SoL said:
Ok I can't explain to you why 948 beats 1066 speed because I don't know crap about RDRAM. But thats besides the point, you didnt answer my question. You said that over 155 you get bad results did you not? So my 3159 MB/s at 155 is right or wrong? If you say its right then why wouldn't me overclocking more give me more memory bandwith? You're confusing the hell out of me and probably everyone else reading this. Your computer not giving you the results that you want doesn't have anything to do with what people are getting with DDR. And btw at 2.76ghz I get 3281MB/s. Thats 59 MB/s off PC1066 So says Sandra. What are you getting for PCMark2002 scores?
I'm sorry that I'm confusing you.
I'm not the only one that thinks that sandra needs a patch.
let me please repeat my self.....why does an AMD system need 180fsb just to hit pc800 rdram speeds and an intel needs less???
are you saying that at 160 fsb a intel based DDR system has better memory bandwith then an AMD at 200fsb????
the results I get are on my mobo....are you going to tell me that all thies numbers make sence to you???
I truely can't say that the problem starts at 155fsb....it could start earlyer.
I know you say that you dont understand RDRAM...yet how can 948 speed beat 1066 speed???
this is like saying that pc2700 DDR is faster then pc3200 DDR....
now the real question is what are you getting in real world results.
Q3A,UT,comanchie4 demo,pcmark,3dmark and so on..........
start to compair and you will see that the sisoft sandra is not the end all be all test at anything........and can be flat out wrong sometimes.
maybe some one else could explain this better then I.
mica