How come this clock was not sucessful

trunks512

Distinguished
Jan 4, 2006
73
0
18,630
I bumped up my fsb to 240 while upping voltage a bit to 1.3625V
the multiplier I left at 11 and the memory I left to 200 as well

I am currently running a 4200+ (200x11 2.2Ghz) chip and have the A8N32-SLI as my motherboard

Can someone please give me a suggestion on why this overclock failed?
Not enough voltages? Have to fiddle with the fsb/mem divider? anything else?

Any help would be appreciated
thanks
 

chuckshissle

Splendid
Feb 2, 2006
4,579
0
22,780
Try clocking down on your ram while you increase the cpu. Usually it's more stable for oc cpu on lower ram speed. Try setting your ram clock to 2T. Also, try increasing your fsb and bridge chip voltage as well. But be extra careful on adjusting voltage as it can result to overheating or frying your system. Also monitor the heat as you oc your rig.
 

EllisD420

Distinguished
Jan 27, 2006
250
0
18,780
Another thing you can try is to use a divider on your ram. Running it at 240Mhz shouldnt be a prob cause you have high quality ram, but you never know...they might not be the best clockers. If you bump your speed down to 166 that would give you 240 x 0.83(166 divider) = 199.2mhz so thats right in spec. Give it a shot doing that and leaving everything the same. If it works, then you might need to up the ram voltage in order to get it to 240mhz 1:1.

Just like Chuck said though...you really gotta watch it while adjusting your voltages, it increases heat and increases the possibility of parts failing. Just watch your temps carefully.
 

EllisD420

Distinguished
Jan 27, 2006
250
0
18,780
Ah yes, i dont know how i forgot about that. Try and make your HTT speed as close to 1000Mhz as possible. So if you are overclocking up to 250FSB, set it to 4x so it will be 4x250=1000Mhz. If you didnt change this during your last attempt, then the HTT bus was way out of spec. You probably had it at 5x240=1200Mhz which isnt likely going to work.

The general rule ive seen while going around forums is:

FSB up to 250 = HTT x4
FSB up to 330 = HTT x3 (people usually say anything from 250-300 you should set to HTT x3 but up to 330 would still put you under spec so i think it should be ok)

and your not going to be setting it higher than 330 so i will just leave it at that.
 

denster

Distinguished
Aug 18, 2005
158
0
18,680
I bumped up my fsb to 240 while upping voltage a bit to 1.3625V
the multiplier I left at 11 and the memory I left to 200 as well

I am currently running a 4200+ (200x11 2.2Ghz) chip and have the A8N32-SLI as my motherboard

Can someone please give me a suggestion on why this overclock failed?
Not enough voltages? Have to fiddle with the fsb/mem divider? anything else?

Any help would be appreciated
thanks

OK I will tell you. First reset everything to default.
Leave your HT Frequency set to AUTO.
Leave Cool and Quiet DISABLED
Leave DRAM set to AUTO

Change your CPU multiplier to 10. Then raise your FSB to 240.

Anything above 20% and you will have to MAX all your VOLTAGES, ram, Vcore, HT, and chipset. Which opens another can of worms, and we can get into that after you run stable at 2400Mhz :)

You should be able to run PRIME 95 all day and night at 240x10 on stock voltages :)
 

SuperFly03

Distinguished
Dec 2, 2004
2,514
0
20,790
240x11 is still a solid overclock s don't be sad about it. But your Vcore seems a bit low for the clock. I assume stock is 1.3v, so your pretty safe up to 1.45v ish (ifyou have moderate to decent cooling). I run on a 250x11 setup now on a single core and the damn thing just won't go higher.

I don't think my venice should stop at 2750, I know its capable of 3000, i just got a bad chip, as you might have as well. Overclocking is an EXTREME exercise in patience, and frustration. It will take hours, it will take days, and if you want the absolute most Mhz possible it could easily take you weeks.
 

ChipDeath

Splendid
May 16, 2002
4,307
0
22,790
Anything above 20% and you will have to MAX all your VOLTAGES, ram, Vcore, HT, and chipset. Which opens another can of worms, and we can get into that after you run stable at 2400Mhz
That's misinformation. Some RAM will overclock worse once you get past a certain voltage, and blindly increasing all the V adjustments is far more likely to give you some overheating issue, without you having any idea which component is the problem.

The only proper approach is take things slowly, and try to find the limits of each part individually.

It takes weeks or perhaps even months of careful testing to find the 'sweet spot' on any one system.

I have a 25% overclock but I haven't maxed a single one of my voltages, and I have a less overclocker-friendly winchester chip too, which if anything would require a lot more voltage than the newer cores.

I know you're just trying to help, but I really don't like the idea of people thinking 'MAX VOLTAGE' is a good idea. It's not, and will likely end up making someone kill some expensive computer parts.
 

SuperFly03

Distinguished
Dec 2, 2004
2,514
0
20,790
Agreed, voltage is probably the No. 1 killer of PC parts. With watercooling i can keep my Venice core under 50c even if I applied 1.9v to it... but that doesn't mean I should since stock is 1.4v. Chip made a great point, take it slowly and incrementaly while keeping an eye on heat and the air flow in your case.

Also maxing the voltages means diffrent things on diffrent mobo's, esp DFI. DFI lets you use EXTREME voltages while ASUS will allow moderate (in general, not all boards are the same). For Ex, *i think* Asus boards top out at 1.55v on vCore for A64 procs (current generation), but DFI has an option called vcore special that allows you to reach over 2v, maybe even 2.25v. So please be careful when you recommend maxing voltages.
 

denster

Distinguished
Aug 18, 2005
158
0
18,680
Anything above 20% and you will have to MAX all your VOLTAGES, ram, Vcore, HT, and chipset. Which opens another can of worms, and we can get into that after you run stable at 2400Mhz
That's misinformation. Some RAM will overclock worse once you get past a certain voltage, and blindly increasing all the V adjustments is far more likely to give you some overheating issue, without you having any idea which component is the problem.

The only proper approach is take things slowly, and try to find the limits of each part individually.

It takes weeks or perhaps even months of careful testing to find the 'sweet spot' on any one system.

I have a 25% overclock but I haven't maxed a single one of my voltages, and I have a less overclocker-friendly winchester chip too, which if anything would require a lot more voltage than the newer cores.

I know you're just trying to help, but I really don't like the idea of people thinking 'MAX VOLTAGE' is a good idea. It's not, and will likely end up making someone kill some expensive computer parts.

You only read part of what I have suggested and throw the rest to the wind?

I never told him to MAX it out, I told him to run on STOCK VOLTAGES to achieve the 240FSB. Read what I wrote before you start to contradict.

Besides that, OC'ing an X2 with an ASUS MB is something you obviously know nothing about.

AND not trying to be a jerk, but your username suggests the very thing we should all like to avoid.
 

denster

Distinguished
Aug 18, 2005
158
0
18,680
Agreed, voltage is probably the No. 1 killer of PC parts. With watercooling i can keep my Venice core under 50c even if I applied 1.9v to it... but that doesn't mean I should since stock is 1.4v. Chip made a great point, take it slowly and incrementaly while keeping an eye on heat and the air flow in your case.

Also maxing the voltages means diffrent things on diffrent mobo's, esp DFI. DFI lets you use EXTREME voltages while ASUS will allow moderate (in general, not all boards are the same). For Ex, *i think* Asus boards top out at 1.55v on vCore for A64 procs (current generation), but DFI has an option called vcore special that allows you to reach over 2v, maybe even 2.25v. So please be careful when you recommend maxing voltages.

Oh Boy here I go again...

not trying to be a jerk, again, but who are you to give advice???

*I THINK* You are running your PROC slightly above 1.6V

HUH? Go FIGURE - What did you say your stock voltage is?

And your FSB? I guess all that is necessary for your 3800 to run? You can forget being stable at that speed.
 

ChipDeath

Splendid
May 16, 2002
4,307
0
22,790
You only read part of what I have suggested and throw the rest to the wind?
No, I read all of what you said, and quoted the part I felt needed 'touching up'.... :D

I never told him to MAX it out, I told him to run on STOCK VOLTAGES to achieve the 240FSB. Read what I wrote before you start to contradict.
You did indeed say :
Anything above 20% and you will have to MAX all your VOLTAGES, ram, Vcore, HT, and chipset.
So yes, you didn't say to increase voltage until trying to get above 2400Mhz, but then you immediately say 'MAX Everything' for anything more. in fact you specifically say to max out RAM voltage, without knowing exactly what his RAM is. it's a proven fact that some ram chips get no benifit from adding too much voltage, and some actually get worse (in terms of maximum attainable clockspeed).

I made the post for the reasons I stated within it. Just trying to avoid a post saying "I MAXed out all my voltages and now my computer doesn't work!" in a few day's time... not necessarily from the OP, just maybe someone else reading this thread....

your username suggests the very thing we should all like to avoid.
Yup... And I have killed a GPU (r9800Pro) through putting 2V through it for a while, but I haven't yet (touch wood) killed a single CPU, and I've probably handled hundreds by now... And not wanting to sound arrogant, but I have been around for a while, and do know whereof I speak. :D
 

ChipDeath

Splendid
May 16, 2002
4,307
0
22,790
Something around 460 or 465 artifact-free (3dmark and suchlike), could probably have gone to around 480ish without in-game artifacts tho.

Was using a VGA Silencer modified with an 80mm fan so more air went through. On reflection, I should've probably stopped at 1.85V or so.... but where's the fun in that :mrgreen: