SOLVED Overclocking FX-4300 Specifically

[Holymartyr]

Er let me clarify.

With the FSB at 200, CPU at 4.0GHz, NB at 2.0GHz, and HT at 2.6 GHz, my system worked fine and was stable.

With the FSB at 200, CPU at 4.0GHz, NB at 2.2GHz, and HT at 2.6 GHz, my system would still not pass Prime95 for more than a few minutes, and Windows Aero DID NOT work. -This showed my Norbridge probably wasn't stable.

With the FSB at 200, CPU at 4.0GHz, NB at 2.2GHz, and HT at 2.0 GHz, my system would still not pass Prime95 for more than a few minutes, and Windows Aero DID NOT work. - This showed it wasn't dependant on my HT.

With the FSB at 200, CPU at 4.0GHz, NB at 2.2GHz, and HT at 2.6 GHz AND upped voltages for CPU-NB and Northbridge, my system would still not pass Prime95 for more than a few minutes, and Windows Aero DID NOT work. -This showed it wasn't voltage dependant.

With the FSB at 240, CPU at 4.0GHz, NB at 2.16GHz (as close to 2.2GHz as possible), and HT at 2.16 GHz with STOCK voltage, my system would still not pass Prime95, BUT Windows Aero WOULD work. -This would seem to indicate it was FSB and voltage related.

With the FSB at 240, CPU at 4.0GHz, NB at 2.16GHz (as close to 2.2GHz as possible), and HT at 2.16 GHz AND upped voltages for CPU-NB and Northbridge, my system would pass Prime95, and Windows Aero WOULD work. -This would seem to indicate it was FSB and voltage related, and is justified by being fixed.

Hope this helps =/

Also just tried upping my Vcore to 1.500 (from 1.475), CPU PLL to 2.80 (from 2.65), and raising my CPU multi from 19 to 19.5 to give me 4680 MHz (240 FSB) and the 2nd core failed almost instantly. All the others seem to be fine. Temps were still way in the green.

 

[Anonaru]

@ Anonaru, have you heard of the PileDrivers being less clockable Mhz wise but actually doing 'more' work even though not reaching the bigger Mhz of the BullDozers?

I know you have a computer store and at that point price becomes a consideration and you have to 'use' what you sell to know what is going on. Well let me say that using makes you more able to serve your customer. Something the big boys only do from an engineering standpoint and likely little real hands-own. I am saying I don't like the cheap UD3, but I can glimpse why some use them. So your using; makes it easier to know what the UD3 will do.

The OPs windows areo goes nuts because CPU_NB is too slow for ram transfer effectiveness or a quirk in bios setting NOT visible to us. It picks up when he FSB overclocks but the same thing could be done by just setting the CPU_NB divider for more speed. Of course the CPU_NB voltage should be raised to complement the increased CPU_NB.

None of this stuff is rocket science nor is what works on one a sure thing on another system. Freeken lot of trial and error.

Happy Holidays "Anonaru". RGone...ster.

Haha, happy holidays! I like these boards personally-- They have allowed me to go just about as far as some of the much better boards, but yes--

They require 50 times more trial and error to get the result you need. A bit on the "ugh" side of things. And yes, the PD are definitly happier per-clock than their BD brothers, particularly beyond 4.3 GHz.. I've not gotten the pleasure of pushing a 4300, but I know my 4100 can get higher freq than my 6300 can at the shop

Edit: You will need something like 1.275 CPU-NBv for anything above 2200 on the NB.. Also keep the NB and HT frequency the same if at all possible! I'd try to run both as close to 2.4GHz as possible

Also just tried upping my Vcore to 1.500 (from 1.475), CPU PLL to 2.80 (from 2.65), and raising my CPU multi from 19 to 19.5 to give me 4680 MHz (240 FSB) and the 2nd core failed almost instantly. All the others seem to be fine. Temps were still way in the green.

I don't think it is your vCore. I'd use that increased multiplier, then decrease your FSB to 238, set your NB and HT link to whatever you have that's closest to 2400, set your CPU-NB to 1.275, set PLL to 2.675 (Or leave alone) then tell me what you get (With LLC on "Very High")

 

[Holymartyr]

Yeah my CPU-NB is running just above 1.3 right now. I think I might be able to lower it to 1.275 and still be stable, but I don't remember why I upped it to 1.3 in the past so I'll mess with that later - lol.

I guess at this point I just want (if possible) some sort of verification that I just got a gimpy 2nd core, or some sort of verification that i'm still doing something wrong (voltage, FSB or multi or w/e wise)

Edit: Will raising my actual Northbridge voltage increase stability? Read that somewhere. Also what about the NB/PCIE/PLL voltage?

Seems some to say to lower that, and it increases stability.

 

[Anonaru]

Yeah my CPU-NB is running just above 1.3 right now. I think I might be able to lower it to 1.275 and still be stable, but I don't remember why I upped it to 1.3 in the past so I'll mess with that later - lol.

I guess at this point I just want (if possible) some sort of verification that I just got a gimpy 2nd core, or some sort of verification that i'm still doing something wrong (voltage, FSB or multi or w/e wise)

Edit: Will raising my actual Northbridge voltage increase stability? Read that somewhere. Also what about the NB/PCIE/PLL voltage?

Seems some to say to lower that, and it increases stability.

Read my edit.

Those values can sometimes help. Set your chip nb and chip ht up a single bump. The NB/PCIE/PLL voltage shouldn't matter in the least at this point. A gimpy core is an easy cop-out.. I can't imagine AMD shipping you a PD that couldn't get past 4.5 GHz

 

[RGone]

Haha, happy holidays!

They require 50 times more trial and error to get the result you need. A bit on the "ugh" side of things. And yes, the PD are definitly happier per-clock than their BD brothers, particularly beyond 4.3 GHz.. I've not gotten the pleasure of pushing a 4300, but I know my 4100 can get higher freq than my 6300 can at the shop

Edit: You will need something like 1.275 CPU-NBv for anything above 2200 on the NB.. Also keep the NB and HT frequency the same if at all possible! I'd try to run both as close to 2.4GHz as possible

"I know my 4100 can get higher freq than my 6300 can at the shop " >> Well that is pretty much inline with what I am beginning to hear. Problem is you cannot know in advance which PD will not haul booty. I know CPU buying is a dice roll anyway, but AMD FX cpus are just all over the map now it seems.

Sure you or I could do that 50 times more trial and error, but for the 'average' user that comes in this forum and wants to overclock...well it just seems counter-intuitive to me. All the trial and error I do in minutes. Told a bud of mine to up his CPU-NB to ~2600Mhz and decided I better try it as well. Dang thing HAD to have added CPU_NB of nearly 1.3V to boot. Then his email came and I knew exactly what to tell him to do.

His CHV won't run over about 2550Mhz CPU_NB but mine will do 2700Mhz. That is how hard it is to tell someone exactly what to use for numbers it is all very much sitting at the keyboard and testing.

RGone...ster.

PS: thanks for the awesome tips on UD3.

 

[Holymartyr]

Just tried what you said:

FSB = 238 MHz
Multi = x19.5
CPU Speed = 2641 MHz
NB/HT = 2380 MHz
CPU PLL = 2.665 V
CPU NB = 1.2875 V
VCore = 1.475 V
LLC = Ultra High

2nd Core failed within 7 seconds.

For LLC options, I have Auto, Regular, Medium, High, Ultra High and Extreme.

Edit:

Highest stable settings for me BTW are:

FSB = 240 MHz
Multi = x19
CPU Speed = 2560 MHz
NB/HT = 2400 MHz
CPU PLL = 2.675 V
CPU NB = 1.3 V
VCore = 1.475 V

 

[Anonaru]

Just tried what you said:

FSB = 238 MHz
Multi = x19.5
CPU Speed = 2641 MHz
NB/HT = 2380 MHz
CPU PLL = 2.665 V
CPU NB = 1.2875 V
VCore = 1.475 V
LLC = Ultra High

2nd Core failed within 7 seconds.

For LLC options, I have Auto, Regular, Medium, High, Ultra High and Extreme.

Edit:

Highest stable settings for me BTW are:

FSB = 240 MHz
Multi = x19
CPU Speed = 2560 MHz
NB/HT = 2400 MHz
CPU PLL = 2.675 V
CPU NB = 1.3 V
VCore = 1.475 V

Try once more with pll at 2.695 just to shoot down a final birdie in the dark. If that fails, try again with the FSB at 236 and the same setting (That is, the higher multiplier and the sort).

If neither work, I'll stop smacking your chip around

 

[Holymartyr]

Failed within 2 seconds with FSB at 238 and took about 30 seconcds with FSB at 236.

Any another ideas? =( lol (Thanks so much for everything thus far BTW)

 

[Holymartyr]

Bump - Update:

Got my water cooler, NB fan and new case all set up with proper cable routing!

NB now is sub 50 degrees 24/7, and CPU/Mobo temps are anywhere between 40-20 degrees on full load.

Any more ideas on my OC situation?

 

[Anonaru]

Bump - Update:

Got my water cooler, NB fan and new case all set up with proper cable routing!

NB now is sub 50 degrees 24/7, and CPU/Mobo temps are anywhere between 40-20 degrees on full load.

Any more ideas on my OC situation?

It very well can be as we have feared-- Your processor appears to have a weak core. We can try the following for giggles:

FSB = 238 MHz (down to 237 if you get a fail)
Multi = x19.5
CPU Speed = 2641 MHz
NB/HT = 2380 MHz
CPU PLL = 2.695 V
CPU NB = 1.3 V
VCore = 1.475 V
LLC = Extreme

These are pretty liberal settings for your current level of OC-- I can't emulate the problem you're having on my own UD3 boards, so I really think it is the chip at this point

 

[RGone]

CPU Speed = 2641 MHz

238x19.5 = 4641 Mhz.

 

[Holymartyr]

I'm not sure what extreme is supposed to do for me for LLC, but when I enabled, my core wouldn't move from 1.424 V when I've set it at 1.475...

So Extreme is out of the question. Redoing the test with LLC bumped down one notch.

 

[Anonaru]

I'm not sure what extreme is supposed to do for me for LLC, but when I enabled, my core wouldn't move from 1.424 V when I've set it at 1.475...

So Extreme is out of the question. Redoing the test with LLC bumped down one notch.

Sounds like your extreme is borked too. Aye, keep Ultra High then Happy running, hope it finally works =_=

 

[Holymartyr]

Tried Ultra High (one notch down) - failed within 20 seconds. Tried:

FSB = 237 MHz
Multi = x19.5
CPU Speed = 2621 MHz
NB/HT = 2370 MHz
CPU PLL = 2.695 V
CPU NB = 1.3 V
VCore = 1.475 V
LLC = Ultra High

I could try with 1.5 VCore. But since I run my comp 24/7, I don't want premature CPU failure to be an issue...I'm not opposed to buying a new CPU (from all the new info I may very well buy an 8350 if It does fail - and if AMD hasn't come out with a new processor to beat the pants off intel lol), but I don't want a few MHz in CPU speed to severely shorten my CPU life just because of 1.5 VCore.

Kinda catch my drift?

 

[Anonaru]

Tried Ultra High (one notch down) - failed within 20 seconds. Tried:

FSB = 237 MHz
Multi = x19.5
CPU Speed = 2621 MHz
NB/HT = 2370 MHz
CPU PLL = 2.695 V
CPU NB = 1.3 V
VCore = 1.475 V
LLC = Ultra High

I could try with 1.5 VCore. But since I run my comp 24/7, I don't want premature CPU failure to be an issue...I'm not opposed to buying a new CPU (from all the new info I may very well buy an 8350 if It does fail - and if AMD hasn't come out with a new processor to beat the pants off intel lol), but I don't want a few MHz in CPU speed to severely shorten my CPU life just because of 1.5 VCore.

Kinda catch my drift?

No, you don't need 1.5v and you shouldn't 24/7 1.5 on just the vCore, especially with a high level of LLC. Guess we're at the ceiling of your chip If you can turn back the NBv a bit and CPU-PLL on your best stable setting you can make your chip run a bit happier / cooler

 

[Holymartyr]

Well that's a little depressing =(

So the verdict is my 2nd core is either defective, or just for some reason way weaker than the rest of my cores?

Thanks for the help.

 

[Anonaru]

Well that's a little depressing =(

So the verdict is my 2nd core is either defective, or just for some reason way weaker than the rest of my cores?

Thanks for the help.

Of course! Sounds like a weak core vs a defective one-- 4.5 is still a decent OC and will net you some nice gains

 

[Ajay57]

How old is the FX-4300 any chance for a RMA for a replacement? Shame he has a nice Mobo there Anonaru buddy. AJ.

 

[Holymartyr]

Just over a month old. No chance for RMA at this point... /sigh

 

[Ajay57]

Shame you have a nice set up well that what happens sometimes sorry to hear that!