aGeoM Posted June 5, 2010 Posted June 5, 2010 Finaly I got the NB Frenquency working on this board, it was a little tricky at the beginning, but now I know how to do it, easely. The BIOS of my M3A79-T doesn´t alow me to set HT ref up to 210, so I had to use the old stuff. Quote
Crew Don_Dan Posted June 5, 2010 Crew Posted June 5, 2010 Damn, what a huge score, now Intel will never have a chance at the MaxxMem DDR2 record again. -.- Congrats, really well done! :-) Quote
Calathea Posted June 6, 2010 Posted June 6, 2010 Wow, amazing. Didn't think my score could be beaten on air... Oh and try bios 1102 or 1203 on your M3A79-T! Quote
aGeoM Posted June 7, 2010 Posted June 7, 2010 Hi, your score could be higher if you use a non L3 cache CPU, like Athlon II at the same settings you used for the 965. Thanks for the tip. Quote
Massman Posted June 10, 2010 Posted June 10, 2010 I'm starting to think this score might be bugged. The latency is very low in comparison to Chew*'s scores (which is a phenomenon occuring with bugged runs). Can you try V1.82? It may just be that V1.74 doesn't support Regor well enough. Quote
aGeoM Posted June 10, 2010 Posted June 10, 2010 (edited) Latency is achieved with memory timings, and yes I will try another version in another board if ASUS engenniers pull out a decend BIOS for M3A79-T at time (refuses to post @210 HT ref w/ Sempron), with this M2N32 board v.1.82 doesn't run, CPU-Z it's eating all resources and got stuck, I can't even close the process. Need to reboot. v1.78 it's fine. My DDR and DDR3 results are with v.1.82. Edited June 10, 2010 by aGeoM Quote
chew* Posted June 10, 2010 Posted June 10, 2010 massman, no need to jump to conclusions. I think the key diff may be 64 bit os choice or platform choice which we both know can be considered a tweak. Quote
aGeoM Posted June 10, 2010 Posted June 10, 2010 I'm starting to think this score might be bugged. The latency is very low in comparison to Chew*'s scores (which is a phenomenon occuring with bugged runs). Can you try V1.82? It may just be that V1.74 doesn't support Regor well enough. Preliminary test w/ M3A79-T, BIOS Version 1203, MaxxMEM v1.82 Uploaded with ImageShack.us Thanks Calathea for BIOS hint Quote
Crew Don_Dan Posted June 10, 2010 Crew Posted June 10, 2010 Massman, at first I thought there was something wrong with this score as well as I have seen many bugged scores getting posted, but since it's only hitting 5ns lower latency then normal I don't think it's bugged, but of course it could be some new kind of bug. Maybe it's due to the NV chipset? I can only comment on Intel chipsets but they are hitting lower latencies as well, check Vendetta's DDR2 score for example. However, I did not expect the chipset on AMD to create such a big difference in MaxxMem. Quote
aGeoM Posted June 10, 2010 Posted June 10, 2010 ... and does not check my new submission, in a couple minutes. Quote
aGeoM Posted June 10, 2010 Posted June 10, 2010 Did you find the super-tweak? ^^ There is no super-tweak, it's all about latency and how to lower it. Quote
knopflerbruce Posted June 10, 2010 Posted June 10, 2010 Yeah, I think tight subtimings are really important. Quote
aGeoM Posted June 10, 2010 Posted June 10, 2010 Yeah, I think tight subtimings are really important. Thanks man, at last some one saw it. Quote
Crew Don_Dan Posted June 10, 2010 Crew Posted June 10, 2010 Guys, I'm not a n00b, I know how important latency is in this benchmark, but a difference of 7ns to the second best score is HUGE! Quote
aGeoM Posted June 10, 2010 Posted June 10, 2010 Guys, I'm not a n00b, I know how important latency is in this benchmark, but a difference of 7ns to the second best score is HUGE! Ok, you know how important it is, now you only need to know how to reduce it, at that point. I don't mind at all to explain how to do it, in a couple weeks Quote
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.