----- Original Message ----- From: "Vincent Trouilliez" <[EMAIL PROTECTED]>
> > There's for sure a lot of potential for improvement, but declaring > > something "top-priority" because that's your opinion won't do > > anything. There are many similar minor improvements that could be > > made, they probably sum up to maybe 10...20 % of possible savings if > > really *all* of them were sorted out > > Really ? Sorry didn't realise that the H/W multiplier was of so little > benefit when doing calculations... maybe Atmel could just remove it from > their chip to sell them cheaper and/or free the space for some more > useful peripheral then... :-) > What could they replace it with mmmmm.... a few D/A converters would be > fine... :-) > The hardware multiplier makes a difference when doing mulitplication, which take a very small percentage of the run time of most real programs - therefore, overall, it's a small issue. Obviously the hardware multiplier is a lot faster than a software routine at the actual multiply, it's just that it makes little difference to most programs as a whole. _______________________________________________ AVR-GCC-list mailing list [email protected] http://lists.nongnu.org/mailman/listinfo/avr-gcc-list
