Folding Forum

Does FahMon support the calc of PPD based on BonusPoints and WU-points, so for exampel a 3GHz 8-core system that fold 24x7 will show aprox 25K PPD?

I don't think FahMon will ever be able to predict what your bonus will be. The estimated completion time is already a guess that requires you to choose one of several different methods because the process gets interrupted and there are way too many things that can delay your results to expect an accurate calculation. You need to treat the bonus as exactly that, a bonus, not part of your inherent PPD.

No Fahmon has no way to tell what the bonus value of a WU is (it's not displayed on psummary or any other source). It will only show PPD based on the point value found in the psummary.

I saw a spreadsheet at overclockers or OCforums a few weeks ago. It said that the bonus was....

Bonus = SQRT(final deadline x constant / WU Time) = sqrt(6 * (2 / wu_time))

Dunno how true that is. This could be added to Fahmon I would image (assuming it is correct)

DriveEuro wrote:I saw a spreadsheet at overclockers or OCforums a few weeks ago. It said that the bonus was....

Bonus = SQRT(final deadline x constant / WU Time) = sqrt(6 * (2 / wu_time))

Dunno how true that is. This could be added to Fahmon I would image (assuming it is correct)

The formula is basically correct -- The "2" and "6" values are not constants they are variables, they could change from time to time based on the WU project. The formula was outined in the messages which announced the new WUs.

The only problem for Fahmon, is that the bonuses are awarded based on the upload time of the WU results, not on when the WU finished. So, Fahmon would need to parse the fahlog details more carefully so see when the WU results where sent.

I think people use Fahmon as an estimate of their PPD. We're aware that PPD isn't accurate unless the time the WU was accepted by Stanford. Even still, it would be nice to see a semi-accurate PPD that includes the bonus.

DriveEuro wrote:I think people use Fahmon as an estimate of their PPD. We're aware that PPD isn't accurate unless the time the WU was accepted by Stanford. Even still, it would be nice to see a semi-accurate PPD that includes the bonus.

But because points are awarded for each WU completed, Fahmon PPD values are pretty accurate (even though points may not show up in Stanford number for a little while).

The same accuracy will not be possible with the -bigadv, unless (as I said below) Fahmon tracks when the WU are submitted.

You'll get a quicker response from the Fahmon developer if you post in the existing Fahmon thread, for which he has a subscription. He may not see this one for several days.

Hey guys. I have an Excel spreadsheet that calculates the Bonus based on the frame time. Currently, you have to figure out the frame times yourself and enter them in minutes then seconds, but I can make it to where all you would have to do is copy and paste a portion of the FAHlog into a preset cell range. I'm trying to do this without using macros because of security.

All the variables/constants in the formula can be entered in cells. There are enough for two different projects and five frame time entries for each project.

There are comments on the spreadsheet to help with usage. Here is a link to my post at EOC. Anyone who wants to use or modify it is welcome and have my permission to host it as well.

http://forums.extremeoverclocking.com/s ... stcount=50