Credit shall be granted by the actual work done from the host. In order to achieve that we grant credit based on the number of models (decoys) completed for each work unit. This relates exactly to the actual scientific contribution. For each type of WU we determine a credit/model ratio which we will grant as "work credit". A faster machine which completes more models in the same time will receive more credits.
For each work unit type, we keep track of the total amount of claimed credits and structures from valid results returned by hosts, and we use these running totals to determine the amount of credit to award per structure. So if your computer returns 2 structures, the amount of credit awarded would be 2 * total_claimed_credit / total_structures where total_claimed_credit and total_structures are the sum of the claimed credits and structures from valid results returned by all hosts prior to your returned result for that particular work unit type, respectively. The first returned result will be awarded the claimed credit, the second returned result will get the average claimed credit per structure between the two multiplied by the number of structures returned by the result, the third returned result will get the average claimed credit per structure between the three multiplied by the number of structures returned by the result, and so forth. There is currently a discussion to wait until the first xxx results are returned before establishing the average. This would further reduce the variation among the first returned results.
Cheating is no longer possible (To avoid misunderstanding: using the optimized BOINC client was never considered cheating from the project team, but there were instances of manually faking reported benchmarks). Credits are granted based on the actual work done and faster machines receive more credit than slower ones accordingly to their actual contribution. Reporting falsificated boinc benchmarks doesn't influence the granted credit and there are no differences in granted credits depending on which version of BOINC you are using. There is a level playing field and a fair competition.
There will be some variation among results. You will sometimes receive more credit and sometimes less. It will average out over time but there will be some kind of fluctuation. This is due to the fact that the amount of computation for each model varies even among equal WU. Furthermore we use many different WU and we need to determine for each type of WU a specific credits/model-ratio. This will result in slightly different credits/hour-ratios. Again this will average out over time but it will create some fluctuation in your average work credit.
The "ugly" (controversial)
There is no ugly aspect of the new credit system, ;-)however there are implication which some will praise while others will condemn them. Under the current system your granted credit varied by about 300% whether you were using a standard BOINC client or a custom one (crunch3er, etc.) and it varied considerably whether you were using Linux or Windows. Under the new credit system people who used customized (optimized) BOINC clients such as 5.5.0 will get considerably less credits/hour than before - probably less than half the credit they got before. Users with the standard BOINC client will receive slightly more credits/hour especially if they are using Linux. So there will be "winners" and "losers". However with the new work credit system the amount of credit will only depend on your actual scientific contribution and not your operating system or the version of BOINC you are using (with the exception of a small random variation for early results). As a side note you shall know that the scientific contribution is the same whether you are using the standard BOINC client or an optimized one like 5.5.0.
P.S.: Please don't post in this thread but post discussions about the new credit system here
I am a forum moderator! Am I?