Very strange behaviour after changing PC case and PSU

[LaZ]

WOooohooo I think I solved 2 things at once now :) I contacted Hiper about my issue with the comp-ON-led light being constantly ON and apperantly that is a known bug/issue with my particular mobo, their solution was to remove one wire from the the PSU´s USB connector, but I simply disconnected the whole USB cable from the PSU to the mobo. So now the led light goes out when I turn computer OFF :)

AND also now my computer cold boots with no problems at all, no more one extra boot cycle. Damn this is gonna be a fine day.

I have now set the Static tRead Phase Adjust to [AUTO] - seems to be working fine.

Now to my next question. I found in memset the EPP settings for my mems but they dont use the same names for the settings as in the bios, I have done some reading and figured out most of the names but can u pls help me with the rest?

Here´s the screenshot of my EPP


It is:

BIOS NAME: Rank Write to Read Delay + Write To Precharge Delay. Those two I cant figure out what the correspond to in memsets EPP setting.

[Lsdmeasap]

Nice to hear!

Yeah, Here is a comparison of memset to BIOS settings, most are there that you will find in your EPP. >>>>

Memset:-


Bios:-
Memset Equivalent

Performance Enhance :
System Memory Multiplier (SPD)
Memory Frequency(Mhz):
DRAM Timing Selectable :
Standard Timing Control
CAS Latency Time: : = 1 Cas# Latency [tCL]
DRAM RAS# to CAS# : =  2 Ras# to Cas Read Delay & 3 Ras# to Cas Write Delay
DRAM RAS# Precharge : = 4 Ras# Precharge [tRP]
Precharge Delay(tRAS) : = 5 Precharge Delay [tRAS]
Advanced Timing Control
ACT to ACT Delay(tRRD) : 11 Act to act Delayed [tRRD]
Rank Write to READ Delay : 10 Write to Read Delayed [tWTR] 
Write To Precharge Delay : 9 Write to Precharge Delayed 
Refresh to ACT Delay :  7 Refresh Cycle Time [tRFC]
Read To Precharge Delay :13  Read to Precharge [tRTP]
tRD : 8 Performance Level 
tRD Phase Adjustment :  Unknown
Command to #CS Timing: : 16 Command Rate [CR]

The only one I dont see in your EPP is tRd (Performance level = Static tRead Value) which like I said before you will have to adjust n test 7-10. This is a MAJOR performance setting, hence the name. So lower is better for sure

[LaZ]

Thanx :) Now I only have a few more question hehe :)

No.1
I have all values figured out now except one. I can´t find the corresponding value in the EPP screen for "Write To Precharge Delay"

Memset says my value now is 12, but I entered 5 in bios. What is my EPP value for "Write To Precharge Delay"?

No.2
Do I need to change the subtimings again if I try overclocking the CPU? Or will loosening the 4-4-4-12 timings be enough if I run into stability/non boot issues?

[Lsdmeasap]

Write To Precharge Delay : 9 Write to Precharge Delayed

That is correct, sometimes 4=11 5=12 and so on for that setting. That is why I suggested you to try an test different values. 4/11 is likely the better setting.(your should also set it to 11 and make sure it isnt giving you 11, if it does leave it that way unless a lower # proves to test better) But sometimes 9-11 actual BIOS setting is better no matter what you actually get in windows, you just have to test each

As for it being in the EPP or not, they do not enter all values often times in the EPP profiles, so it is either there or not, which is looks like not.

[LaZ]

Setting "Write to precharge delay" to 11 in BIOS gives me a value of 18 in memset so 4=11 5=12 seems correct. I´ll try the 4=11 now.

And also I edited my last post at the same time as u answered the last post so here is the question again :) Will these subtimings work if I overclock the cpu/fsb and change the SPD to keep the mems at around 800? Or do I need to redo the subtimings again?
My goal is to get to set the FSB to 400 (3.6GHz) and have the mems at 1:1

[Lsdmeasap]

Well what you need to do is set the CPU now and then do the subtimings. Much easier that way. But na, you should be able to run 444-12 at 800 no problem, the subtimings may be getting to tight if you do them before you do your FSB/CPU so just do it first then the ram subtimings

[LaZ]

Hehe yeah I thought so. I wouldnt have gone this far into these subtimings at this point if I had known it was the USB connector from the PSU to the mobo that was the cause of the stability issues I had. Anyway I am gonna have myself a bite to eat now and try some cpu clocking afterwards. Running Orthos large FFTs to stress the mems now to see what happens.

[Lsdmeasap]

Orthos Large FFT only stresses CPU. NONE on the ram. Blend stresses CPU and Ram, Small and large FFT CPU ONLY. Just so ya know

Memtest86+ is how you test that ram only

I'll see ya tonight, good luck in you overclocking

[LaZ]

oh, hmmm ok... I know about memtest+ though. will switch to that. Orthos says "Large In place FFTs - stress some RAMS. thats why I ran it.

I will most likely be posting again soon... and thanx again for taking the time to help me!

[Lsdmeasap]

Ahh well yeah I forgot about that. What that means is Stress CPU ALOT, Barely stress RAM.

Blend is 50/50. But orthos does not have a RAM Only test is the point of me saying to test ram use Memtest86+