The Next Big Development for GIMPS

[kruoli]

When workers are testing wildly different numbers, such as ECM on a small
Mersenne number in one window and an LL test on a 100 million digit number
in another window, then the "iterations between screen outputs" setting will
cause the windows to output at wildly different rates.  You can workaround
this in prime.txt:
	ScaleOutputFrequency=0 or 1	(default value is 0)
If you set this to 1, then "iterations between screen outputs" will be
automatically scaled so that windows produce output at roughly the same rate.
An LL test on M50,000,000 will not scale at all, ECM on a small number will
scale to a much larger value and an LL test on a 100 million digit number
will scale to a much smaller value.

That's what I'm using and it helps a lot. Of course, this alone cannot be perfect, but in this case I can live without perfection.

[chalsall]

George... Where can I find the latest version you want to be tested, Linux mprime?

I've got a machine I want to "prove the sanity of" before deploying next week.

Thanks (this is so cool!!! )

[kriesel]

Originally Posted by chalsall

Where can I find the latest version you want to be tested, Linux mprime?)

Until George puts a later one out, v30.1b1 is here

[chalsall]

Originally Posted by kriesel

Until George puts a later one out, v30.1b1 is here

Yeah, thanks. But I was wondering if the version which does the MD5 checksum was ready for testing?

Even beyond the Proxy issue, I've had at least one case where a direct comms CERT run still failed to match. On known good kit.

[Prime95]

v30.2 is imminent. I'll post links here.

[Prime95]

Windows 64-bit: https://www.dropbox.com/s/iz417i8hpt...win64.zip?dl=0
Linux 64-bit: https://www.dropbox.com/s/q4haz69wti...64.tar.gz?dl=0

This is much closer to a beta version. I haven't tested it as much as I should.

This version can upload proof files. There is a new Resource Limits menu choice / dialog box. LL work preferences are gone, if you upgrading, your LL work preferences will be converted to PRP.

Please review the overhauled readme.txt.

Suggestions for further changes are welcome. I have a few ideas I'll put forward soon.

[firejuggler]

About to finish my first 'proof' file. hope it will help. As I upgraded to 30.2 I do not have to manually upload the proof anymore, right?

[chalsall]

Originally Posted by Prime95

This is much closer to a beta version. I haven't tested it as much as I should.

Thanks. Now running on a 6-core E-2136 (the machine's name is "rdt1", if you have any use for seeing the server-side of things).

[Prime95]

Originally Posted by firejuggler

About to finish my first 'proof' file. hope it will help. As I upgraded to 30.2 I do not have to manually upload the proof anymore, right?

Correct. 30.2 will upload any proof files it finds in the directory containing the executable.

[firejuggler]

I guess that's it? And since it is a DC there is no need to upload the proof?

[Prime95]

Paging Oliver:

Did anything strange happen during the run of 10471007? The proof failed certification. Do you have the proof file by chance?