GHz days saved update
 

Does the GHz days saved column not update along with the work done in the Workers Overall Progress? Mine doesn't seem to be moving.

[QUOTE=Chuck;290370]Does the GHz days saved column not update along with the work done in the Workers Overall Progress? Mine doesn't seem to be moving.[/QUOTE]

According to the datasets, you have not found a factor since 02.18, even though you have submitted work every day since then. That is why your GHz Days Saved graph hasn't been updated.

Note to self: Render every day of interest; even if zero....

[QUOTE=Chuck;290370]Does the GHz days saved column not update along with the work done in the Workers Overall Progress? Mine doesn't seem to be moving.[/QUOTE]

Just to be clear, we're talking about [URL="http://www.gpu72.com/reports/worker/390a11e492481f62d201b6578395bd14/"]this[/URL]. Correct?

Note that a savings metric is only measured when a factor is found.

DUUHHHHHHHH — of course, what was I smoking?

[QUOTE=Chuck;290386]DUUHHHHHHHH — of course, what was I smoking?[/QUOTE]

Well, Chuck, you are not alone in seeing a declining "Work Saved" metric. But I am in the same boat as you: the "few or no factors found" boat.

One saves work only if he finds a factor. Otherwise there is no work saved, no matter ho much TF he does, the LL and DC tests would still need to be done. Except if he TF up to the square root of the number (to a number of bits equal to half of exponent).

This is clear and the table on gpu272 page is right. What regression are you talking about? There is no need for any regression there. The total of the work saved, or a daily average (not EMA!) would be more than enough.