25.3
 

Hello. I just switched from distributed.net to mersenne.org and I really like this project.

About 25.3 I like the multi threading. It's working well, as far as I can tell, on both my Core 2s and HT Prescotts.

I like the idea of having an SDI that can't be closed. The message sources could be differentiated by labels or colors. Alternatively, a &View menu with the child windows listed could help users accustomed to using MDIs.

If all the child windows are closed, how does one resume testing? Maybe I'm overlooking something here.

Thanks.

[QUOTE=Prime95;111269]In Options/CPU dialog box, has prime95 properly identified your machine as having 4 CPUs?[/QUOTE]

Yes, it identifies four cores. However, the clock is not detected correctly though.
I think the explanation by drew solves my issues. I've tested to switch to any COMBINATION (1x4, 2x2, 4x1) to setup four cores and it works. Thanks, drew!

-D

I get this error often when sending results or expected completion dates. What does it mean?
[code]
pnResult=3
PrimeNet error 20003: Unknown error code[/code]

[QUOTE=S00113;111334]I get this error often ... pnResult=3[/QUOTE]

That means the v4 server database is down or the exponent is not assigned to your computer.

[quote=Prime95;111341]That means the v4 server database is down or the exponent is not assigned to your computer.[/quote]
Do the servers talk much? 42851911 is in status.txt on the old server, but the factoring work was completed and the exponent returned on the new server with 25.3. Another exponent which give this error has probably expired and got reissued to another user on the old server. Is it safe to move assignments from a v24 worktodo.ini to a v25 worktodo.txt, or should I avoid this?

[QUOTE=S00113;111345]42851911 is in status.txt on the old server, but the factoring work was completed and the exponent returned on the new server with 25.3. Is it safe to move assignments from a v24 worktodo.ini to a v25 worktodo.txt, or should I avoid this?[/QUOTE]

The servers do not talk to each other. v25.3 should only be used with the old server for now. It is safe to move assignments from v24 worktodo to v25 worktodo as long as you talk to the v4 server.

[QUOTE=Bundu;111099]Also, having installed this version over a 24.14 version i copied onto my brother's new dual core laptop i noticed that when I increased the number of worker threads this new version returns the first thread back to the start of its TF.. oops
[/QUOTE]

I cannot reproduce this bug. Can you try again and if successful, document the exact steps you took and send me any save files I'll need to reproduce the problem? Thanks.

[QUOTE=AES;111275]I like the idea of having an SDI that can't be closed. The message sources could be differentiated by labels or colors. Alternatively, a &View menu with the child windows listed could help users accustomed to using MDIs.[/QUOTE]

Ugh. I just looked into how to make an MFC application switch between MDI and SDI. It looks real hard.

While I think this is a good idea, it is a cosmetic change. I'm postponing any further work on it until all core functionality is working properly.

[QUOTE=Prime95;111370]I just looked into how to make an MFC application switch between MDI and SDI.[/QUOTE]
My two cents...I don't think it's such a fantastic idea. You'll never please everyone with a UI, and there's no sense trying to adapt to everyone's tastes. Especially with a single-purpose, functional application like this.

Just pick one configuration. People will use it the same way whether there's 1 window or 4. After all, most people will probably leave it running in the system tray and ignore it altogether.

I've been running p-1 for the Seventeen or Bust project, so ALL my work involves p-1. After reading this thread, I think that allocating 1200MB for p-1 is WAY too much, so I'll change it. But related to that, I have a request to make.

It would be nice if the Status Report could be made to tell someone whether or not Stage 2 is done right after Stage 1, or if there is a delay. The reason for this is if Stage 1 takes longer than Stage 2, and Stage 2 has to wait for available memory, Stage 1 files will build up as Stage 2 falls behind. Also, if people want to run a different project, it's difficult to tell when there will be a situation where all Stage 1s have been done and, because of a Stage 2 RAM situation, one or more cores is(are? I hate grammar) free.

Sometimes a worker thread window fails to report a line of progress until the next line is displayed. Two lines will appear at the same time, instead.

I've seen this while trial factoring and in P-1 (I forget which stage).