-----Original Message----- From: Toon Knapen [mailto:toon.knapen@fft.be]

...

Thanks for the update Toon. I think we're going to migrate to the gcc compiler on HP-UX for all our future work. That or we're going to go with Linux and bag HP altogether.

OK but nevertheless some related info : HP says that gcc only obtains 1/3th of the performance of aCC (never checked myself though). And you should also now that I tried to compile boost with gcc (oops it's not shown anymore on the status pages, I'll have to update it) and I had trouble compiling it (mostly to do with posix threads AFAICT).

Not sure where that performance statistic comes from, but I didn't really notice it when we went from aCC to gcc a while back (but that was on 10.20, acc 1.18 -> gcc 2.9x). I'd recommend you use at least gcc 3.3 if you are going to go to gcc. There are a few big bugs. On 3.2.x we've had to drop to -O1 due to what appear to be optimizer bugs. GCC says they fixed some of those with 3.3, though I haven't had the chance to verify. In any case, back when Jens Maurer and I were trying to get Boost on aCC going, we were told Boost was going to become part of their test suite, but the lack of progress seems to indicate that didn't happen. And, if you have the opportunity, run as fast as you can away from HP to Linux. When Douglas Schmidt singles out a platform as being hard to work with (can't remember the exact reference, but it wasn't complimentary), you know its a pain. ----------------------------------------------------------------------- DISCLAIMER: Information contained in this message and/or attachment(s) may contain confidential information of Zetec, Inc. If you have received this transmission in error, please notify the sender by return email. -----------------------------------------------------------------------