Quote Originally Posted by Cecil View Post
Gskill Ecos on ASUS 890GX,

7-8-7-20 1600Mhz

Super Pi 1m - 17.488
Super Pi 32m - 18:4.951
Wprime 32m - 9.436
Wprime 1024m - 301.437
PiFast - 27.94
Everest Read - 9314
Everest Write - 6858
Everest Copy - 10845
Everest Latency - 46.9

6-8-7-20 1600Mhz

Super Pi 1m - 17.472
Super Pi 32m - 18:1.769
Wprime 32m - 9.672
Wprime 1024m - 305.448
PiFast - 27.99
Everest Read - 9488
Everest Write - 6859
Everest Copy - 10960
Everest Latency - 45.9

6-8-6-20 1600Mhz

Super Pi 1m - 17.488
Super Pi 32m - 18:2.564
Wprime 32m - 9.626
Wprime 1024m - 306.573
PiFast - 27.68
Everest Read - 9559
Everest Write - 6857
Everest Copy - 10950
Everest Latency - 45.8


7-7-7-20 1600Mhz

Super Pi 1m - 17.457
Super Pi 32m - 18:2.345
Wprime 32m - 9.623
Wprime 1024m - 304.793
PiFast - 27.82
Everest Read - 9297
Everest Write - 6859
Everest Copy - 10758
Everest Latency - 46.8

6-7-6-20 1600Mhz

Super Pi 1m - 17.472
Super Pi 32m - 17:55.934
Wprime 32m - 9.701
Wprime 1024m - 303.563
PiFast - 27.63
Everest Read - 9563
Everest Write - 6857
Everest Copy - 10984
Everest Latency - 45.3

6-8-6-18 1600Mhz

Super Pi 1m - 17.503
Super Pi 32m - 18:0.801
Wprime 32m - 9.45
Wprime 1024m - 301.286
PiFast - 27.68
Everest Read - 9491
Everest Write - 6859
Everest Copy - 11041
Everest Latency - 45.7

6-7-6-18 1600Mhz

Super Pi 1m - 17.456
Super Pi 32m - 17:55.435
Wprime 32m - 9.517
Wprime 1024m - 302.565
PiFast - 27.57
Everest Read - 9562
Everest Write - 6857
Everest Copy - 11054
Everest Latency - 45.2


Is it possible that you could test that with a higher CPU_NB clock, say 2600 or 2800? At 1600, memory bandwidth is bottlenecked by your NB clock...thats why nothing is really changing other than standard deviation when you tighten timings.