I guess that's how the 74 got in there - the normal prime95 limit is 72 for exponents between 71M & 79.3M.

Yes you can set the override to whatever you want. It's just that setting it higher takes a hell of a lot of time.

lets see ... a 69.5M # from 61 to 62 takes 7.5min, so that same # from 62 to 73 would take an additional 7.5*2046 = 15345 minutes, or roughly 11 days.

I'm afraid not! the step from 62 to 63 and then from 64 to 65 MORE than doubles the amount of time required. So it will be more like 20 days.

those the change in FFT sizes used? i'm not completely familiar with the algorithm, and it's only the second week back in classes so i'm not sure my mind would fully understand anything complex this gets in response - but try it and i'll try to comprehend again in a few weeks :)

No it is not the FFT sizes. There was some super efficient algo for non-SSE2 CPUS that was developed that works reall well for < 62 bits. Hence the jump from 62 to 63. 64bits can fit in two words whereas 65 bits need three. So the jump there is due to that.

is there a way to have my computer stop using the SSE2 and use something else to make it more efficient at lower bits?

No unfortunately not. P4 actually sucks at non-SSE2 calculations on a per-clock basis. It only performs better as it uses SSE2 for 64 bits and above. That is why I reommended using a non-P4 for under 62 bits.

how about P3's?

P3s are fine. I think if you just do a quick test at factoring a number to 62 bits on both a P4 and a P3 you will be surprised at the results. For example look at these timing tests I did a while back. the 1333 Athlon outperforms the 2533 P4 upto 62 bits! :shock:

[code:1]

Time taken to complete 14366959 to 0.98%

To Bit On PII 450 On P4 2533 Improv On 1333TB Improv

59 6 2.5 2.4 1.85 3.25
60 12 5 2.4 3.7 3.25
61 24 10 2.4 7.4 3.25
62 48 20 2.4 14.8 3.25
63 170 37.5 4.5 51 3.33
64 340 75 4.5 102 3.33
65 1540 180 8.5 480 3.21
[/code:1]

This data indicates that it is best to do upto 62 bits on a non-P4, maybe even upto 64 bits. And after 64 bits, the P4 smokes due to SSE2.

once my p3 finishes its current assignment from primenet, i'll have some numbers of my own to post about p3 v p4 - yey - dance time :banana:

also, what's a 1333TB?