Credit versus PPD

[cyclometric]

Hi,

I am running VMWare Player on Vista 64, with the Notfred's SMP client, on 2 cores of my Phenom x4 @ 3675ghz, and I use FahMon to track progress. I am a bit confused by the discrepancy between the terms "PPD" and "Credit" which are given in the main view. For example, right now PPD says "2359.72", and Credit says "1920 points". Actually, the value for Credit is always 1920, and that happens to be the same amount which I believe is being added to my actual FAH points every day.

I assumed that PPD meant "points per day", but I don't understand why the same static value (1920) is added daily, instead of an actual number of points that FahMon seems to reflect (and btw which averages about 350-400 more points per day...)

Do I need to reset something, or what? It seems unusual to me that the same number of points is added daily, as if a cap had been set somewhere among my settings, or on some server perhaps?

Thank you for any insight you can give me.

cyclometric

[John Naylor]

PPD does mean points per day, but that assumes that you constantly work. Data transfer overheads (as in, uploading/downloading) will reduce the real value slightly (or a lot, if you have a slow connection). Each unit is inidividually worth 1920 points, so when you complete the unit, that is what is awarded to your username. You will find that every so often you will get 3840 points in one day, as all the extra time not used doing 1 unit adds up to mean you complete one unit early in the morning and another later in the day. Consider this as the extra "PPD" over the "credit" for each unit over the past few days being awarded retrospectively.

If your PPD is higher than the credit value of the unit, this means that you are completing units in less than one day. So if (for example) you completed every unit in 22 hours, eventually a day would occur where you would finish one unit at 2am and another at midnight and they would be credited on the same day. This is where the "extra" points will appear.

[cyclometric]

Ah, OK, I get it! That's makes perfect sense. And I suppose in the statistics web applications, until someone has been folding long enough for some of those XXtra-value-points days to have accrued, the system only can predict that you will be advancing 1 WU per day, i.e. 1920, when in actuality, over time, that # will be higher, on average.

Thanks a lot for a great explanation.

[John Naylor]

until someone has been folding long enough for some of those XXtra-value-points days to have accrued, the system only can predict that you will be advancing 1 WU per day, i.e. 1920, when in actuality, over time, that # will be higher, on average.

It will be, but the statistics system does not use predictions, it only awards credit when units are returned. This cycle of missing a load of points then occasionally getting a dump of points one day will continue forever.

Thanks a lot for a great explanation.

You're welcome

[cyclometric]

It will be, but the statistics system does not use predictions, it only awards credit when units are returned. This cycle of missing a load of points then occasionally getting a dump of points one day will continue forever.

I wasn't clear about which system I was referring to, that is, I was talking about the statistics apps on such as on extremeoverclocking.com, which predict future results, not the actual live points tabulation on the Stanford side. I'll just have to keep watching that curve and see how the slope of that line changes.

Yeah, big surprise, a geek, huh?

[John Naylor]

I'm one too, don't worry

[cyclometric]

I figured I would be in good company at a place like this. When I worked at MSFT in Redmond, my Dad used to kid me by saying he could imagine the geeks walking the halls, shirts unbuttoned wrong, taped eyeglasses, mismatched socks, and the like --- and some days, that was pretty much spot on how it was, too. LOL.