View Single Post
2021-09-17, 04:09   #753
petrw1
1976 Toyota Corona years forever!

"Wayne"
Nov 2006

3×1,613 Posts
I guess as the "inventor" of this project...

Quote:
 Originally Posted by chalsall Would anyone like to coordinate the last bit of P-1'ing needed in 22.1, 22.7, and 22.8M? Oliver "did his thing".
Most importantly speak for your ranges to avoid toe stepping

There are actually 21 ranges that Oliver has completed to 74 bits (22.8 is special ... he is finished).
P.S. within a few weeks he will probably complete the rest (fingers crossed)

I have ordered them by "ToGo".
Then I have suggested P1 effort.
Always work on the exponents with B1=B2 first.
Do not fear that some ranges have NO B1=B2 remaining.
Most of these exponents have had very meagre P1 based on today's hardware.
In these cases look for low B1/B2 here .
Sort by B2 (click on the column) and start at the top.
The top exponents will likely have a current P1 success rate of under 2%; most will be under 3%
Choosing a B1/B2 that give 4% (+2%) for example will require on average 50 exponents per factor.
Each assignment will take about 2 GhzDays ... not a whole lot, right?

Code:
Range	To Go	B1=B2	Low P-1
This first bunch have so few remaining; don't pick B1/B2 with over 3-4% success rate
22.8M	1		302
29.6M	1	579	122
25.1M	2	497	162
24.6M	3		189
22.1M	4		265
23.6M	7	602	258
27.4M	9	761	163
22.7M	10	130	322
26.3M	10	722	145
28.1M	17	679	161
27.3M	17	712	153
27.7M	23	770	159
28.2M	24	610	157
28.5M	24	330	153
28.9M	25	643	154
27.5M	26	382	175

For these we'll need about a 5-5.5% success rate
27.8M	41	791	179
29.0M	47	715	176
28.0M	50	531	163

These 7-8% but probably better another bit of TF and then something more reasonable
27.2M	89	821	150
29.8M	113	789	100
Does anyone need actual recommended B1/B2 for PMinus1 or B1 for PFactor?