GPU72 / MISFIT use for 100M digit range?

[kracker]

Nice! Just pulled some in, is there a way to see who is working on the range there? Right now this is the only way I can tell it is being assigned to people...

[chalsall]

Originally Posted by kracker

Nice! Just pulled some in, is there a way to see who is working on the range there? Right now this is the only way I can tell it is being assigned to people...

Because the depth can be much deeper for LMH than regular LLTF, I'm going to have to figure out a way to expose that without making the page (to much) larger than it already is.

Still thinking on that....

[kracker]

Originally Posted by chalsall

Because the depth can be much deeper for LMH than regular LLTF, I'm going to have to figure out a way to expose that without making the page (to much) larger than it already is.

Still thinking on that....

Ah, I see.

...

M332412151 has a factor: 3465422469903073527641 [TF:71:72*:mfakto 0.12-Win barrett15_75]

[chalsall]

Originally Posted by kracker

M332412151 has a factor: 3465422469903073527641 [TF:71:72*:mfakto 0.12-Win barrett15_75]

Thanks. That's the next thing I was waiting for. How does the system deal with the "value" of factors found. James gave me an elegant formula, which has not been implemented yet....

[Chuck]

Originally Posted by Chuck

OK I took one of these just for fun...I have not factored above 73 before.

Factor=N/A,332350xxx,73,78

This really cuts into the normal throughput; normally I was doing 436 GHz-d/days, now only getting 394

[firejuggler]

usually get 210, I get 180

[chalsall]

Originally Posted by firejuggler

usually get 210, I get 180

Yeah... I usually get ~214, I'm now getting ~192.

Oliver... Counsel?

More data:

got assignment: exp=62989193 bit_min=71 bit_max=72 (7.59 GHz-days)
Starting trial factoring M62989193 from 2^71 to 2^72 (7.59 GHz-days)
 k_min = 18742764662160
 k_max = 37485529326194
Using GPU kernel "barrett76_mul32_gs"
[date    time]  exponent   : percent  class #, seq    |     GHz |    time |    ETA |    #FCs |      rate | SieveP. | CPU wait |
[Feb 27 12:10] M62989193   : 100.0% 4615/4620,960/960 |  214.35 |  3.188s |  0m00s |   4.06G | 1272.5M/s |   82485 |    n.a.% |
no factor for M62989193 from 2^71 to 2^72 [mfaktc 0.20 barrett76_mul32_gs]
tf(): total time spent: 51m  0.827s

got assignment: exp=332410703 bit_min=71 bit_max=72 (1.44 GHz-days)
Starting trial factoring M332410703 from 2^71 to 2^72 (1.44 GHz-days)
 k_min = 3551605314180
 k_max = 7103210637097
Using GPU kernel "barrett76_mul32_gs"
[date    time]  exponent   : percent  class #, seq    |     GHz |    time |    ETA |    #FCs |      rate | SieveP. | CPU wait |
[Feb 27 12:21] M332410703  : 100.0% 4617/4620,960/960 |  191.83 |  0.675s |  0m00s | 768.75M | 1138.9M/s |   82485 |    n.a.% |
no factor for M332410703 from 2^71 to 2^72 [mfaktc 0.20 barrett76_mul32_gs]
tf(): total time spent: 10m 48.484s

[Chuck]

Originally Posted by chalsall

Yeah... I usually get 214.35, I'm now getting 191.83.

Oliver... Counsel?

I reduced my GPUSievePrimes to 70000 some time ago (it actually uses 69941); wonder if that needs to be revisited?

[firejuggler]

I wonder what happens if we switch back to 0.19 for the 100 million digit range.
otherwise it's a 10 to 15% performance hit

[chalsall]

Originally Posted by firejuggler

I wonder what happens if we switch back to 0.19 for the 100 million digit range.

Try it. We won't know until we do.

[firejuggler]

Factor=N/A,332350xxx,73,74

0.19
CPU
Sieveprime @57xx
1H11

0.20
GPU
sieveprime @ 70xxx
45 min.

So even if we loose a bit of speed, 0.20 seem still faster (Mfaktc)