recent cleared list

ramgeis · 2016-05-11, 20:51

Hi there,

I just sorted the "recent cleared list" by descending GHz-days and one would expect to also have a descending order in the exponent column for the LL results but at the moment exponent 78358727 seems to be out of line for some reason. Any explanation for this?

chalsall · 2016-05-11, 21:11

Quote:

Originally Posted by ramgeis

Any explanation for this?

Yes.

This was assigned as a Cat 4 to an unknown worker. Said worker worked it quickly.

ramgeis · 2016-05-11, 21:19

Quote:

Originally Posted by chalsall

Yes.

This was assigned as a Cat 4 to an unknown worker. Said worker worked it quickly.

224.2 days doesn't sound quickly to me.

ric · 2016-05-11, 21:23

I guess the OP is asking "given that similar exponents in the 78M2 range earn about 240 GHz/days each, and exponents in the 84M7 range earn about 270 GHz/days, how is it that a candidate in 78M3 earns 280+ GHz/days?"

Pls correct me if this ain't the right question. No idea how or why this happened, though.

EDIT: the only legit reason coming to my mind was that the candidate was checked with a larger FFT-size (a former version of Prime95 maybe, or a deliberate choice with a different program), thus giving the extra-yield. Just a guesstimate, however.

Quote:

Originally Posted by chalsall

Said worker worked it quickly.

Hmmm... 224 days is the new quick? ;-P

chalsall · 2016-05-11, 21:28

Quote:

Originally Posted by ric

Hmmm... 224 days is the new quick? ;-P

Better than many (many!) years which some clients gleefully report...

chalsall · 2016-05-11, 21:45

Quote:

Originally Posted by ric

I guess the OP is asking "given that similar exponents in the 78M2 range earn about 240 GHz/days each, and exponents in the 84M7 range earn about 270 GHz/days, how is it that a candidate in 78M3 earns 280+ GHz/days?"

OK. This has been covered before. But let's do this yet again...

The credits awarded for work is a function of the FFT size used.

ric · 2016-05-11, 21:57

Quote:

Originally Posted by chalsall

The credits awarded for work is a function of the FFT size used.

yup, we cross-posted.

ramgeis · 2016-05-12, 13:39

Thanks for the clarification...

Must have missed that post, sorry for asking a faq.

Madpoo · 2016-05-14, 23:51

Quote:

Originally Posted by ric

EDIT: the only legit reason coming to my mind was that the candidate was checked with a larger FFT-size (a former version of Prime95 maybe, or a deliberate choice with a different program), thus giving the extra-yield. Just a guesstimate, however.

That result was run by a CPU running this version of Prime95, for some reason:
Windows64,Prime95,v25.11,build 2

That is a pretty old version... so maybe that solves this mystery? It used a much larger FFT size than it would have with a newer version.

FYI, the CPU that did it reports itself as:
Intel(R) Xeon(R) CPU E5504 @ 2.00GHz

2016-05-11, 20:51	#1
ramgeis Apr 2013 2²×29 Posts	recent cleared list Hi there, I just sorted the "recent cleared list" by descending GHz-days and one would expect to also have a descending order in the exponent column for the LL results but at the moment exponent 78358727 seems to be out of line for some reason. Any explanation for this?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Recent Results/Factoring	Primeinator	PrimeNet	4	2014-10-26 00:56
Recent assignment	kladner	Software	20	2011-09-02 02:59
Can "Recent Result" list be longer?	ATH	PrimeNet	1	2011-07-08 14:48
PM box cleared out...	gd_barnes	No Prime Left Behind	17	2008-10-05 07:48
List of recent exponents that have been passed out	jasong	PrimeNet	5	2007-05-18 22:56

2016-05-12, 13:39	#8
ramgeis Apr 2013 2²×29 Posts	Thanks for the clarification... Must have missed that post, sorry for asking a faq.