I installed a optimized client just to see the difference.


27065786 This is optimized client

CPU time 42315.59375

Claimed credit 284.981345829876

Granted credit 284.981345829876

27065786

27007905 Boinc regular.

CPU time 42927.53125

Claimed credit 114.911514533627

Granted credit 114.911514533627



This don't seem to right . Its close to same cpu time but 2.47 times the credit.



Is this optimized or cheating???????

---

I installed a optimized client just to see the difference.


27065786 This is optimized client

CPU time 42315.59375

Claimed credit 284.981345829876

Granted credit 284.981345829876

27065786

27007905 Boinc regular.

CPU time 42927.53125

Claimed credit 114.911514533627

Granted credit 114.911514533627



This don't seem to right . Its close to same cpu time but 2.47 times the credit.



Is this optimized or cheating???????

Strictly speaking its cheating. Their is currently no optimized rosetta app and the optimized boinc client claims just more credit but does not increase the speed which with your computer runs Rosetta. It just claims more credit.

However this kind of "cheating" is tolerated since many people run the optimized boinc client for various reasons. Hopefully soon after CASP is over Rosetta will change the credit system to a one which grants credit based on the actual work done and this "cheating" won't longer work.

---

Strictly speaking its cheating.

If you ask the top teams, 95% of the members will tell you it's not.
Most of them run optimized.

If you ask the top teams, 95% of the members will tell you it's not.

They would though, wouldn't they.

---

Wave upon wave of demented avengers march cheerfully out of obscurity into the dream.

Strictly speaking its cheating.

If you ask the top teams, 95% of the members will tell you it's not.
Most of them run optimized.

Reminds me of that t-shirt

"Eat sh*t!

1 Billion flies can't be wrong"

---

I think there's a plan to sort this out following the CASP trial.

I think something needs to be done about this from BOINC though, as there's currently disparity between the projects. Maybe the BOINC team could produce a benchmark modle for all the projects to incorporate into their applications so there's some standardisation again.

---

Before I begin, I should state that I run the standard client, just for the record.

I guess I am not the crazy competitive type.. But..... What's the "oh my devine entity" deal here?

My view is that as long as the work is getting done, and there is no value (Monitary, prizes, etc) for Credits, this should not cause strife on the project.

If in the future, the founder/administrators/people in charge of this project comes out and states "DO NOT USE MODIFIED CLIENTS AS IT IS MESSING UP THE END RESULT OF THE WORK", then anyone that persists in using the modified or "optimized" clients should be told to stop, and if they won't, then more drastic measures taken. I don't think this is the case though. I think the science is not being effected at all, but just the braging rights.

If the founder/administrators/people in charge of this project come out and state "THE MODIFIED CLIENTS DO NOT MESS ANYTHING UP, BUT DON'T USE THEM BECAUSE OF THE CREDIT SYSTEM", then since the science is not being effected, it should be up for debate, with the founder/administrors having the final word.

Take my point of view for what it's worth, (the whole two cents/grain of salt thing)...

---

The best argument for a standard client that is freed from the Boinc benchmarks is that a fair and universal Rosetta credit system will eliminate the problems of Linux getting lesser amounts of credit from the standard Boinc client - and encourage those that are waiting for a system free of the optimized clients before they'll take part in Rosetta.

And then the stats freaks will have to concentrate on getting their systems to run longer, run Rosetta with a larger piece of the Boinc allocation of time, improve the performance of their machines (more ram, faster processor, overclock, etc) and running another system.. :)

---