Yet, another query - re: vastly different sieve returns
 

[strike]I have a machine that is returning only ~12% as many relations as others. I'm curious as to why and whether I can do anything about it. I tried it and another machine with identical poly files several times with the same results each time. I have the poly and one of the runs from each machine below:[/strike]

For some reason, I incorrectly believed that B[SUP]2[/SUP]'s sievers did not run on AMD. This erroneous idea must have been from an earlier experience with one of my antiques.:sad:

B[SUP]2[/SUP]'s sievers are, indeed, returning many more relations in a speedier manner.:smile:

Rels number for lpba != lpbr
 

[SIZE=3][FONT=Calibri]Roughly speaking, I need 40M (unique) relations for lpba=lpbr=29 and 80M relations for lpba=lpbr=30. (More exact estimations are out of interest now.)[/FONT][/SIZE]

[SIZE=3][FONT=Calibri]My question is, How many relations I need for lpba != lpbr, i.e.[/FONT][/SIZE]
[SIZE=3][FONT=Calibri]1) lpbr=29, lpba=30[/FONT][/SIZE]
[SIZE=3][FONT=Calibri]2) lpbr=30, lpba=29 ?[/FONT][/SIZE]

[SIZE=3][FONT=Calibri]Does the answer depend on a side of sieving, i.e.[/FONT][/SIZE]
[SIZE=3][FONT=Calibri]r) rational side[/FONT][/SIZE]
[SIZE=3][FONT=Calibri]a) algebraic side[/FONT][/SIZE]
[SIZE=3][FONT=Calibri]b) both sides.[/FONT][/SIZE]

[SIZE=3][FONT=Calibri]It seems to me, I need the maximal number in any case, i.e.[/FONT][/SIZE]

[SIZE=3][FONT=Calibri]1r) 80M[/FONT][/SIZE]
[SIZE=3][FONT=Calibri]1a) 80M[/FONT][/SIZE]
[SIZE=3][FONT=Calibri]1b) 80M[/FONT][/SIZE]

[SIZE=3][FONT=Calibri]2r) 80M[/FONT][/SIZE]
[SIZE=3][FONT=Calibri]2a) 80M[/FONT][/SIZE]
[SIZE=3][FONT=Calibri]2b) 80M[/FONT][/SIZE]

[SIZE=3][FONT=Calibri]Is it correct? Otherwise, give your variants, please.[/FONT][/SIZE]

[FONT=Calibri][SIZE=3]Thank you in advance.[/SIZE][/FONT]

Split the difference. Let's say, 60M.

I've done a few like that. 57^145-1 with LPBA=29 and LPBR=30 needed 56385091 relations after removing duplicates (53556727 relations was too few).

I don't think it would vary with which side has the larger LPB. But I've not done one with larger LPBA.

Chris

If raising both sides one bit doubles the relations needed, raising one side should scale relations by sqrt2. Chris' 56M fits this exactly.

I do not think it matters which side is up one bit, though I imagine 29/30 vs 30/29 would sieve with different efficiencies.

LPBR>LPBA is appropriate for SNFS polys with lower degree than optimal. That was a quartic at about 200 digits SNFS difficulty.

LPBA>LPBR would be appropriate if the degree is too high. Eg degree 7 or 8. But I've not done any such.

Chris