Credit for trial factoring

[bayanne]

I notice that neither George or James have posted in this post, but I still say that the formula for working the factoring credit changed sometime earlier this month, about the time that I initiated this thread. This is what my results history shows.
You obviously have made your statement, and I am unable to argue with you, as I do not have your technical knowledge. However I believe that I am not alone in my thoughts, but don't wish to antagonize anyone at all.
As Bdot said, the good times are over ....

[LaurV]

Bdot can easily make typing mistakes in reports like the one in post #8 He is not a trustful person...

Originally Posted by Bdot

OTOH, so far I NEVER got less credit for a factor than for completing a bitlevel with NF. Typically, I'd say, it is 100-200% of the credit of an NF for the same range.
As in

INFO:  M65531953 submitted; 29.1922 GHz Days credit.   (73 -> 74)
<snip>
INFO:  M65771527 submitted; 14.5429 GHz Days credit.   (72 -> 73)
INFO:  M65771899 submitted; 37.8749 GHz Days credit.   (73.3 bit factor)
INFO:  M65772071 submitted; 14.5428 GHz Days credit.   (73 -> 74)

Joking apart, I don't say that the method to calculate the credit did not change. You may be right. Or Bdot may be right. Only George and James can tell. I just say that for me it was ever like that, I never got a bigger credit for a factor, than I would get for a "no factor" in the same bitlevel. And (if no bonus is given) it is normal to get a lower credit, because you spent less time. That is the only thing I said, I don't want to argue with anybody. For me, everything looks normal, as it ever was. I may be getting older, however

I however think that you people are trusting too much your scripts and you don't check how much credit PrimeNet REALLY gave you for assignments, those reports like in my post #12. And when you suddenly and inadvertently find out, you are very fast in claiming that "something changed"

[bayanne]

I never spend less time, because I always factor the exponent to the end of its bit level, rather than finishing at the end of the class. So the time spent factoring would be the same time as for an exponent which did not have a factor found.

I have merely made an observation based on a reasonable sample size.

[axn]

Originally Posted by LaurV

I however think that you people are trusting too much your scripts and you don't check how much credit PrimeNet REALLY gave you for assignments, those reports like in my post #12. And when you suddenly and inadvertently find out, you are very fast in claiming that "something changed"

All of my factors.

[TheMawn]

Regarding factoring to the end of the bit level even if a factor was found, LaurV (quickly) converted me to stopping.

If the project ever turns into a find-the-factors project (because we have infinity billion GPU's suddenly) then we will want to factor to the end regardless. In fact, partial factoring will be worth zero because you have to redo the whole bit level if you don't know how the classes were set up.

On the other hand, a factor is worth one or two LL tests so maybe it's fair that you get more credit? Actually I think I've brought myself through this thought process before and it got nowhere.

[Mini-Geek]

Originally Posted by TheMawn

If the project ever turns into a find-the-factors project ...

...then, since PrimeNet doesn't record whether you did a partial or to-the-end search when you found a factor, we'll assume that we need to redo the whole bit range anyway. So nothing is gained by completing the bit range (except in ranges where we are in a find-the-factors mode; but those aren't typically TF'd).

[bayanne]

There was another thread which discussed the ethics of performing an incomplete factoring to a bit level and the consensus seemed to be that it was 'better' to do this, rather than just to the class level.
If that consensus has now moved the other way, I am happy to change

[Bdot]

Originally Posted by LaurV

Bdot can easily make typing mistakes in reports like the one in post #8 He is not a trustful person...

Boy, someone's reading my stuff to the dot. Cool, keep going

Originally Posted by LaurV

I however think that you people are trusting too much your scripts and you don't check how much credit PrimeNet REALLY gave you for assignments, those reports like in my post #12. And when you suddenly and inadvertently find out, you are very fast in claiming that "something changed"

I briefly thought you could be right with this, and the new web page layout could have tricked the submit_spider. However, the primenet report clearly shows the change between 2014-02-05 and 2014-02-08 (same bad formatting as your #12 ):

CPU Name        Exponent    Result Type    Received    age days    Result              GHz-Days
Manual testing    65983289    F    2014-02-25 10:28    0.0    4971076954942570156649    1.0734
Manual testing    65477507    F    2014-02-24 11:44    0.0    12663102147556814658191    12.3600
Manual testing    65986153    F    2014-02-23 10:26    0.0    7240382971298531589289    8.9374
Manual testing    65986033    F    2014-02-22 10:27    0.0    4822533015970008558073    0.4389
Manual testing    65761351    F    2014-02-18 10:35    0.0    10853841280216457006249    5.8362
Manual testing    65753309    F    2014-02-10 10:39    0.0    16189503994930825683457    22.6198
Manual testing    65759651    F    2014-02-09 10:26    0.0    7323761882888134203151    9.2084
Manual testing    65761541    F    2014-02-08 10:25    0.0    8699247776308188055079    12.8198
Manual testing    65761207    F    2014-02-05 10:25    0.0    8375528634063682210849    26.5666
Manual testing    65765723    F    2014-02-04 11:39    0.0    9108843388977589917593    28.3259
Manual testing    65733469    F    2014-02-04 10:39    0.0    7612157239665569840441    24.5716
Manual testing    65771899    F    2014-02-03 21:32    0.0    11646199706640592599871    37.8749
Manual testing    65531951    F    2014-02-01 19:49    0.0    7907998769962613503369    25.4500
Manual testing    65571031    F    2014-01-28 22:10    0.0    8710789282820948395649    27.4696
Manual testing    65541913    F    2014-01-28 10:38    0.0    14605095256947365624183    47.5409
Manual testing    65572421    F    2014-01-21 10:26    0.0    6246733051929437688191    20.4716
Manual testing    65346877    F    2014-01-19 10:38    0.0    11411856718724448587071    37.2628
Manual testing    65371883    F    2014-01-15 11:26    0.0    5894394064099461926527    19.3089
Manual testing    65371703    F    2014-01-14 10:27    0.0    8140547550554612286943    26.1242
Manual testing    65360959    F    2014-01-12 10:39    0.0    15259096434338874034457    49.5223
Manual testing    65360531    F    2014-01-12 10:38    0.0    10252037559785378333881    32.7294
Manual testing    65374669    F    2014-01-12 10:25    0.0    6224352925289852013919    20.4578
Manual testing    65381737    F    2014-01-10 10:26    0.0    5820201298913626509767    19.0386
Manual testing    65319637    F    2014-01-08 21:59    0.0    10309586108828411918503    32.9864
Manual testing    65261507    F    2014-01-07 10:25    0.0    7907603518155528695537    25.5544
Manual testing    65261087    F    2014-01-05 10:26    0.0    8418955939659876734993    26.8796
Manual testing    65264273    F    2014-01-05 10:25    0.0    4917056359367000200921    15.5075
Manual testing    65267087    F    2014-01-04 10:26    0.0    6862351050580977896071    22.5547
Manual testing    65263507    F    2014-01-01 10:26    0.0    6234283038789276522001    20.5263
Manual testing    65266489    F    2014-01-01 10:25    0.0    8599645111057124380439    27.3263
Manual testing    65259373    F    2013-12-30 10:26    0.0    9104963855127898545943    28.5367
Manual testing    65262979    F    2013-12-30 10:25    0.0    9133487627104265977649    28.6012
Manual testing    65248993    F    2013-12-24 10:36    0.0    8871736832475716204039    27.9924
Manual testing    65157737    F    2013-12-23 10:25    0.0    5065687354011880256489    16.1635
Manual testing    65154941    F    2013-12-20 10:26    0.0    8407674051753018332063    26.8949
Manual testing    65157143    F    2013-12-18 10:25    0.0    7136434613407836933433    23.4222
Manual testing    65152141    F    2013-12-16 10:25    0.0    6158131564100228807321    20.3011
Manual testing    65147681    F    2013-12-16 10:25    0.0    5506849441358956254329    17.9348
Manual testing    65149351    F    2013-12-14 10:25    0.0    8796374039777563596913    27.8545
Manual testing    64750783    F    2013-12-11 10:26    0.0    6987103442023177161023    23.1185
Manual testing    64928503    F    2013-12-08 18:09    0.0    6470876830040947559303    21.4239

@axn: your view confirms the change between 2014-02-03 and 2014-02-19 where you get the same credit for a much bigger factor (of a slightly bigger exponent).

[c10ck3r]

Might I suggest a separate statistic for Total Factors Found (or something to this effect)?

[TheMawn]

Originally Posted by c10ck3r

Might I suggest a separate statistic for Total Factors Found (or something to this effect)?

This is the "successes" in your TF statistics.

EDIT: These were once upon a time displayed in user statistics but it seems they are not, now. Still appear in top producers, though.

[Bdot]

EDIT: Forget it, I was not paying attention - nothing has changed (this time).

Originally Posted by Bdot

Seems the good times are over, and this has been "fixed"

They are back ... I just submitted 3 factors and got credit like in the good old times

Manual testing    66610373    F    2014-03-07 16:08    0.0    8494034439425749743959    12.1619
Manual testing    65578871    F    2014-03-07 15:48    0.0    17415995573166599970359    25.7533
Manual testing    65587583    F    2014-03-07 15:44    0.0    8610939780518590585921    12.6391