Any efforts to address benchmarking disparities are greatly appreciated.

As to the "cheater" penalty, from the wu's I have looked at, it seems to be working as intended. Still don't think it was necessary and believe it sent the wrong message, but if you felt you had to act, fine.

I remain bewildered why you messed with the rest of quorum system. I have looked at a lot of wu's and am glad that the changes have had as little effect as they have. I commend the programmer/analysts for doing a good job given their directive. But what was the point?

Some quorums score a little less, some the same, some more. Unfortunately, it seems more less, so there appears to be a small drop-off in points. Which I guess is ok from a relative standpoint because everyone in the quorum drops off together.

The problem, though, is that now, instead of continuing to build momentum adding machines to the cause, people are wondering why their points are dropping.

Now, before I get barraged by the "points don't matter" crowd, I'd like to say that they do. Not for their own sake, and not only because they serve as competitive motivation, but becuase points are the closest thing we have to measure science contributed.

Run time doesn't acknowledge science produced. Results returned depends on the size of the wu's. Points are the only measure we have to gauge how much science we're contributing on a daily basis.

So when the point system is changed in response to wild accusations from an irrational few, and dedicated, serious crunchers feel negatively impacted by the changes, it can't help but disturb momentum and erode confidence concerning the priorities of the project leaders.

Rob725,

There are a few points that I would like to share with you:

1) 5% of results being returned were claiming 2-3 times higher. After looking at the results, these computers were not doing work any faster then other computers - they were simply claiming more points. The existing variation in the 'stock' benchmarks is no where near this broad thus the points they are claiming does not represent more work being accomplished.

It was important to address this for exactly the reasons you mentioned i.e. people do enjoy the competition but it is only a fun competition if it is as fair as possible. Also - if people use the points to measure how much science they are doing and comparing their contributions to others, then the only way to make that a valid competition is to do the best to keep the field level.

2) These users who claimed high benefited in two ways. The most dramatic is when two people claiming high return results for the same workunit. This allowed them to get pretty close to what they claimed.

The second way that they benefited was that by making sure they were also the high claim, then the credits granted would sometimes be higher. Let me use an example to explain. Under the old system lets say there were three computers who returned 50, 45 and 43. The credit granted would be 45. However, lets say that the user who had the 45 had instead used a different client and claimed 102. Then the credit granted would have been 50 - a boost of 11.1% . Under the new system the granted credit would be the average of 50+43 or 46.5. This further eliminates the advantage of claiming high as well as it takes more values into consideration when determining granted credit.

That's a point I'm trying to research right now. It's very complicated...

KeithHenry,

u need to qualify the "somewhat less points for HDC". I run the HDC 1/5 to 1/2 of the time than FAAH. If you're talking per hour.... i have the exact opposite, but it's purely because the partners in the quorum are generally higher claimers.... the averaging is nicely smoothing it out now.

cheers

WCG has always had a policy of only supporting the official client. This has been clarified on numerous occasions.

In particular, the specific example of using optimised clients for the sole purpose of inflating scores has been thoroughly discredited. If you were in any doubt that this is regarded as cheating, then you must have been deliberately ignoring the discussion on the subject.

If that wasn't enough, the XS team got an official ruling on the question, and changed their team recommendation as a result.

As you can see, the penalty system is wonderfully effective at bringing the problem to your attention, and forcing you to take action.

Nobody is blaming you for using the optimised files (unless you were using them in an attempt to cheat) but I hope you can understand that people doing this are causing problems with the credit system. WCG have found a solution.