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

Need help with FX-4170

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
Even if it might be a software error, everything is reading slightly higher. I guess there's not much to do about it then. :shrug:
 
Even if it might be a software error, everything is reading slightly higher. I guess there's not much to do about it then. :shrug:

could it being slightly higher cause the error though?

ill double check bios but im 99.99% all of that is untouched
 
I don't know if it would make that much of a difference. You can try adding slightly more voltage, like 1.26 or 1.27.
 
I don't know if it would make that much of a difference. You can try adding slightly more voltage, like 1.26 or 1.27.

gave it a .025 bump
idk how to go any less then that, since from what i can see thats all my bios will allow me to go up or down by
so im @1.2875 per the bios
however CPUz is reading it @ 1.280

also, im going to be looking for a new heatsink/fan now as well, just to help keep it cool

im under, "normal room conditions" *i had the AC pumping pretty good with a fan pushing the cold air directly @ comp before XD* and @ 1.28v w/ 21x multiplier im at about 41*C
id s/s but im getting tired :p

anyway, hopefully this works out just fine
 
Well, at least the voltage drop brought the temps down to normal. That's less than what mine was on the stock cooler with stock speeds and undervolted. Is the CPU stable now at least?
 
Well, at least the voltage drop brought the temps down to normal. That's less than what mine was on the stock cooler with stock speeds and undervolted. Is the CPU stable now at least?

its definately better
i ended up passing out in my chair last night after getting P95 running, so it ran all night

i stopped it when i woke up but i got

"FATAL ERROR: Rounding was 0.5, expected less than 0.4
Hardware failure detected, consult stress.txt file." on core 3 after 1 hour and 52mins

but this on core 4 after 22 mins *which woulda been after i passed out :p *
FATAL ERROR: Final result was 625EC669, expected: 09C7ADE4.
Hardware failure detected, consult stress.txt file.

/edit

im going to run v25.11 of p95 today and see if i get different results on that
 
well, no matter what i did this thing just wouldnt become stable

so im shipping it out tomorrow in hopes of getting it replaced
no pins are bent and theres no damage to it so it shouldnt be a problem to replace it
if they deny it, itll do what i need it to till the Piledriver cores come out, hopefully

either way, thanks for the help alda
 
I'm sorry to hear that. :( Hopefully they're replace it since it's not even stable at it's stock speed. Good luck! :thup:
 
I'm sorry to hear that. :( Hopefully they're replace it since it's not even stable at it's stock speed. Good luck! :thup:

hopefully they do, and ill let you know what happens, and if they do replace it ill let you know how the new one is.

if i have the same problem, co-worker of mine suggested it may be the ram so ill have to look into that then if this continues with a new one
 
well RMA was approved so hopefully ill have the new core by next week thursday or friday

alda,

would u suggest i double check bios after installing and lower voltage down to 1.25 right away? cause im sure my bios will put it back to 1.4 stock again

i imagine it would be a good idea, but better to be safe not sorry XD
 
well RMA was approved so hopefully ill have the new core by next week thursday or friday

alda,

would u suggest i double check bios after installing and lower voltage down to 1.25 right away? cause im sure my bios will put it back to 1.4 stock again

i imagine it would be a good idea, but better to be safe not sorry XD

Yeah, I would try that. Glad you're getting a replacement! :thup:
 
knock on wood *knock* *knock*
replacement core is in, voltage is set at 1.275 in bios, CPU-Z and HWMonitor both say its @ 1.26 however

P95 is running, it has been running a little over 15 mins, all cores have "Self-test 720k passed!", temps are @ 43*C at the moment and so far so good

Mid Typing Edit XD
getting close to 20 mins now
 
knock on wood *knock* *knock*
replacement core is in, voltage is set at 1.275 in bios, CPU-Z and HWMonitor both say its @ 1.26 however

P95 is running, it has been running a little over 15 mins, all cores have "Self-test 720k passed!", temps are @ 43*C at the moment and so far so good

Mid Typing Edit XD
getting close to 20 mins now

I have my vCore set to 1.25 in BIOS but CPUz and HWMonitor both report it as 1.26 as well. Looks good so far, I hope no cores are defective on this one for you! I let prime95 run for about 8 hours on mine when I overclocked it to 4.2GHz. It was 32C when I went to bed and 28C when I woke up the next morning, probably from the air getting cooler during the night, but it stays perfectly stable. :D If all goes well, enjoy your new CPU! :bday:
 
all hasnt gone well
core 4 has been failing
but im not gonna immediately jump to "its defective" this time

its gotta be something else
i get it, there are lemons
but i find it difficult to believe that 2/2 cores ive had would be defective upon receiving them

im getting some other programs to test it with to see how they perform

either way, im prolly not gonna send this back since ill be building a new machine when the piledriver cores come out
hopefully XD
 
I think you've got another problem that you're going to have to address, you have two different memory kits that your running as one and that never a wise thing. I know, your going to say that they're the same make and model so they are the same but that's not true. Those two kits could have been made a month apart and have some minor differences that keep them from being the same.
 
one kit *the 8gig kit obviously*
1.3v
4.2ghz clock

what it looks like now


edit 9:29pm est

21 min run - second image
 

Attachments

  • cpu1.png
    cpu1.png
    200.9 KB · Views: 108
  • cpu2.png
    cpu2.png
    197.3 KB · Views: 106
Last edited:
40 mins in
i just noticed, and this is under the assumption that the programs are reading accurately, that the voltage even under constant load is @ 1.28v

if that drop happened @ 1.25v *which in the programs read 1.25 max* or @ 1.275 bios *which in the programs read 1.26*
that would have put me @ 1.23-1.24v

and with just the single kit in, i think its good
i am gonna let this run through the night, and start looking for a better cooler as well, cause i dont like these temps, even tho ill never be under that extreme load for very long if at all
 

Attachments

  • cpu3.png
    cpu3.png
    198.6 KB · Views: 107
40 mins in
i just noticed, and this is under the assumption that the programs are reading accurately, that the voltage even under constant load is @ 1.28v

if that drop happened @ 1.25v *which in the programs read 1.25 max* or @ 1.275 bios *which in the programs read 1.26*
that would have put me @ 1.23-1.24v

and with just the single kit in, i think its good
i am gonna let this run through the night, and start looking for a better cooler as well, cause i dont like these temps, even tho ill never be under that extreme load for very long if at all

Hello there! Just a few questions, it seems like I'm having a similar situation here... I also have an fx-4170, installed on an asus board (m5a97 pro) with 2x4gb gskill sniper 1866 cl 9-10-9.
I'm using the stock heatsink/fan, my cores aren't failing, but I noticed the temps were too high under load... on stock voltage they went up to 64°C stressing the CPU with p95 blend test, 51°C per core... seeing your post I decided to check the votages, and the vid shows a range between 0.8125min and 1.4125v max... the cpu varies between those values depending on the duty it has to perform...
1.4125v seems an overkill to run the processor on stock speed, so I decided to drop it down lo 1.25v +/- 0.022, gettin 1.272 under load, and 1.232 on idle... temps went down to 56°C cpu and 42°C per core under stress... tested stability with p95, blend test +2hours...
My question is, once you got the replacement, did you set up you voltages manually or the cpu came with 1.25 vid stock?
I'm really mad right now with my system, I'm having difficulties to drop the temps down, and it seems like the stock voltage is too high for the stock multiplier... setting the voltage manually makes me loose the oportunity to idle at 0.8125vcore... any help would be very apressiated!!

EDIT: BTW, you may want to check your PSU voltages... they seem unstable, or else, the sensors are getting wrong data... you have -12 at -14.27; -5v at -13.31; +5vcch at 3.63... +12v rail is almost close to the acceptable error limit (+/- 5% error, you still have +0.18v of headroom).
 
Last edited:
Hello there! Just a few questions, it seems like I'm having a similar situation here... I also have an fx-4170, installed on an asus board (m5a97 pro) with 2x4gb gskill sniper 1866 cl 9-10-9.
I'm using the stock heatsink/fan, my cores aren't failing, but I noticed the temps were too high under load... on stock voltage they went up to 64°C stressing the CPU with p95 blend test, 51°C per core... seeing your post I decided to check the votages, and the vid shows a range between 0.8125min and 1.4125v max... the cpu varies between those values depending on the duty it has to perform...
1.4125v seems an overkill to run the processor on stock speed, so I decided to drop it down lo 1.25v +/- 0.022, gettin 1.272 under load, and 1.232 on idle... temps went down to 56°C cpu and 42°C per core under stress... tested stability with p95, blend test +2hours...
My question is, once you got the replacement, did you set up you voltages manually or the cpu came with 1.25 vid stock?
I'm really mad right now with my system, I'm having difficulties to drop the temps down, and it seems like the stock voltage is too high for the stock multiplier... setting the voltage manually makes me loose the oportunity to idle at 0.8125vcore... any help would be very apressiated!!

EDIT: BTW, you may want to check your PSU voltages... they seem unstable, or else, the sensors are getting wrong data... you have -12 at -14.27; -5v at -13.31; +5vcch at 3.63... +12v rail is almost close to the acceptable error limit (+/- 5% error, you still have +0.18v of headroom).

The "green" settings are what slows the processor down and decreases the voltage at idle. Are they still on? If you're planning on doing some more serious overclocking then you'll want those turned off anyway, but if you're like me and your computer is sitting idle most of the time and you don't plan on doing a heavy overclock, then leaving them on is fine in my opinion.
 
Back