Trial factoring Zero for Seven Thousand?

[TheMawn]

I don't mean to pry into someone else's business who might not want the spotlight put on them, but a user by the name of Craig Meyers is 14th for trial factoring, and at the time of my checking, has 151132 GHz-Days across 7486 attempts and zero successes. Average assignment value is 20GHz-Days which, I think, corresponds roughly to Makes Sense ^TM.

My personal results with Makes Sense^TM were roughly 1/(# of bits), so is this really bad luck? Faked results? Hardware issues?

#22 Bill Staffen looks a bit suspect too.

[TheMawn]

A couple of Zero Successes down around ranks 75 and 76, also, for 1000+ attempts. I'm going to stop looking into this until I get a bit of feedback. I might just be missing something.

[kracker]

Craig Meyers is nucleon, Bill Staffen is Aramis Wyler in these forums.

I think the "Successes" are for the year, the attempts aren't... I am guessing because nucleon has had his gpu's rest for a while now (I think)

I may be wrong.

[TheMawn]

Funny. I just saw a post from less than an hour ago from Nucleon, signed Craig. I thought to myself "hmmm... THE Craig?" I guess it was.

He was mentioning putting his Titans to work on P-1 so you could be right... On the other hand, aren't the GHz-Days also on a per year thing? Because that's still 150THz-Days with zero successful factoring attempts...

[kladner]

Craig is in Australia. He initially went quiet for southern hemisphere summer for reasons of economy. The amount of hardware he runs is expensive to feed at any time of the year.

[Mini-Geek]

This has been noticed before, with other users (myself included), and is probably due to PrimeNet thinking his factors are P-1, not TF. He has 255 "P-1" successes for 2651 attempts - that's 9.6%. You'd expect about 100 factors to be found in that many TFs, so if you assume that 100 of those factors are really TF, and the rest of the P-1 successes and attempts are truly P-1, then you get the more reasonable P-1 success rate of 6.1%.

[Aramis Wyler]

Originally Posted by TheMawn

#22 Bill Staffen looks a bit suspect too.

!! I'm a shady character I guess! But my results are a little off because I run LMH numbers in the 332M range to 79 bits for Uncwilly in exchange for him keeping his minions on p-1 work. Those are really long jobs (16-23 hours each on my 580) and I've only had 2 or 3 successes.

mfaktc v0.20 (32bit built)
Compiletime options
  THREADS_PER_BLOCK         256
  SIEVE_SIZE_LIMIT          32kiB
  SIEVE_SIZE                193154bits
  SIEVE_SPLIT               250
  MORE_CLASSES              enabled
Runtime options
  SievePrimes               25000
  SievePrimesAdjust         1
  SievePrimesMin            5000
  SievePrimesMax            100000
  NumStreams                3
  CPUStreams                3
  GridSize                  3
  GPU Sieving               enabled
  GPUSievePrimes            70000
  GPUSieveSize              128Mi bits
  GPUSieveProcessSize       16Ki bits
  WorkFile                  worktodo.txt
  Checkpoints               enabled
  CheckpointDelay           900s
  Stages                    enabled
  StopAfterFactor           class
  PrintMode                 compact
  V5UserID                  (none)
  ComputerID                (none)
  AllowSleep                no
  TimeStampInResults        no
CUDA version info
  binary compiled for CUDA  4.20
  CUDA runtime version      4.20
  CUDA driver version       5.0
CUDA device info
  name                      GeForce GTX 580
  compute capability        2.0
  maximum threads per block 1024
  number of multiprocessors 16 (512 shader cores)
  clock rate                1544MHz
Automatic parameters
  threads per grid          1048576
running a simple selftest...
Selftest statistics
  number of tests           92
  successfull tests         92
selftest PASSED!
got assignment: exp=332237161 bit_min=77 bit_max=79 (276.38 GHz-days)
Starting trial factoring M332237161 from 2^77 to 2^78 (92.13 GHz-days)
 k_min = 227421470545500
 k_max = 454842941099622
Using GPU kernel "barrett87_mul32_gs"
found a valid checkpoint file!
  last finished class was: 3704
  found 0 factor(s) already
Date    Time | class   Pct |   time     ETA | GHz-d/day    Sieve     Wait
Aug 10 13:52 | 4616 100.0% | 23.479   0m00s |    353.15    69941    n.a.%
no factor for M332237161 from 2^77 to 2^78 [mfaktc 0.20 barrett87_mul32_gs]
tf(): time spent since restart:    1h 13m 14.219s
      estimated total time spent:  6h 15m 58.557s
Starting trial factoring M332237161 from 2^78 to 2^79 (184.26 GHz-days)
 k_min = 454842941095620
 k_max = 909685882199244
Using GPU kernel "barrett87_mul32_gs"
Date    Time | class   Pct |   time     ETA | GHz-d/day    Sieve     Wait
Aug 11 02:28 | 4616 100.0% | 47.257   0m00s |    350.91    69941    n.a.%
no factor for M332237161 from 2^78 to 2^79 [mfaktc 0.20 barrett87_mul32_gs]
tf(): total time spent: 12h 35m 33.114s
got assignment: exp=332241941 bit_min=76 bit_max=79 (322.44 GHz-days)
Starting trial factoring M332241941 from 2^76 to 2^77 (46.06 GHz-days)
 k_min = 113709099305340
 k_max = 227418198613023
Using GPU kernel "barrett77_mul32_gs"
Date    Time | class   Pct |   time     ETA | GHz-d/day    Sieve     Wait
Aug 11 05:28 | 4615 100.0% | 11.297   0m00s |    366.97    69941    n.a.%
no factor for M332241941 from 2^76 to 2^77 [mfaktc 0.20 barrett77_mul32_gs]
tf(): total time spent:  3h  0m 45.176s
Starting trial factoring M332241941 from 2^77 to 2^78 (92.13 GHz-days)
 k_min = 227418198610680
 k_max = 454836397226046
Using GPU kernel "barrett87_mul32_gs"
Date    Time | class   Pct |   time     ETA | GHz-d/day    Sieve     Wait
Aug 11 08:03 | 1899  40.8% | 23.646   3h43m |    350.65    69941    n.a.%
M332241941 has a factor: 154379640999923508617839
found 1 factor for M332241941 from 2^77 to 2^78 (partially tested) [mfaktc 0.20 barrett87_mul32_gs]
tf(): total time spent:  2h 34m 26.662s
got assignment: exp=332242007 bit_min=75 bit_max=79 (345.47 GHz-days)
Starting trial factoring M332242007 from 2^75 to 2^76 (23.03 GHz-days)
 k_min = 56854538354460
 k_max = 113709076718156
Using GPU kernel "barrett76_mul32_gs"
Date    Time | class   Pct |   time     ETA | GHz-d/day    Sieve     Wait
Aug 11 09:29 | 4609 100.0% |  5.371   0m00s |    385.93    69941    n.a.%
no factor for M332242007 from 2^75 to 2^76 [mfaktc 0.20 barrett76_mul32_gs]
tf(): total time spent:  1h 26m  5.054s
Starting trial factoring M332242007 from 2^76 to 2^77 (46.06 GHz-days)
 k_min = 113709076713540
 k_max = 227418153436312
Using GPU kernel "barrett77_mul32_gs"
Date    Time | class   Pct |   time     ETA | GHz-d/day    Sieve     Wait
Aug 11 12:29 | 4609 100.0% | 11.279   0m00s |    367.56    69941    n.a.%
no factor for M332242007 from 2^76 to 2^77 [mfaktc 0.20 barrett77_mul32_gs]
tf(): total time spent:  3h  0m 34.482s
Starting trial factoring M332242007 from 2^77 to 2^78 (92.13 GHz-days)
 k_min = 227418153431700
 k_max = 454836306872624
Using GPU kernel "barrett87_mul32_gs"
Date    Time | class   Pct |   time     ETA | GHz-d/day    Sieve     Wait
Aug 11 18:48 | 4609 100.0% | 23.602   0m00s |    351.30    69941    n.a.%
no factor for M332242007 from 2^77 to 2^78 [mfaktc 0.20 barrett87_mul32_gs]
tf(): total time spent:  6h 18m 35.098s
Starting trial factoring M332242007 from 2^78 to 2^79 (184.25 GHz-days)
 k_min = 454836306868020
 k_max = 909672613745248
Using GPU kernel "barrett87_mul32_gs"
Date    Time | class   Pct |   time     ETA | GHz-d/day    Sieve     Wait
Aug 12 07:25 | 4609 100.0% | 47.311   0m00s |    350.51    69941    n.a.%
no factor for M332242007 from 2^78 to 2^79 [mfaktc 0.20 barrett87_mul32_gs]
tf(): total time spent: 12h 37m 13.750s
got assignment: exp=332242091 bit_min=75 bit_max=79 (345.47 GHz-days)
Starting trial factoring M332242091 from 2^75 to 2^76 (23.03 GHz-days)
 k_min = 56854523981640
 k_max = 113709047969353
Using GPU kernel "barrett76_mul32_gs"
Date    Time | class   Pct |   time     ETA | GHz-d/day    Sieve     Wait
Aug 12 08:51 | 4609 100.0% |  5.366   0m00s |    386.29    69941    n.a.%
no factor for M332242091 from 2^75 to 2^76 [mfaktc 0.20 barrett76_mul32_gs]
tf(): total time spent:  1h 26m  2.219s
Starting trial factoring M332242091 from 2^76 to 2^77 (46.06 GHz-days)
 k_min = 113709047967900
 k_max = 227418095938706
Using GPU kernel "barrett77_mul32_gs"
Date    Time | class   Pct |   time     ETA | GHz-d/day    Sieve     Wait
Aug 12 11:52 | 4609 100.0% | 11.275   0m00s |    367.69    69941    n.a.%
no factor for M332242091 from 2^76 to 2^77 [mfaktc 0.20 barrett77_mul32_gs]
tf(): total time spent:  3h  0m 32.929s
Starting trial factoring M332242091 from 2^77 to 2^78 (92.13 GHz-days)
 k_min = 227418095935800
 k_max = 454836191877412
Using GPU kernel "barrett87_mul32_gs"
Date    Time | class   Pct |   time     ETA | GHz-d/day    Sieve     Wait
Aug 12 18:10 | 4609 100.0% | 23.622   0m00s |    351.00    69941    n.a.%
no factor for M332242091 from 2^77 to 2^78 [mfaktc 0.20 barrett87_mul32_gs]
tf(): total time spent:  6h 17m 51.954s
Starting trial factoring M332242091 from 2^78 to 2^79 (184.25 GHz-days)
 k_min = 454836191876220
 k_max = 909672383754824
Using GPU kernel "barrett87_mul32_gs"
Date    Time | class   Pct |   time     ETA | GHz-d/day    Sieve     Wait
Aug 12 20:21 |  796  17.3% | 47.277  10h25m |    350.76    69941    n.a.%