!

BUT...
Unfortunately that doesn't change the way this is scored/credited.

True, but watching the graphs progress may help pass the time while you ease into things.

---

I finish a bad one in 5 minutes or a tough one in 3 hours. Call it luck of the draw but to insert some artificial benchmark and base my "granted credit" on it from other systems is what SUCKS. I did the wu now give me 1 credit like the classic did.
Daemon
BTW - I tried the optimized client.

let's look at this in reverse. As if I was "forced" to go from Boinc to Classic.

I get 3 credits for a bad one, or 35 credits for a tough on in 3 hours. Call it luck of the draw but to insist I get only one credit regardless of the science done is what sucks. I want my 28 credits like boinc did.

Another wierd exchange. The focus of this project should be on the amount of space we have 'explored', not on the bone-headed credits. Somebody shouldn't leave 'us' because of the credit nonsense (you act as though it is real money). But one should leave if the quality of the science is lacking.

Unfortunately, the fundamental problem here is that we don't know how much science is getting done, just the number of work units being computed. So I think someone should create a metric based on scientific measures. For example, the solid angle of space of each work unit. This would need to be multiplied by some measure of bandwidth. And so on. I'm sure it isn't easy, but it would make more sense than what we are doing now.

---

May this Farce be with You

I did the wu now give me 1 credit like the classic did.

Well, if that's all you want, you should be absolutely ecstatic that you get more than one credit per WU! :)

All two of you?

Maybe you should do your research a little better.

What better research would you suggest?
Team DZ has two members. Fact.
OTH if you're saying that under classic you're a part of much larger team, then I would hardly call just 2 of you trying out BOINC a significant sample.
BOINC is the future, but - as usual - the future follows the present.

---

What better research would you suggest?
Team DZ has two members. Fact.
OTH if you're saying that under classic you're a part of much larger team, then I would hardly call just 2 of you trying out BOINC a significant sample.
BOINC is the future, but - as usual - the future follows the present.

6 members actually, but only 2 have gotten credit.

---

Raithmir's SPARC64/UltraSPARC Linux Builds
http://www.kulthea.net/boinc/

Also, there are more projects than just SETI@Home. For those that want to only do SETI, that is fine ... but BOINC is multiple projects, and within projects multiple applications.

For UCB we are doing only SETI for Arecibo, in the near future we will be doing Astropulse ... and later, we hope (that is, me and the mouse in my pocket), doing data from other telescopes ...

The credit system is intended, though it is not working well for this, to allow us to all measure contribution to science, regardless of exactly which project you are supporting...

---

Your right. Screw the artificial benchmark. This isn't Future Mark and I don't need to know how fast my video card is with it's OPTIMIZED DRIVERS. I do a wu and get 1 wu credit. That's all there should be to it. Yes, sometimes there will be the bad or noisy packet but that is life. I finish a bad one in 5 minutes or a tough one in 3 hours. Call it luck of the draw but to insert some artificial benchmark and base my "granted credit" on it from other systems is what SUCKS. I did the wu now give me 1 credit like the classic did.
Daemon
BTW - I tried the optimized client.

and get screwed by all the cheaters?

Andy

Why some people think that going back to one credit per work unit means, "stop validating" is farther out in space then the ones we hope to discover.

Keep and even improve the validation process and give one credit per validated work unit.
If you think that the work units need to be smaller so you can get more numbers then make them smaller.

All the people that seem to claim not interested in the accuracy of the benchmarks and granted credit seem to object the most to the one credit per work unit method, what does that tell you.

Many of the people in these forums mention work units that finish in 1 - 3 hours boy I wish I had such a powerful computer when mine are 6 - 19 hours except for my G5 2.5 which is 2.5 hours, it is my exception.

Finish a work unit get a credit.
Turn in a bad work unit no credit.

---

Why some people think that going back to one credit per work unit means, "stop validating" is farther out in space then the ones we hope to discover.

Keep and even improve the validation process and give one credit per validated work unit.
If you think that the work units need to be smaller so you can get more numbers then make them smaller.

All the people that seem to claim not interested in the accuracy of the benchmarks and granted credit seem to object the most to the one credit per work unit method, what does that tell you.

Many of the people in these forums mention work units that finish in 1 - 3 hours boy I wish I had such a powerful computer when mine are 6 - 19 hours except for my G5 2.5 which is 2.5 hours, it is my exception.

Finish a work unit get a credit.
Turn in a bad work unit no credit.

OK you're right about the validation, but as Paul has said there will be different crunch time Seti units when the other parts come on line. Do you want 1 credit for two hrs and 1 credit for 20 hrs work on the same computer? It is fairer for x credits/time period no matter what project you are working on and at the moment that means averaging the claimed credits of those that successfuly complete the same unit. What it means for present day Seti Units is that on average you will be awarded about 27 credits. And for Einstein a somewhere in the 80's but they take 3 times longer to complete ( on my old machine).

Andy