Newbie type questions - ANONYMOUS user etc

Hi there - just returned to GIMPS after a couple of years out. Seem to have mprime v25.9 running OK on SuSE 11.1.

3 dumbo issues:
1) While I tried to give my primenet userID as a parameter when asked, it seems to have got lost in the shuffling I did (setting up as a service etc). To the extent prime.txt now has
V5UserID=ANONYMOUS
I cant find the simple way to fix this (have tried mprime -m etc) - can I just edit prime.txt?

2) ditto tried to config memory available as 128Mb, but local.txt still has
Memory=8 during 7:30-23:30 else 8
(I have 4Gb of ram and not much else going on so can afford to be generous!) Again should I just edit local.txt or is there a more appropriate way?

3) Is there a description of the phases mprime goes through somewhere? At first my machine spent about 2 days climbing up to 100%, except instead of finishing it is now doing Optimal P-1 factoring on the same work units. As this seems to be progressing at around 1%/day it would be interesting to know how much further I have to go before getting some movement on the producers list!

Thanks

[joblack]

Originally Posted by dukla2000

Hi there - just returned to GIMPS after a couple of years out. Seem to have mprime v25.9 running OK on SuSE 11.1.

3 dumbo issues:
1) While I tried to give my primenet userID as a parameter when asked, it seems to have got lost in the shuffling I did (setting up as a service etc). To the extent prime.txt now has
V5UserID=ANONYMOUS
I cant find the simple way to fix this (have tried mprime -m etc) - can I just edit prime.txt?

2) ditto tried to config memory available as 128Mb, but local.txt still has
Memory=8 during 7:30-23:30 else 8
(I have 4Gb of ram and not much else going on so can afford to be generous!) Again should I just edit local.txt or is there a more appropriate way?

3) Is there a description of the phases mprime goes through somewhere? At first my machine spent about 2 days climbing up to 100%, except instead of finishing it is now doing Optimal P-1 factoring on the same work units. As this seems to be progressing at around 1%/day it would be interesting to know how much further I have to go before getting some movement on the producers list!

Thanks

If prime95 is runing you can change the file but if you kill the process it will written back. So check out if it is running, kill it and then edit the parameters in the file (don't forget to make an account at mersenne.org before that).

[cheesehead]

Originally Posted by dukla2000

3) Is there a description of the phases mprime goes through somewhere?

AFAIK, it's the same phases prime95 goes through:

1) trial factoring, if TF hasn't already been done up to the default limit for that exponent's range

2) P-1, if that hasn't already been done

3) L-L

At first my machine spent about 2 days climbing up to 100%

That would have been the TF. Apparently your exponent had not already had TF done to the default limit.

except instead of finishing it is now doing Optimal P-1 factoring on the same work units.

... because, apparently, your exponent had not previously had the P-1 done.

As this seems to be progressing at around 1%/day it would be interesting to know how much further I have to go before getting some movement on the producers list!

1) P-1 at 1% per day? That's pretty slow for exponents in the 40M-60M range. How fast is your CPU? Is your exponent larger than 100 million?

Or do you mean that the L-L test has started after P-1 completion? 1%/day is okay for the L-L test.

2) No one gets credits (e.g., on the producers list) for work until it has been reported as completed (i.e., 100%).

[Uncwilly]

Originally Posted by cheesehead

1) P-1 at 1% per day? That's pretty slow for exponents in the 40M-60M range. How fast is your CPU? Is your exponent larger than 100 million?

Sounds like it to me (100m)

Many thanks for the responses. Have got my options sorted (I think): stopped mprime and played with the options under mprime -m. Must have been blind previously. Have noticed with more memory 1 "Worker" has started processing relative primes.

Originally Posted by cheesehead

1) P-1 at 1% per day? That's pretty slow for exponents in the 40M-60M range. How fast is your CPU? Is your exponent larger than 100 million?

Got an AMD 4850e which is dual core 2500Mhz, but most of the time with CnQ kicked in both cores are 1000MHz.