• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Fx-4100 overclock stuck at 4.4ghz

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

falcon93

Registered
Joined
Jun 18, 2013
Ive been trying for a couple days to get my Fx-4100 to a higher speed. My first goal was 4.8, huge jump I know. Then I set my self lower to 4.6 when I couldn't reach that 4.8. I seem to not be able to reach any higher than 4.4ghz with out bluescreening.

During the overclock attempts, I kept my voltage at 1.5 and did not raise it from there. Some forums were warning not to go over 1.5.

My 4.4ghz is stable while running any of the 3 stress tests from prime95. Temperature, at its highest over 8 hours on stress test, was 47c.

Any help getting this to around 4.6-4.8 would be very appreciated, I am trying to get a higher streaming quality for my viewers.

Current specs:

Clock - 4.400ghz
Bus - 200
Multiplier - 22
Votage - 1.5
CnQ - disabled
Turbo - disabled
Ram - 1600Mhz Corsair Vengeance gaming memory
Mobo - Asus M5A88 Evo
 

Attachments

  • cpu-z.png
    cpu-z.png
    75.4 KB · Views: 2,092
Last edited:
your v core might be a little on the high side. I would start by tuning it down and bit and see if the system is still stable at 4.4.
 
:welcome: to OCF falcon93

Have you given this guide a read through?

It will give you a lot of insight into oC'ing that type of chip. Typically we go the other way when overclocking. Raising frequencies till it's unstable then adding voltage
 
Sure, I would try something in the 1.3 range, since your voltage is high for the given clock speed. but again, Like Johan said, there's more than one way to skin a cat. Go ahead and check that guide out and report any questions, probably your best bet.
 
You overclock the same way as I do, set the max vcore you're willing to put into the CPU and start from there (saves a lot of minor adjusting) anyway... I would raise the HT to 2400 and give the CPU multiplier another try, if that doesn't work then its time to use the reference clock for the last MHz, you may be able to get another 200mhz from that
 
ty all for your suggestions, Ill try it out tomorrow(well today technically :p) and get back to you.

PS: Kenny, I read one of your posts about the voltage i believe, thats why I did it :p
 
You overclock the same way as I do, set the max vcore you're willing to put into the CPU and start from there (saves a lot of minor adjusting) anyway...

Makes perfect sense "Keny" to start will all the Vcore you are "willing" to use. If one is not willing to ever use more Vcore than the amount set, then starting at the top of your "willing" Vcore removes the Vcore as the problem with all those BSODing baby steps that just wonk out the operating system.

Starting at the Max Vcore I am 'willing' to use also will let me know d*mn quick if I am going to be overheating when I get a pretty fair overclock going on. If it overheats at next to n0 overclock but the max Vcore I will use, then I know it will not run big Mhz numbers when I get thru baby-stepping up the Vcore ladder.

The people able to get outside the box and realize that if they start at the max Vcore they are willing to use, then the overclocking process is greatly shortened. Those I have personally told to decide the max amount of Vcore that they are 'willing' to use and start with that cpu voltage are able to reach their max overclock in at least half the amount of time they would use in baby-stepping up the Vcore ladder and having to wade thru BSOD and failing P95 Blends as they try a speed and fail and have to up Vcore.

I don't guess baby-stepping up the Vcore ladder is "wrong" just slow and often allows too many BSODs to occur which I am unwilling to accept and think my O/S is still in lovely shape.

I got over that old way of baby-stepping up the Vcore ladder over 10 years ago when it was explained to me that if I am never 'willing' to use more than 1.525Vcore; then use that Vcore to begin with. Then every cpu speed I test and move upwards to the next speed, the failure is not caused by too little VCore since I am already at the most Vcore I am willing to use. I am never going to be 'willing' to use more Vcore than I have set to begin with, so when I finally come to a cpu speed that does fail...the failure is not going to be remedied by more Vcore because I am already at the Max Vcore I am 'willing' to use. Cuts a heck of a lot of BS and false starts and guessing out of way.

Most just don't get it. They hold to some method described to them over the years and cannot understand the principle of beginning with all the cpu voltage you will use and use that amount to begin with, takes away all the guessing about the cpu voltage issue from the beginning.

It took me less than 20 mins to get my first FX-8120 to a stable 4.6Ghz using the begin at Max Vcore to begin with instead of starting at stock voltage and bumping up the Vcore when the overclock fails and beginning again with just a little more Vcore and go until it fails. Then doing that over and over again until I reach the Max Vcore I am "willing" to use. Max Vcore I am 'willing' to use is just that; the most Vcore I am "willing" to use and that voltage number is the max. Start at Max and never have to go thru all those baby-stepping Vcore fails and up Vcore and repeat until I get to the Max Vcore I am 'willing' to use.

Once the principle was explained to me; I never again played the baby-stepping game. It took too long and I saw no sense in baby-stepping up to the most Vcore I was 'willing' to use. An overclock is determined by how much Vcore the cpu can get and if it will be able to run cool enough at the max Vcore I am 'willing' to use. I just shortcut all those baby-stepping Vcore bumps after some speed failure from too low Vcore and begin the overclock already using the Max Vcore I am 'willing' to use. Short and sweet and makes complete sense once we break out of the old mould.
RGone...
 
I felt that as a newcomer the theOC arena falcon may benefit from a slower approach to develop a sense for the chip. You know the old do as I say not what I do!! Like Rgone said, once he gets to 1.5 and decides that his temps are good and wants to go a bit higher "baby steps" become a bit more important. Or a feel for I know I need at least .XX V-core for the next 100 Mhz + 1 for heat and voila.
 
I felt that as a newcomer the theOC arena falcon may benefit from a slower approach to develop a sense for the chip. You know the old do as I say not what I do!! Like Rgone said, once he gets to 1.5 and decides that his temps are good and wants to go a bit higher "baby steps" become a bit more important. Or a feel for I know I need at least .XX V-core for the next 100 Mhz + 1 for heat and voila.

Hey man I have no problems with baby steps as it were. It seems a little convoluted to me after having the "apply the most Vcore to begin with" explained to me back 10 years ago. I hate BSODs and I bypass most of that by going to all the Vcore I am 'willing' to apply and clocking the cpu speed up from there. Plus by going the Max Vcore I am 'willing' to use at first that actually lets me know if I am going to be okay on temps from the get go. It is just a way of looking at the end result quicker and getting there in a shorter time. I know what my Max Vcore will be from the beginning. Set that Vcore and clock up to the max cpu speed that Vcore allows and job done as my UK buddies used to say.
RGone...
 
Either way It's kind of a moot point as it is. I don't do it either.
I'm in it with Kenny though. The NB definately needs to be sped up since you're looking for better video performance. Might not hurt to get a HWmonitor shot under load at some point as well.
 
two posts and rgone has already written a book... lol

As for the baby steps, you can do what you want I wouldn't want anyone to make that decision for you. If you feel you want to go for one O/C approach and not the other, that's your prerogative.
 
Thank you all for helping with this. I liked that you guys put your reasoning for why you overclock the way you do. Good to know I have been doing it properly to this point xD

Ill be upping my NB either in 1hr or some time tomorrow, depending on if I get too tired. I will also update this post with my HWM under 1hr of full load(in about 50 mins).

Edit: My cooling system is a Cooler Master Liquid Cpu Cooler, Seidon 120m
Edit2:Uploaded SS underload after 1 hr, note that it is at night atm but still quite warm in my room. I will do another later today for 4-8 hours underload as well as during the day.
 

Attachments

  • underload.png
    underload.png
    172.3 KB · Views: 1,118
Last edited:
3 things >_<

1. I got the NB to 2400mhz but it still was unstable when I tried making the multiplier higher.

2. After that I lowered my voltage to 1.45. I tried 1.4 but that was unstable, I haven't tried anywhere in between.

3. I noticed something weird with cpuz when doing that 8 hour stress test, that I didn't notice before. It is toggling the voltage AND my multiplier, when under high stress.

adding pictures of bios, and cpu-z when its flicked to lower volt/mult
 

Attachments

  • why.png
    why.png
    20.2 KB · Views: 1,168
  • IMG_20130622_155011_716.jpg
    IMG_20130622_155011_716.jpg
    928.4 KB · Views: 1,179
  • IMG_20130622_155023_771.jpg
    IMG_20130622_155023_771.jpg
    889 KB · Views: 1,118
  • IMG_20130622_155103_356.jpg
    IMG_20130622_155103_356.jpg
    923.4 KB · Views: 1,165
Disabling APM in bios? It makes the CPU throttle when under load (to prevent VRM's going "pop"!).

Am I right rGone?
 
would that be safe at my voltages? o.o I read into those today and saw some horror stories.
 
Disabling APM in bios? It makes the CPU throttle when under load (to prevent VRM's going "pop"!).

Am I right rGone?

That is one of the problems with the earlier chipset motherboards, there is no menu choice for Disable APM. I looked thru all his bios pics and did not see that menu item.

So it will throttle the cpu back when the VRMs go under a heavy load. A load that is considered high for the ability of the VRM circuit. So it is certainly not a 4 core overclock at 4.4Ghz as it stands.
RGone...
 
would any of these help?
 

Attachments

  • IMG_20130622_173536_481.jpg
    IMG_20130622_173536_481.jpg
    828.8 KB · Views: 1,121
  • IMG_20130622_173520_393.jpg
    IMG_20130622_173520_393.jpg
    928.1 KB · Views: 1,155
No those are for >> The Advanced Power Management (APM) Specification defines the following power states:

Ready
Stand-by
Suspended
Hibernation
Off

That is not the new Application Power Management spec involved with the AMD FX core processors.
RGone...
 
APM has to be disabled with MsrTweaker with this motherboard and the FX chips. I haven't been able to find a working download in a long time, so if you need help with it shoot me a PM.

v--- I know your feel bro. ;)
 
Back