v24.13 release candidate 1
 

Version 24.13 RC1 is ready.

1) Better Celeron D timings.
2) Fixes all reported trial factoring bugs on big exponents.
3) Error 2252 fixed for some users.
4) AMD64 torture test now runs both prefetchw and non-prefetchw versions of the FFT. Anecdotal evidence suggests that version 23's non-prefetchw FFTs were tougher on the CPU.
5) Prime.log file now limited to 2M bytes rather than 250K.
6) Fixed crash bug when P-1 stage 2 didn't have enough memory and an LL save file already existed.

See [url]http://www.mersenneforum.org/showpost.php?p=56866&postcount=6[/url] for download links.

Is this a beta version or an actual version?

At [url="http://www.mersenne.org/gimps/"]http://www.mersenne.org/gimps/[/url] the [url="http://www.mersenne.org/gimps/p95v2413.exe"][color=#0000ff]p95v2413.exe[/color][/url] file is only 27 bytes long.

Bug?
 

When running the Linux ver. 24.13 with this in worktodo.ini

Pfactor=55459,2,9449770,1,49,1.6

the program stall and then on kill exits on with this statement

[1]+ Floating point exception./mprime -d

It runs fine with ver. 24.12.

Anyhow though I just wanted to report this ... I going on with my factoring for SoB with ver. 24.12.

Thank you for all your hard work George.

Frodo42, I'll have a fix later today.

Are these newer versions of Mprime safe to use for day to day testing?

[QUOTE=mathemajikian]Are these newer versions of Mprime safe to use for day to day testing?[/QUOTE]

Yes. If no serious bugs are found 24.13 will become the official version.

This is with mprime2413.tar.gz

[CODE]mv@debian:~/Desktop/temp$ ./mprime -m
./mprime: error while loading shared libraries: libstdc++.so.6: cannot open shared object file: No such file or directory[/CODE]

I got the same problem with 24.12 RC2 as well. I downloaded the statically linkes version and it worked fine. My machine has the latest kernel but it seems to only have libstdc++5.0.7

For all those FreeBSD users in the world like myself to make 24.13 run just download the sprime version of this release and do:

# brandelf -t Linux mprime

of course you have to have linux emulation enabled. How to enable linux emulation can be found here:

[url]http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/linuxemu.html[/url]

OK I don't know if anyone else is getting performance improvements as I am, but I'm getting around an 8%+ increase in speed on this P4 that I'm posting my benchmarks on here. Wait until you see my Sempron laptop, though (I'll let you calculate it yourself :banana::banana::banana::banana:)

This computer has a 512Mb and a 256Mb ram, don't know what kind or speed, but I think 133Mhz.


[code]
Intel(R) Pentium(R) 4 CPU 2.40GHz
CPU speed: 2393.98 MHz
CPU features: RDTSC, CMOV, PREFETCH, MMX, SSE, SSE2
L1 cache size: 8 KB
L2 cache size: 512 KB
L1 cache line size: 64 bytes
L2 cache line size: 128 bytes
TLBS: 64
Prime95 version 23.8, RdtscTiming=1
Best time for 384K FFT length: 15.781 ms.
Best time for 448K FFT length: 18.723 ms.
Best time for 512K FFT length: 21.454 ms.
Best time for 640K FFT length: 25.970 ms.
Best time for 768K FFT length: 31.410 ms.
Best time for 896K FFT length: 36.911 ms.
Best time for 1024K FFT length: 41.689 ms.
Best time for 1280K FFT length: 54.527 ms.
Best time for 1536K FFT length: 65.898 ms.
Best time for 1792K FFT length: 78.910 ms.
Best time for 2048K FFT length: 90.543 ms.
[Sun Jul 10 16:06:49 2005]
Compare your results to other computers at http://www.mersenne.org/bench.htm
That web page also contains instructions on how your results can be included.
[/code][code]
Intel(R) Pentium(R) 4 CPU 2.40GHz
CPU speed: 2394.13 MHz
CPU features: RDTSC, CMOV, Prefetch, MMX, SSE, SSE2
L1 cache size: 8 KB
L2 cache size: 512 KB
L1 cache line size: 64 bytes
L2 cache line size: 128 bytes
TLBS: 64
Prime95 32-bit version 24.13, RdtscTiming=1
Best time for 512K FFT length: 18.890 ms.
Best time for 640K FFT length: 24.375 ms.
Best time for 768K FFT length: 30.051 ms.
Best time for 896K FFT length: 34.774 ms.
Best time for 1024K FFT length: 39.263 ms.
Best time for 1280K FFT length: 50.691 ms.
Best time for 1536K FFT length: 62.606 ms.
Best time for 1792K FFT length: 72.740 ms.
Best time for 2048K FFT length: 81.712 ms.
Best time for 2560K FFT length: 107.720 ms.
Best time for 3072K FFT length: 137.351 ms.
Best time for 3584K FFT length: 159.894 ms.
Best time for 4096K FFT length: 184.159 ms.
Best time for 58 bit trial factors: 12.161 ms.
Best time for 59 bit trial factors: 12.063 ms.
Best time for 60 bit trial factors: 12.218 ms.
Best time for 61 bit trial factors: 12.299 ms.
Best time for 62 bit trial factors: 13.170 ms.
Best time for 63 bit trial factors: 13.175 ms.
Best time for 64 bit trial factors: 16.291 ms.
Best time for 65 bit trial factors: 16.230 ms.
Best time for 66 bit trial factors: 16.246 ms.
Best time for 67 bit trial factors: 16.234 ms.
[/code]