Date: Wed, 4 Oct 95 13:43:05 +0100 To: djgpp AT sun DOT soe DOT clarkson DOT edu From: til AT zentrum DOT phys DOT chemie DOT tu-muenchen DOT de (Tilmann Haeberle) Subject: Re: Bugg on Pentium HP vectra XU > > > [...] > > On other pentiums, everything is allright! > > Using BC to compile gives good result! > > > > So why??????????? > > > >Hmmm. Isn't there a flag in gcc-2.6.3 called >"-fI_am_a_boken_pentium" ? >Don't know if this has to the with the pentium fdiv bug, >but it might be worth thinking of. FDIV bug should behave always the same. IMHO the / operator delivers a double (not float) result, so you should try #include void main() { printf("%f\n",(float) (1024.0f/378.0f)); } ______________________________________________________________________ Tilmann Haeberle e-mail: til AT zentrum DOT phys DOT chemie DOT tu-muenchen DOT de Institut fuer physikalische Chemie I Tel: 089/3209-3373 TU Muenchen Lichtenbergstr. 4 Fax: 089/3209-3026 D-85747 Garching, Germany