In CUDALucas, do I have to wait until 100% completion before knwoing that my cadidate is composite?

[science_man_88]

Originally Posted by ICWiener

Iteration times are around 60 ms, and the PC is on 24/7. Note that I'm only using a home PC, nothing fancy with a GeForce GTX Ti 1050. My exponent isn't in't the billions range, as someone suggested. Instead, it's in the 100 million.

So ~432M range as that's near the 100 million digit range.

[ICWiener]

Originally Posted by kriesel

Welcome to the hunt. (Futurama fan?)

a) yes
b) only in the sense that it has not obviously crashed yet. Only the last Lucas-Lehmer sequence output tells of primality status, the other iterations are necessary steps to reach it.

Did you use good practice, fft benchmarking the card, in CUDALucas for the fft lengths of interest, threads benchmarking the card, run the residue selftest, run memory test over the whole card, run a couple of low double checks to verify it's working at differing fft lengths, before tackling a several months or some years' length assignment? Check out the exponent? Confirm that 64-bit residues are changing from one status line to the next, and known-bad behavior not happening? What version CUDALucas is running? (V2.06 May 5 2017 beta?)

Yes, thanks for noticing :)

Yes, I did all those benchmarks prior to launching my test. The residues are changing at every output. I am running version 2.05.1.

[ICWiener]

Originally Posted by science_man_88

So ~432M range as that's near the 100 million digit range.

Correct. Although \(log_2(10^{10^8})\approx 332 192 809\), so any prime exponent above that number yields a 10^8+ digit number.

[kriesel]

Originally Posted by ICWiener

Yes, thanks for noticing :)

Yes, I did all those benchmarks prior to launching my test. The residues are changing at every output. I am running version 2.05.1.

2.06beta has some error checking that 2.05.1 doesn't.

[kriesel]

Originally Posted by ICWiener

Iteration times are around 60 ms, and the PC is on 24/7. Note that I'm only using a home PC, nothing fancy with a GeForce GTX Ti 1050. My exponent isn't in the billions range, as someone suggested. Instead, it's in the 100 million.

For comparison, from the same model. Timing looks to be a 100-mega-digit, ~332 million exponent. From FFT file timings predict total run time, 60msec * 332 million / (86400 sec/day) =~230 days.

Device GeForce GTX 1050 Ti
Compatibility 6.1
clockRate (MHz) 1468
memClockRate (MHz) 3504

fft max exp ms/iter
1 22133 0.1015
...
2592 48471289 7.0917
...
4096 75846319 10.7888
4608 85111207 12.7806
...
5600 103000823 16.6240
...
18432 330441847 54.1352
18816 337176443 60.5082
20480 366326371 62.0885
...
24192 431175197 79.0719
...
32768 580225813 99.6362
36864 651102253 112.1126
...
65536 1143276383 205.1426

[ICWiener]

Originally Posted by kriesel

For comparison, from the same model. Timing looks to be a 100-mega-digit, ~332 million exponent. From FFT file timings predict total run time, 60msec * 332 million / (86400 sec/day) =~230 days.

Device GeForce GTX 1050 Ti
Compatibility 6.1
clockRate (MHz) 1468
memClockRate (MHz) 3504

fft max exp ms/iter
1 22133 0.1015
...
2592 48471289 7.0917
...
4096 75846319 10.7888
4608 85111207 12.7806
...
5600 103000823 16.6240
...
18432 330441847 54.1352
18816 337176443 60.5082
20480 366326371 62.0885
...
24192 431175197 79.0719
...
32768 580225813 99.6362
36864 651102253 112.1126
...
65536 1143276383 205.1426

Nice analysis, you're not far off :)

If I switch to 2.06, can I continue from my last 2.05.1 checkpoint? Would I gain a significant temporal boost by using the new version?

[kladner]

Originally Posted by ICWiener

Nice analysis, you're not far off :)

If I switch to 2.06, can I continue from my last 2.05.1 checkpoint? Would I gain a significant temporal boost by using the new version?

You will save time if errors are detected earlier in the process.

[ATH]

Originally Posted by ICWiener

Iteration times are around 60 ms, and the PC is on 24/7. Note that I'm only using a home PC, nothing fancy with a GeForce GTX Ti 1050. My exponent isn't in't the billions range, as someone suggested. Instead, it's in the 100 million.

GeForce Ti 1050 is one of those cards better suited for trial factoring than LL tests:
https://en.wikipedia.org/wiki/GeForce_10_series

Single Precision speed (trial factoring): 1981 GFLOPs
Double Precision speed (LL test): 62 GFLOPs