Overclock.net › Forums › AMD › AMD CPUs › ASRock 990FX Fatal1ty overclock settings
New Posts  All Forums:Forum Nav:

ASRock 990FX Fatal1ty overclock settings - Page 2

post #11 of 48
Quote:
Originally Posted by computerparts View Post

Something is wrong. I was able to get 4.5 ghz on 1.4v just by increasing the multi. What bios version do you have? I'm running version 1.5.Also what cooler do you have on your 8120?

Well, I just reinstalled the 1.5 bios to make sure. I thought I had upgrades the bios initially with a 955 chip.

Regardless, I just ran prime for like 6-7 tests under 20x @ 1.45 volts. Maybe I just botched the bios upgrades when I tried doing it with a 955. So I will continue my overclocking efforts.

My current cooler is a Zalman CNPS9900ALED but I have a Phanteks in the mail
Science Oven
(15 items)
 
  
CPUMotherboardGraphicsRAM
2500k p8z77 v pro asus 780 ripjaws 
Hard DriveHard DriveCoolingCooling
seagate Crucial m4 h100i spectre 
CoolingOSKeyboardPower
cougar pwm windows 8.1 Ducky Shine 3 cooler master 850w silent 
CaseMouseAudio
600t razer 2014 Asus Xonar 
  hide details  
Reply
Science Oven
(15 items)
 
  
CPUMotherboardGraphicsRAM
2500k p8z77 v pro asus 780 ripjaws 
Hard DriveHard DriveCoolingCooling
seagate Crucial m4 h100i spectre 
CoolingOSKeyboardPower
cougar pwm windows 8.1 Ducky Shine 3 cooler master 850w silent 
CaseMouseAudio
600t razer 2014 Asus Xonar 
  hide details  
Reply
post #12 of 48
Thread Starter 
Quote:
Originally Posted by Trev0r269 View Post

Well, I just reinstalled the 1.5 bios to make sure. I thought I had upgrades the bios initially with a 955 chip.
Regardless, I just ran prime for like 6-7 tests under 20x @ 1.45 volts. Maybe I just botched the bios upgrades when I tried doing it with a 955. So I will continue my overclocking efforts.
My current cooler is a Zalman CNPS9900ALED but I have a Phanteks in the mail

With your Cooler you should easily get 4.5Ghz.... just it going to be a bit noise fun at full speed.
Try settings I provided at the first post.
Make sure you instal CoreTemp gadget to control CPU temperature.
Ice and Fire
(13 items)
 
Toxic Bulldozer
(14 items)
 
 
CPUMotherboardGraphicsGraphics
FX-8120/4.8Ghz-1.48v ASUS M5A97evo Gigabyte HD 6970 SAPPHIRE HD 6950 TOXIC 960/5800 
RAMHard DriveOptical DriveCooling
A-Data 2000g Vortex 3 120Gb BR-LG BH10LS30 Corsair H100 Mod with 2 water tanks 
OSMonitorPowerCase
Windows7-64bit ASUS 24"ms238 & ACER19" Antec TruePower Quattro TPQ-1000 Cooler Master 690II Adv 
MouseAudio
Laser 2000dpi Audigy 2 Front Panel 
  hide details  
Reply
Ice and Fire
(13 items)
 
Toxic Bulldozer
(14 items)
 
 
CPUMotherboardGraphicsGraphics
FX-8120/4.8Ghz-1.48v ASUS M5A97evo Gigabyte HD 6970 SAPPHIRE HD 6950 TOXIC 960/5800 
RAMHard DriveOptical DriveCooling
A-Data 2000g Vortex 3 120Gb BR-LG BH10LS30 Corsair H100 Mod with 2 water tanks 
OSMonitorPowerCase
Windows7-64bit ASUS 24"ms238 & ACER19" Antec TruePower Quattro TPQ-1000 Cooler Master 690II Adv 
MouseAudio
Laser 2000dpi Audigy 2 Front Panel 
  hide details  
Reply
post #13 of 48
Quote:
Originally Posted by Kolovrat View Post

With your Cooler you should easily get 4.5Ghz.... just it going to be a bit noise fun at full speed.
Try settings I provided at the first post.
Make sure you instal CoreTemp gadget to control CPU temperature.

I used the settings that you posted in the first post. Thanks for the screenshots, those make it all a bit easier. I tried that overclock from the first page and I couldn't boot into windows. I cranked out some iterations with the cpuV being between 1.4-1.515 and cpu multiplier ranging from 20x (4.0) to 23.5x. Nothing seems stable (BSODs, nigh immediate freezing in prime), with any setting for LLC. On the prime95 tests I was able to start, I noticed the first 2 cores would stop working relatively early on, like during or after test 1.

Next, I decided to see those 2 cores would also stop working on lesser overclock. I ran a prime blend test on 19x (3.8) @ 1.35v and the same thing happened; workers #1 and #2 stop.

There isn't anything goin on crazy with temps, I've been monitoring those from the beginning. I'm using bios 1.5.

What do you guys think? Should I RMA the chip, or am I getting ahead of myself? Thanks for your help everyone.
Science Oven
(15 items)
 
  
CPUMotherboardGraphicsRAM
2500k p8z77 v pro asus 780 ripjaws 
Hard DriveHard DriveCoolingCooling
seagate Crucial m4 h100i spectre 
CoolingOSKeyboardPower
cougar pwm windows 8.1 Ducky Shine 3 cooler master 850w silent 
CaseMouseAudio
600t razer 2014 Asus Xonar 
  hide details  
Reply
Science Oven
(15 items)
 
  
CPUMotherboardGraphicsRAM
2500k p8z77 v pro asus 780 ripjaws 
Hard DriveHard DriveCoolingCooling
seagate Crucial m4 h100i spectre 
CoolingOSKeyboardPower
cougar pwm windows 8.1 Ducky Shine 3 cooler master 850w silent 
CaseMouseAudio
600t razer 2014 Asus Xonar 
  hide details  
Reply
post #14 of 48
Thread Starter 
From my readings every FX chip can overclock higher than 4.5Ghz , and I'm talking about all FX series 4100/6100/8120
Try to reinstall your processor again, reset your BOARD, reflash BIOS, things happens and it's looks like something went wrong.
If no luck at the second time then RMA this chip faster as possible.
Quote:
Originally Posted by Trev0r269 View Post

I used the settings that you posted in the first post. Thanks for the screenshots, those make it all a bit easier. I tried that overclock from the first page and I couldn't boot into windows. I cranked out some iterations with the cpuV being between 1.4-1.515 and cpu multiplier ranging from 20x (4.0) to 23.5x. Nothing seems stable (BSODs, nigh immediate freezing in prime), with any setting for LLC. On the prime95 tests I was able to start, I noticed the first 2 cores would stop working relatively early on, like during or after test 1.
Next, I decided to see those 2 cores would also stop working on lesser overclock. I ran a prime blend test on 19x (3.8) @ 1.35v and the same thing happened; workers #1 and #2 stop.
There isn't anything goin on crazy with temps, I've been monitoring those from the beginning. I'm using bios 1.5.
What do you guys think? Should I RMA the chip, or am I getting ahead of myself? Thanks for your help everyone.
Ice and Fire
(13 items)
 
Toxic Bulldozer
(14 items)
 
 
CPUMotherboardGraphicsGraphics
FX-8120/4.8Ghz-1.48v ASUS M5A97evo Gigabyte HD 6970 SAPPHIRE HD 6950 TOXIC 960/5800 
RAMHard DriveOptical DriveCooling
A-Data 2000g Vortex 3 120Gb BR-LG BH10LS30 Corsair H100 Mod with 2 water tanks 
OSMonitorPowerCase
Windows7-64bit ASUS 24"ms238 & ACER19" Antec TruePower Quattro TPQ-1000 Cooler Master 690II Adv 
MouseAudio
Laser 2000dpi Audigy 2 Front Panel 
  hide details  
Reply
Ice and Fire
(13 items)
 
Toxic Bulldozer
(14 items)
 
 
CPUMotherboardGraphicsGraphics
FX-8120/4.8Ghz-1.48v ASUS M5A97evo Gigabyte HD 6970 SAPPHIRE HD 6950 TOXIC 960/5800 
RAMHard DriveOptical DriveCooling
A-Data 2000g Vortex 3 120Gb BR-LG BH10LS30 Corsair H100 Mod with 2 water tanks 
OSMonitorPowerCase
Windows7-64bit ASUS 24"ms238 & ACER19" Antec TruePower Quattro TPQ-1000 Cooler Master 690II Adv 
MouseAudio
Laser 2000dpi Audigy 2 Front Panel 
  hide details  
Reply
post #15 of 48
Quote:
Originally Posted by Trev0r269 View Post

I used the settings that you posted in the first post. Thanks for the screenshots, those make it all a bit easier. I tried that overclock from the first page and I couldn't boot into windows. I cranked out some iterations with the cpuV being between 1.4-1.515 and cpu multiplier ranging from 20x (4.0) to 23.5x. Nothing seems stable (BSODs, nigh immediate freezing in prime), with any setting for LLC. On the prime95 tests I was able to start, I noticed the first 2 cores would stop working relatively early on, like during or after test 1.
Next, I decided to see those 2 cores would also stop working on lesser overclock. I ran a prime blend test on 19x (3.8) @ 1.35v and the same thing happened; workers #1 and #2 stop.
There isn't anything goin on crazy with temps, I've been monitoring those from the beginning. I'm using bios 1.5.
What do you guys think? Should I RMA the chip, or am I getting ahead of myself? Thanks for your help everyone.

I would recommend replacing that Zalman asap, it's not a sufficient heatsink for an 8 core cpu. What are your temps? Make sure you're monitoring temps with core temp.
post #16 of 48
Quote:
Originally Posted by computerparts View Post

I would recommend replacing that Zalman asap, it's not a sufficient heatsink for an 8 core cpu. What are your temps? Make sure you're monitoring temps with core temp.

I have a Phanteks comes in the mail this week. This fan

Also, I remounted my mobo (which was missing a screw + standoff for one of the mobo mounting holes thinking.gif . I also re-seated the 8120 and zalman over it as recommended by Kolovrat, I also flashed the bios to 1.5 again.

As I type this. I'm running prime on 1.45 v and 4.0 mhz. After test #3, CoreTemp has cpu #0 as 50C. Speedfan has the CPU @ 71c. I'm not sure which to believe. Hardware monitor says the 8 cores for the 8120 are 50 (agrees with core temp), and HWM also says CPUTIN under the motherboard menu tree says 71C. Whats the difference between core temps and CPUTIN?

Next, I guess I'll see if I can crank the multipler @ 1.45v. Unless you guys have any other plans of action or questions.

Edit: After being frustrated and realizing this is the last day that I can RMA the 8120, I suited it up and took it to UPS. That chip just didn't want to be stable at what seemed like any cpuV 1.4+ and clock frequencies over 4.0. I opted to just put the reliable 955 be back in (which is priming @ 4.0 mhz as we speak). I'm also seeing much more continuity from the all monitoring programs. as far as temperatures being reported. Hopefully my next 8120 can be safely OC'd to 4.0+, but for whatever reason, that last one wasn't..

thanks for everyone's input and hopefully the next 8120 along with a phateks cooler will be able to give me some decently high oc's.
Edited by Trev0r269 - 12/28/11 at 3:59pm
Science Oven
(15 items)
 
  
CPUMotherboardGraphicsRAM
2500k p8z77 v pro asus 780 ripjaws 
Hard DriveHard DriveCoolingCooling
seagate Crucial m4 h100i spectre 
CoolingOSKeyboardPower
cougar pwm windows 8.1 Ducky Shine 3 cooler master 850w silent 
CaseMouseAudio
600t razer 2014 Asus Xonar 
  hide details  
Reply
Science Oven
(15 items)
 
  
CPUMotherboardGraphicsRAM
2500k p8z77 v pro asus 780 ripjaws 
Hard DriveHard DriveCoolingCooling
seagate Crucial m4 h100i spectre 
CoolingOSKeyboardPower
cougar pwm windows 8.1 Ducky Shine 3 cooler master 850w silent 
CaseMouseAudio
600t razer 2014 Asus Xonar 
  hide details  
Reply
post #17 of 48
Thread Starter 
Good luck on the New Year Chip smile.gif
I'm sure it's going to be PERFECT!!!
Quote:
Originally Posted by Trev0r269 View Post

I have a Phanteks comes in the mail this week. This fan
Also, I remounted my mobo (which was missing a screw + standoff for one of the mobo mounting holes thinking.gif . I also re-seated the 8120 and zalman over it as recommended by Kolovrat, I also flashed the bios to 1.5 again.
As I type this. I'm running prime on 1.45 v and 4.0 mhz. After test #3, CoreTemp has cpu #0 as 50C. Speedfan has the CPU @ 71c. I'm not sure which to believe. Hardware monitor says the 8 cores for the 8120 are 50 (agrees with core temp), and HWM also says CPUTIN under the motherboard menu tree says 71C. Whats the difference between core temps and CPUTIN?
Next, I guess I'll see if I can crank the multipler @ 1.45v. Unless you guys have any other plans of action or questions.
Edit: After being frustrated and realizing this is the last day that I can RMA the 8120, I suited it up and took it to UPS. That chip just didn't want to be stable at what seemed like any cpuV 1.4+ and clock frequencies over 4.0. I opted to just put the reliable 955 be back in (which is priming @ 4.0 mhz as we speak). I'm also seeing much more continuity from the all monitoring programs. as far as temperatures being reported. Hopefully my next 8120 can be safely OC'd to 4.0+, but for whatever reason, that last one wasn't..
thanks for everyone's input and hopefully the next 8120 along with a phateks cooler will be able to give me some decently high oc's.

Edited by Kolovrat - 12/29/11 at 3:36am
Ice and Fire
(13 items)
 
Toxic Bulldozer
(14 items)
 
 
CPUMotherboardGraphicsGraphics
FX-8120/4.8Ghz-1.48v ASUS M5A97evo Gigabyte HD 6970 SAPPHIRE HD 6950 TOXIC 960/5800 
RAMHard DriveOptical DriveCooling
A-Data 2000g Vortex 3 120Gb BR-LG BH10LS30 Corsair H100 Mod with 2 water tanks 
OSMonitorPowerCase
Windows7-64bit ASUS 24"ms238 & ACER19" Antec TruePower Quattro TPQ-1000 Cooler Master 690II Adv 
MouseAudio
Laser 2000dpi Audigy 2 Front Panel 
  hide details  
Reply
Ice and Fire
(13 items)
 
Toxic Bulldozer
(14 items)
 
 
CPUMotherboardGraphicsGraphics
FX-8120/4.8Ghz-1.48v ASUS M5A97evo Gigabyte HD 6970 SAPPHIRE HD 6950 TOXIC 960/5800 
RAMHard DriveOptical DriveCooling
A-Data 2000g Vortex 3 120Gb BR-LG BH10LS30 Corsair H100 Mod with 2 water tanks 
OSMonitorPowerCase
Windows7-64bit ASUS 24"ms238 & ACER19" Antec TruePower Quattro TPQ-1000 Cooler Master 690II Adv 
MouseAudio
Laser 2000dpi Audigy 2 Front Panel 
  hide details  
Reply
post #18 of 48
Quote:
Originally Posted by Kolovrat View Post

Good luck on the New Year Chip smile.gif
I'm sure it's going to be PERFECT!!!

The new one is acceptable; it's prime stable on 1.4v @ 4.2 MHz, which is something that the last one couldn't come close to doing.

I've noticed some interesting things:

1. If I bump the voltage up over 1.4 using Overdrive, then the chip will downclock to 2.x MHz. If I set the voltage to >1.4 in the bios, then this problem does not occur. I've turned off the proper settings in the bios.

2. I played with LLC a bit and saw something funny: I cranked the voltage to 1.45 and the clock to 4.2 MHz at one point, figuring that I'd up the cpu multiplier if 1.45v & 4.2 MHz seemed OK. Well, it failed prime during the 1st test. I decided to turn LLC from disable/auto to 100%. After changing LLC to 100%, prime would run for a while but one core would stop running. Does that mean LLC 100% just sort of the chokes the V enough to in effect shut down one core?

I'll play around with it more later on. I'd like to get something with a voltage greater than 1.4 and a multiplier greater than 21x (4.2). It seems like V above 1.4 causes stability issues.
Science Oven
(15 items)
 
  
CPUMotherboardGraphicsRAM
2500k p8z77 v pro asus 780 ripjaws 
Hard DriveHard DriveCoolingCooling
seagate Crucial m4 h100i spectre 
CoolingOSKeyboardPower
cougar pwm windows 8.1 Ducky Shine 3 cooler master 850w silent 
CaseMouseAudio
600t razer 2014 Asus Xonar 
  hide details  
Reply
Science Oven
(15 items)
 
  
CPUMotherboardGraphicsRAM
2500k p8z77 v pro asus 780 ripjaws 
Hard DriveHard DriveCoolingCooling
seagate Crucial m4 h100i spectre 
CoolingOSKeyboardPower
cougar pwm windows 8.1 Ducky Shine 3 cooler master 850w silent 
CaseMouseAudio
600t razer 2014 Asus Xonar 
  hide details  
Reply
post #19 of 48
You might want to look into cooling your mosfets with a fan. I am currently having a similar problem but my phase change isnt as effective as yours but i do have more airflow in that area than you have. My problem is throttling tho when they get a certian temp.
Workspace
(12 items)
 
  
CPUMotherboardGraphicsRAM
AMD FX-8120 ASRock 880G Pro3 AMD Radeon HD 6700 Series (5770 bios flashed to... SpecTek Incorporated  
RAMRAMRAMCooling
A- Team Group Inc. Team Group Inc. Spire TherMax Eclipse II CPU Cooler 
OSMonitorPowerCase
Windows 7 Home Premium Westinghouse 1280 x 1024 PC Power & Cooling Silencer 760w 80 Plus Silver PowerUp! G54-40840 XL Mid-Tower ATX Case 
  hide details  
Reply
Workspace
(12 items)
 
  
CPUMotherboardGraphicsRAM
AMD FX-8120 ASRock 880G Pro3 AMD Radeon HD 6700 Series (5770 bios flashed to... SpecTek Incorporated  
RAMRAMRAMCooling
A- Team Group Inc. Team Group Inc. Spire TherMax Eclipse II CPU Cooler 
OSMonitorPowerCase
Windows 7 Home Premium Westinghouse 1280 x 1024 PC Power & Cooling Silencer 760w 80 Plus Silver PowerUp! G54-40840 XL Mid-Tower ATX Case 
  hide details  
Reply
post #20 of 48
Quote:
Originally Posted by Yuji View Post

You might want to look into cooling your mosfets with a fan. I am currently having a similar problem but my phase change isnt as effective as yours but i do have more airflow in that area than you have. My problem is throttling tho when they get a certian temp.

If you made a thread on that issue, I think I read it. Cooling my mosfets might be hard due to the size of the Phanteks hsf I have. That thing covers ram and mosfets, too. Although I will check when I get home to make sure.

I only noticed the weird down-clock only when I increased cpuV to something greater than 1.4 using Overdrive. I've had temps at 60 and a bit over (~65) during prime so far and no throttle down occurs then.

Side note: writing my thesis and all I want to do is go home and work on hot-rodding my computer. rolleyes.gif
Science Oven
(15 items)
 
  
CPUMotherboardGraphicsRAM
2500k p8z77 v pro asus 780 ripjaws 
Hard DriveHard DriveCoolingCooling
seagate Crucial m4 h100i spectre 
CoolingOSKeyboardPower
cougar pwm windows 8.1 Ducky Shine 3 cooler master 850w silent 
CaseMouseAudio
600t razer 2014 Asus Xonar 
  hide details  
Reply
Science Oven
(15 items)
 
  
CPUMotherboardGraphicsRAM
2500k p8z77 v pro asus 780 ripjaws 
Hard DriveHard DriveCoolingCooling
seagate Crucial m4 h100i spectre 
CoolingOSKeyboardPower
cougar pwm windows 8.1 Ducky Shine 3 cooler master 850w silent 
CaseMouseAudio
600t razer 2014 Asus Xonar 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: AMD CPUs
Overclock.net › Forums › AMD › AMD CPUs › ASRock 990FX Fatal1ty overclock settings