mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   PrimeNet (https://www.mersenneforum.org/forumdisplay.php?f=11)
-   -   Joke of a manual report (https://www.mersenneforum.org/showthread.php?t=18643)

bloodIce 2013-10-01 22:12

Joke of a manual report
 
From my manual report:
"Factors found: 1
Processing result: M2000153 has a factor: 941867337385320490849 [TF:69:70*:mfaktc 0.19 barrett76_mul32]
Insufficient information for accurate CPU credit.
For stats purposes, assuming factor was found using ECM with B1 = 50000.
CPU credit is 0.0846 GHz-days."

This is just a joke, isn't it?

chalsall 2013-10-01 22:14

[QUOTE=bloodIce;354836]This is just a joke, isn't it?[/QUOTE]

Sort of...

bloodIce 2013-10-01 22:28

The programmers are busy... again...
 
How many GHzYears will need to pass, until GIMPS server is able to catch a mfaktc report. The same, all over again... What more information is needed? You see mfaktc in a string, then you subtract the factor bit length from the current bit level and multiply by a coefficient.

The only explanation I see in this forum is that the programmers are busy with other things? Probably true, but come on, for "a one liner"? Are those programmers in North Korea?

kladner 2013-10-01 22:35

[QUOTE=bloodIce;354839]How many GHzYears will need to pass, until GIMPS server is able to catch a mfaktc report. The same, all over again... What more information is needed? You see mfaktc in a string, then you subtract the factor bit length from the current bit level and multiply by a coefficient.

The only explanation I see in this forum is that the programmers are busy with other things? Probably true, but come on, for "a one liner"? Are those programmers in North Korea?[/QUOTE]

The submission system gets confused if an mfaktc report begins with a Factor Found line. Make sure to always lead with a No Factor line, even if it has been previously submitted. Alternatively, use MISFIT, which will not submit a factor first.

chalsall 2013-10-01 22:49

[QUOTE=kladner;354841]The submission system gets confused if an mfaktc report begins with a Factor Found line. Make sure to always lead with a No Factor line, even if it has been previously submitted. Alternatively, use MISFIT, which will not submit a factor first.[/QUOTE]

But... bloodice raises an important point.

All the information needed for the correct interpretation is presented to the submission script.

It should take no more than one hour to fix this issue. And it's been an issue for quite some time.

Why should one know (or use scripts which know) how to submit the results so they'll be interpreted properly? ("Oh, didn't you know? You need to submit a NF before a F, silly!")

PrimeNet should know already....

Prime95 2013-10-01 23:52

Maybe James, the PHP pattern matching guru, can cobble together a fix.

swl551 2013-10-02 00:22

[QUOTE=chalsall;354844]But... bloodice raises an important point.

All the information needed for the correct interpretation is presented to the submission script.

It should take no more than one hour to fix this issue. And it's been an issue for quite some time.

Why should one know (or use scripts which know) how to submit the results so they'll be interpreted properly? ("Oh, didn't you know? You need to submit a NF before a F, silly!")

PrimeNet should know already....[/QUOTE]
Well said. Solve the root problem and all the workarounds and frustration goes away.

kladner 2013-10-02 00:32

[QUOTE=chalsall;354844]But... bloodice raises an important point.

All the information needed for the correct interpretation is presented to the submission script.

It should take no more than one hour to fix this issue. And it's been an issue for quite some time.

Why should one know (or use scripts which know) how to submit the results so they'll be interpreted properly? ("Oh, didn't you know? You need to submit a NF before a F, silly!")

PrimeNet should know already....[/QUOTE]

Probably so. Is not my department. Stating the known work around is more my speed, though without taunts. :no:

davieddy 2013-10-02 00:32

[QUOTE=chalsall;354837]Sort of...[/QUOTE]
Drat.
You stole my line.

chalsall 2013-10-02 00:43

[QUOTE=davieddy;354860]Drat. You stole my line.[/QUOTE]

Get used to it.

bloodIce 2013-10-02 07:53

[QUOTE=kladner;354841]The submission system gets confused if an mfaktc report begins with a Factor Found line. Make sure to always lead with a No Factor line, even if it has been previously submitted.[/QUOTE]

However, I did not have "No Factor lines". The reason is that I submitted them a day before (well, to submit them again you say :smile:, logical). The next day (which is reasonable for such assignment) I decided to report a factor. As pointed above, a user should not keep such details (first that, then that, if not then that) in mind all the time. We already have all information we need in my report to assign a proper method, and credit it.

P.S. Sorry, it seems that I repeated things, which are already said by chalsall.[URL="http://www.mersenneforum.org/member.php?u=144"][COLOR=red][/COLOR][/URL]

kladner 2013-10-02 15:45

The situation is not ideal, for certain! It's just one of those things that hasn't gotten fixed.

James Heinrich 2013-10-02 16:23

I'll see if I can take a look sooner rather than later and rewrite the manual results parsing.

LaurV 2013-10-02 16:37

MISFIT is automatically taking care of this situation, changing the lines in the result file in the proper way. Unless James have time to provide a fix...

kracker 2013-10-02 16:37

If it ain't broke don't fix it(NOT) :razz:

bloodIce 2013-10-02 20:59

[QUOTE=James Heinrich;354960]I'll see if I can take a look sooner rather than later and rewrite the manual results parsing.[/QUOTE]

Thank you James. I am sure it would be worthed at the end. Imagine how many threads like this one you would prevent to happen :-).

swl551 2013-10-02 23:31

humans are so odd sometimes
 
This bug has been causing grief for years and dozens of posts have been written with countless tales of woe and no one cared enough to fix it. Then this one thread comes along and the wand is waived and people suddenly respond.

BloodIce you should run for office because when you speak... people listen. YOU HAVE MY VOTE.:bow::bow::bow:

Prime95 2013-10-03 02:36

[QUOTE=swl551;355014]This bug has been causing grief for years and dozens of posts have been written with countless tales of woe and no one cared enough to fix it.[/QUOTE]

That is not true. Originally, there was no mfaktc support in manual reports. Since mfaktc only output the string "mfaktc" on no factor lines, the current solution was put in place -- if any no factor line containing mfaktc was found then it was assumed all factors were found by mfaktc. This was fix #1.

Fix #2 was the change to mfaktc to output the string "mfaktc" on factor found lines.

We are now awaiting James fix #3 to make use of fix #2.

Admittedly, progress has been quite slow. Sorry.

bloodIce 2013-10-03 08:43

[QUOTE=swl551;355014]BloodIce you should run for office because when you speak... people listen. YOU HAVE MY VOTE.:bow::bow::bow:[/QUOTE]

Not me, if James fixes it, he will be promoted. He already has my vote :-).

cheesehead 2013-10-04 05:32

[QUOTE=swl551;355014]This bug has been causing grief for years and dozens of posts have been written with countless tales of woe and no one cared enough to fix it. Then this one thread comes along and the wand is waived and people suddenly respond.[/QUOTE]Sometimes, when the wheel squeaks, it's just not as urgent, right then, as the loud banging and thumping from the smoking motor ...

Later, when all else is quiet, the squeak happens again, and the mechanics say, "Hey, remember that squeak? Let's fix it this time." It's not that the squeak was more persuasive that time.


All times are UTC. The time now is 23:18.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2022, Jelsoft Enterprises Ltd.