On 11/4/05, Peter Moulder <Peter.Moulder@...38...> wrote:

On Thu, Nov 03, 2005 at 11:06:57AM +0100, Ralf Stephan wrote: Update: Have made some more progress today, with Michael's help.

Apparently there is a bug in gcc 4.0 on ppc with -O2: even a simple multiplication of `(double) 0xb28 * 0x66' (which should be exact even with float or unsigned arithmetic) gets the wrong answer.

Do you have further info such as a URL for this bug? Google gives tens of thousands of results for likely keywords.

On the basis that exchange is no robbery, this URL might be of interest to people wanting some background on how to choose floats or ints. http://ridiculousfish.com/blog/?p=24

Ben