Slartibartfast 2003-10-08 05:43:33 | Just as valid as a browser war, I guess. I could be way off here, but it looks as the usual holds true again for Muon1. That an Athlon XP outperforms a P4 about 2:1 every clock cycle This has so far proved true for United Devices, D.Net RC5, ClimatePrediction, F-a-D, Seti, F@H, G@H. Only project that the P4 came out better per MHz was GIMPS. My observation however, I'm not sure if anyone else compared chips on various projects and came up with the same results? |
abyzz 2003-10-26 00:49:18 | Hi there, for about a week i have 3 machines running MUON1 and it seems that the AMD outperforms Intel. The 3 machines consists of: AMD XP2200+ (~1800 MHz - no OC) P4 2.4 GHz P4 2.0GHz mobile ...all with plenty of memory The AMD deliverys more results pr. time than the Intel cpu's. I made the same experiment on my SETI-accout and here the picture was the same. AMD Rules. BR ABYSS |
Stephen Brooks 2003-10-26 02:58:54 | I've had a look into this recently, and the advantage is that AMD's native floating-point performance is much better than Intel's. With integer arithmetic it is the other way around. Also, if I compiled with SSE2 extensions and fiddled around for ages optimising my code for those instructions, Intel chips be faster on floating-point, but I think what they failed to realise is that the vast majority of developers will not go to that amount of effort unless they're writing a DivX player or something. It doesn't make any sense: that's why they call it "virtual" |