[QUOTE=Viliam Furik;573076][URL="https://www.mersenne.ca/k/3960"]This link[/URL] should show factors with k=3960, but doesn't. Same for any other k.[/QUOTE]Thanks, fixed.
I also adjusted that page to allow you to filter by exponent range.

[QUOTE=James Heinrich;573120]Thanks, fixed.
I also adjusted that page to allow you to filter by exponent range.[/QUOTE]

Great, thanks!
:awesome:

Would it be possible for you to make something similar to the GIMPS progress tool for the factored exponents, which would display TF and P-1 range counts?

I don't know how much work it took to make the progress tool for unfactored exponents, but I assume it's not quick and simple.

My picture of the tool, which could be named "GIMPS factoring progress", is that it would have the same layout as the original, but instead of TF or LL, it would have the option to display TF or P-1. TF would work the same, only on the factored exponents, the factor count could be ignored. P-1 counts would be divided into ranges by the B1 value and grouped by GHz-D that the P-1 with the given B1 costs. For example, the ranges could be 0-1, 1-5, 5-10, 10-20, 20-50, 50-100, 100-1000, 1000+, each value in GHz-D.

Is this possible without too much work on your side?

[QUOTE=Viliam Furik;573444]I don't know how much work it took to make the progress tool for unfactored exponents, but I assume it's not quick and simple.
...
Is this possible without too much work on your side?[/QUOTE]Making a tool pretty involves much work. Making the GIMPS status tool also involves storing and processing a large amount of data daily so no it was neither quick nor simple. :smile:

The P-1 data you're asking for might be more directly available. Would something like this be the kind of information you're after? (Ignore the specific counts, my test copy of the database is from 2020-Dec-31.) This sample query showing counts of 10-11M exponents, grouped by log10(effort), and where the null row is exponents without P-1:[code]mysql> SELECT FLOOR(`exponent` / 1000000) AS `mrange`, AVG(`ghd_pm1`) AS `ghd_avg`, ROUND(LOG10(`ghd_pm1`)) AS `ghd_log10`, COUNT(*) AS `howmany` FROM `primenet_pminus1` WHERE (`exponent` BETWEEN 10000000 AND 12000000) GROUP BY `mrange` ASC, `ghd_log10` ASC;
+--------+----------------------+-----------+---------+
| mrange | ghd_avg | ghd_log10 | howmany |
+--------+----------------------+-----------+---------+
| 10 | 0 | NULL | 2411 |
| 10 | 0.025563424977008253 | -2 | 32 |
| 10 | 0.15486457931879102 | -1 | 17888 |
| 10 | 1.2316391576959007 | 0 | 3218 |
| 10 | 0.5188636872579206 | -0 | 1487 |
| 10 | 5.64300323846772 | 1 | 508 |
| 10 | 54.16449991861979 | 2 | 3 |
| 11 | 0 | NULL | 5407 |
| 11 | 0.02895539936920007 | -2 | 3 |
| 11 | 0.1788080590799759 | -1 | 13152 |
| 11 | 1.461653229588009 | 0 | 479 |
| 11 | 0.473643012711471 | -0 | 9226 |
| 11 | 5.342132771456683 | 1 | 54 |
+--------+----------------------+-----------+---------+[/code](yes, I know there's a weirdness with -0, ignore that for the purpose of this discussion).

For the P-1, yes, that is pretty much what I imagined.

For the TF stats, I understand if it's a lot of work. But at the same time, it would make a few people happy, including me. So I'll leave it entirely up to you whether you decide to make it or not.

Please re-explain what you're looking for with regards to TF, preferably with a simple example (make up numbers is fine, I just don't have a good grasp of what kind of data you're looking for, and/or what layout you envision.

[QUOTE=James Heinrich;573454]Please re-explain what you're looking for with regards to TF, preferably with a simple example (make up numbers is fine, I just don't have a good grasp of what kind of data you're looking for, and/or what layout you envision.[/QUOTE]
Same as the "GIMPS progress visualization" e.g. [URL="https://www.mersenne.ca/status/"]here[/URL], but the exponent counts will be only from the already factored exponents, and there could be perhaps a column containing PRP and fully factored exponent count in the given range(s).

There is a lot of incomplete data in that query. As far as GIMPS is concerned once a factor is known the exponent is no longer interesting, so for a large number of exponents there is little to no data about what factoring effort was done, especially for factors found more than a few years ago. Taking a quick-rough look at 10M range for example I see ~10k exponents with known TF depth and ~30k unknown.

I could probably come up with a report that shows the numbers that are known, with the strong caveat that these numbers are incomplete (and possibly inaccurate) for factored exponents. Something like this:[code]Range 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77
0M 14547 477 272 316 87 58 43 1 2 2 0 0 0 0 0
1M 19236 345 722 901 437 430 16 47 421 1 10 0 1 1 1
2M 13063 454 715 638 465 1206 5 57 27 4 2 0 1 3 0
3M 10550 295 513 350 267 593 94 66 10 1 0 0 0 0 0
4M 9659 351 346 528 132 182 15 31 24 14 0 0 0 0 1
5M 8812 303 401 129 242 541 179 73 48 8 2 1 0 0 0
6M 7416 236 346 54 180 488 160 59 24 1 0 0 0 0 0
7M 7282 271 238 65 131 190 208 51 26 1 9 0 0 0 0
8M 7117 241 258 56 66 215 170 32 3 2 1 0 0 0 0
9M 6681 159 239 50 68 141 175 41 29 5 0 0 4 0 2
10M 9053 217 403 147 131 89 138 382 8 0 0 0 0 1 0
11M 6971 172 361 180 149 68 256 17 3 23 0 0 0 0 0
12M 5899 154 228 88 230 119 251 15 11 0 0 0 0 0 0
13M 5474 138 184 19 89 17 52 10 0 1 0 0 0 0 0
14M 5891 147 157 2 2 0 27 14 105 7 1 0 0 0 0
15M 5620 147 148 6 6 1 73 90 1 0 0 0 1 0 0
16M 5288 152 126 4 2 5 101 150 21 5 0 0 0 0 0
17M 6057 172 171 7 1 0 160 96 3 0 1 0 0 0 0
18M 5556 147 161 5 3 5 183 115 3 0 0 0 0 0 0
19M 5028 138 139 8 4 2 188 150 25 10 0 0 0 0 0
20M 2467 75 87 1 1 0 5 64 43 19 0 0 0 0 0
21M 2288 57 57 3 0 0 0 17 28 20 6 0 0 0 0
22M 2466 66 57 2 3 1 0 11 23 7 0 0 0 0 0
23M 1103 31 31 0 0 0 0 14 2 0 0 0 0 0 0
24M 1166 28 29 0 1 0 0 11 0 0 0 0 0 0 0
25M 786 14 28 2 1 1 0 3 3 0 0 0 0 0 0
26M 505 21 15 0 3 0 0 13 5 2 0 0 0 0 0
27M 565 13 17 1 0 0 0 10 23 31 0 0 0 1 0
28M 494 17 23 1 1 1 0 15 63 0 0 0 0 1 0
29M 583 21 22 1 1 1 0 44 156 120 47 0 0 0 0
30M 1044 21 28 0 2 0 0 0 199 221 218 129 31 2 0
31M 1056 25 34 1 2 0 0 0 183 226 200 25 15 0 1
32M 999 21 15 1 1 0 0 0 176 270 122 34 15 4 0
33M 974 22 26 0 0 0 0 0 176 248 157 34 15 0 0
34M 816 20 20 1 3 1 0 0 156 224 149 6 51 0 0
35M 906 27 21 2 2 0 0 0 121 164 115 3 18 3 0
36M 1033 21 26 2 0 0 0 0 97 167 81 13 0 0 0
37M 1275 31 28 2 0 0 0 0 67 142 106 28 11 0 0
38M 1464 15 26 0 2 0 1 0 39 133 131 20 69 5 0
39M 740 19 22 0 2 0 0 0 6 291 141 31 26 3 0[/code]Is that kind of what you were thinking of?

Yes, that's what I am looking for.

I think the problem with no TF data can be solved by trusting the user TJAOI, that there are no more factors under 67 bits (technically he is at 66.990, but in a week or two he is done). Maybe some check-box "I trust TJAOI" could be made, such that when checked, considers all exponents to be TFed to at least 67 bits (or the current TJAOI-TF integer bit-level achieved), and when not, takes what is in the database. Or no check-box with a disclaimer that data is based on TJAOI's work.

[QUOTE=Viliam Furik;573547]Yes, that's what I am looking for.[/QUOTE]It's perhaps not terribly pretty or GUI-fancy like the normal [url=https://www.mersenne.ca/status/]status page[/url], but I can do this with minimal fiddling with the database, you'll get a new file generated every day at 0100h UTC.
[url]https://www.mersenne.ca/factored_counts_report_tf/[/url]
You can automatically retrieve the most recent report at [url]https://www.mersenne.ca/factored_counts_report_tf/?latest=1[/url]
Past data (not before 2021-03-12) can be accessed using obvious URL edits.

[QUOTE=James Heinrich;573554]It's perhaps not terribly pretty or GUI-fancy like the normal [url=https://www.mersenne.ca/status/]status page[/url], but I can do this with minimal fiddling with the database, you'll get a new file generated every day at 0100h UTC.
[url]https://www.mersenne.ca/factored_counts_report_tf/[/url]
You can automatically retrieve the most recent report at [url]https://www.mersenne.ca/factored_counts_report_tf/?latest=1[/url]
Past data (not before 2021-03-12) can be accessed using obvious URL edits.[/QUOTE]

Thank you! It's great and has pretty much everything necessary. As the saying goes, don't judge the book by the cover -> it's more important for a table to have correct data than to look pretty.

:awesome: