Feedback for new MPQS utility sought

[jasonp]

Originally Posted by BlisteringSheep

What has really surprised me is the poorer performance for gcc-4.1.1, both on the Xeon and the PPC970. With sr2sieve, gcc-4.1.1 gives much better performance, especially on the PowerPC.

The PPC970 machine I have access to is a dual 2GHz running OS X 10.4; on the 76-digit input, msieve compiled with gcc 4.1.1 needs 7:51 and compiled with gcc 3.4.6 needs 7:48, which isn't close to the times you're seeing.

On a dual 2GHz opteron running FC2, msieve compiled with gcc 4.1.1 needs 5:17 and compiled with gcc 3.4.6 needs 7:18. This is really strange, my tests reach the opposite conclusion that yours do.

Could the fact that these are dual-processor machines have anything to do with the difference? The RAM has to get clocked slower to handle the cache coherency protocol, and neither of my machines has high-performance RAM to begin with...

jasonp