To overclocking. Is there anything I should avoid doing that would immediately damage my system? I have read Zebo's useful isolate and consolidate guide to overclocking, but still have some questions. Pardon me for the little I know..
I know my cpu is HTT*Multiplier; 200mhz * 9 = 1,800mhz
So basically the idea is to increase the HTT by increments of 5% while creating a balance by changes in the vcore.
I allow 5 minutes for every 5% increment until it should fail, which should ultimately be the final speed? Ex: 250mhz * 9 = 2250mhz (Seems low..)
Does that mean the memory speed is an isolated equation and wouldn't matter if it were to remain unchanged?
Ok. Step 1; change the multiplier to 6 or 7? Instead of 9..?
Set memory to SPD and 100.
Increase HTT/FSB up until it fails to boot then down 5% for the max safe HTT.
Step 2; Change the multiplier to 9
Change the vcore to 1.50-1.55
Set mem to SPD and 100, again?
Set the HTT to increments of 5% up to the boards max found earlier., down 5% when it fails Prime95 resulting in most probably max CPU speed, and prime test for 12 hours stable.
Step 3;
Set mem setting to 200 and 2.5-4-3-7 1T and give it max volts. Or 200 and 2-2-2-7 1T for LL. Question: What is LL? Does that speed apply or work cas latency 2 ram?
Give the ram the boards max volts. (2.9Vdimm safe for all ram) Would it be ok to give it something lower like 2.8V, just to be safe?
Set CPU multipler to 6-7 again.
Raise HTT to boards max or until it fails memtest 5 and 6.until the max MEM speed is found.
Final SteP; Consolidate.
Mem Divider.. (later)
Now some questions..
1) If the process if followed safely is there still a chance that something might subsequently get damaged?
2) With my specifications: DFI Ultra-D board, 3000+ Venice, OCZ 2-3-3-6 would it be likely to achieve 2400mhz or even 2600mhz?
3) Pardon for this long burdonsome post, thank you in advance and zebo for his guide..
Finally how does all this pay off?
I know my cpu is HTT*Multiplier; 200mhz * 9 = 1,800mhz
So basically the idea is to increase the HTT by increments of 5% while creating a balance by changes in the vcore.
I allow 5 minutes for every 5% increment until it should fail, which should ultimately be the final speed? Ex: 250mhz * 9 = 2250mhz (Seems low..)
Does that mean the memory speed is an isolated equation and wouldn't matter if it were to remain unchanged?
Ok. Step 1; change the multiplier to 6 or 7? Instead of 9..?
Set memory to SPD and 100.
Increase HTT/FSB up until it fails to boot then down 5% for the max safe HTT.
Step 2; Change the multiplier to 9
Change the vcore to 1.50-1.55
Set mem to SPD and 100, again?
Set the HTT to increments of 5% up to the boards max found earlier., down 5% when it fails Prime95 resulting in most probably max CPU speed, and prime test for 12 hours stable.
Step 3;
Set mem setting to 200 and 2.5-4-3-7 1T and give it max volts. Or 200 and 2-2-2-7 1T for LL. Question: What is LL? Does that speed apply or work cas latency 2 ram?
Give the ram the boards max volts. (2.9Vdimm safe for all ram) Would it be ok to give it something lower like 2.8V, just to be safe?
Set CPU multipler to 6-7 again.
Raise HTT to boards max or until it fails memtest 5 and 6.until the max MEM speed is found.
Final SteP; Consolidate.
Mem Divider.. (later)
Now some questions..
1) If the process if followed safely is there still a chance that something might subsequently get damaged?
2) With my specifications: DFI Ultra-D board, 3000+ Venice, OCZ 2-3-3-6 would it be likely to achieve 2400mhz or even 2600mhz?
3) Pardon for this long burdonsome post, thank you in advance and zebo for his guide..
Finally how does all this pay off?
Facebook
Twitter