Overclock.net › Forums › AMD › AMD CPUs › 11hr 38min stable
New Posts  All Forums:Forum Nav:

11hr 38min stable - Page 2

post #11 of 27
Thread Starter 
Failed after 8hrs @ following voltages

CPU VID 1.525
CPU/NB 1.40
DRAM 1.65

I'm thinking of bumping the CPU VID up to 1.55 and lowering the CPU/NB down to 1.35 again to test... Raising the CPU/NB to 1.4 raised my temps almost 4C
post #12 of 27
Quote:
Originally Posted by blkdoutgsxr View Post
Failed after 8hrs @ following voltages

CPU VID 1.525
CPU/NB 1.40
DRAM 1.65

I'm thinking of bumping the CPU VID up to 1.55 and lowering the CPU/NB down to 1.35 again to test... Raising the CPU/NB to 1.4 raised my temps almost 4C
Drop your CPU/NB Volt @ 1.30V
AMD-1
(13 items)
 
AMD-2
(8 items)
 
INTEL-1
(8 items)
 
CPUMotherboardGraphicsRAM
FX-8350 ASUS CrosshairV Formula Power Color AX7970 G.Skill Ripjaws Z 2400 10-12-12-31 
Hard DriveCoolingOSPower
A-Data 256G EK Supreme HF, Water Chiller Windows 8 x64 Seasonic Gold X850 
Case
LianLi PC-A77 
CPUMotherboardGraphicsRAM
FX-8320 Gigabyte 990FXA-UD7 Sapphire HD6970 1600 11-11-11-28 
Hard DriveCoolingPowerCase
OCZ Vertex3 120Gb Heatkiller 3.0, GTX140 + GTX120 Seasonic Gold X650 Lian Li PC-7FN 
CPUMotherboardGraphicsRAM
i7-3770K Maximus V Gene Sapphire HD7950 OCZ Platinum 1600 7-7-7-24 
Hard DriveCoolingPower
SanDisk 256G Enermax ETD-T60-TB 650W 
  hide details  
Reply
AMD-1
(13 items)
 
AMD-2
(8 items)
 
INTEL-1
(8 items)
 
CPUMotherboardGraphicsRAM
FX-8350 ASUS CrosshairV Formula Power Color AX7970 G.Skill Ripjaws Z 2400 10-12-12-31 
Hard DriveCoolingOSPower
A-Data 256G EK Supreme HF, Water Chiller Windows 8 x64 Seasonic Gold X850 
Case
LianLi PC-A77 
CPUMotherboardGraphicsRAM
FX-8320 Gigabyte 990FXA-UD7 Sapphire HD6970 1600 11-11-11-28 
Hard DriveCoolingPowerCase
OCZ Vertex3 120Gb Heatkiller 3.0, GTX140 + GTX120 Seasonic Gold X650 Lian Li PC-7FN 
CPUMotherboardGraphicsRAM
i7-3770K Maximus V Gene Sapphire HD7950 OCZ Platinum 1600 7-7-7-24 
Hard DriveCoolingPower
SanDisk 256G Enermax ETD-T60-TB 650W 
  hide details  
Reply
post #13 of 27
Yeah fine tuning OC's when instability is in the 8+ hrs in prime95 is a lesson in patience.

I have this instability too and upping voltage on CPU-NB allows prime95 to run longer without BSOD or losing workers (after the 8+ hr mark, before system are fine at lower voltage).

Also try using the custom settings to stress the IMC. After around 8 hrs (for me) prime95 switches to large ftt tests so you can just use custom settings to run those tests if instability only occurs at the 8+ hr mark.

Also weirdly though, I did find voltage settings that would pass the large ftt test but not the 512k test. Can't find any information on this aside that you should run the 512k test.

I'm thinking about scaling back my NB freq to lower the needed voltage.

I was prime95 stable at 2400 at 1.3 for 24 hrs. Yesterday I was at 2600 at 1.375 (got up to this voltage after 3 days of crashing at the 8+ hr mark in prime95 blend or 2-3 hrs in prime95 custom large ftts). Woke up today and lost a worker so trying 1.4 v to see temps before I scale back.

All the people running such low voltage on their CPU-NB and high freq can you show me a 24 hr prime95 blend stable system? I have a feeling most people are skimping on the large FTT tests.

I can run prime95 on lower CPU-NB voltage just can't pass prime95 blend 8+ hrs (time seems to vary on voltage).
Edited by evilferret - 1/15/11 at 11:15am
Main Rig
(17 items)
 
*Not so* Main Rig
(13 items)
 
 
CPUMotherboardGraphicsRAM
I7-2600k Asus Maximus Gene-Z Zotac 970 32 gig (4x8gig) Cosair Dominator 1866 CL9 
Hard DriveHard DriveHard DriveOptical Drive
Samsung 830 - 128 gig VelociRaptor 300gig Random Toshiba 3TB Liteon 24x DVD-RW OEM  
CoolingOSMonitorKeyboard
Cosair H100 Windows 7 Pro x64 Xstar 1440p Realforce TKL 
PowerCaseMouseAudio
Seasonic x750 Arc Mini RAT7 Fiio E10 
CPUMotherboardGraphicsRAM
Phenom x4 955 BE Gigabyte 870a-UD3 XFX 6870 4 gig G-skills 1600 mhz CL9 
Hard DriveOptical DriveOSMonitor
64 gig Microcenter SSD / 1TB Samsung F3 Random DVD-RW W7 64 Bit Ultimate Asus VH236H 
KeyboardPowerCase
U9BL Cosair 750 TX CoolerMaster 690 II 
  hide details  
Reply
Main Rig
(17 items)
 
*Not so* Main Rig
(13 items)
 
 
CPUMotherboardGraphicsRAM
I7-2600k Asus Maximus Gene-Z Zotac 970 32 gig (4x8gig) Cosair Dominator 1866 CL9 
Hard DriveHard DriveHard DriveOptical Drive
Samsung 830 - 128 gig VelociRaptor 300gig Random Toshiba 3TB Liteon 24x DVD-RW OEM  
CoolingOSMonitorKeyboard
Cosair H100 Windows 7 Pro x64 Xstar 1440p Realforce TKL 
PowerCaseMouseAudio
Seasonic x750 Arc Mini RAT7 Fiio E10 
CPUMotherboardGraphicsRAM
Phenom x4 955 BE Gigabyte 870a-UD3 XFX 6870 4 gig G-skills 1600 mhz CL9 
Hard DriveOptical DriveOSMonitor
64 gig Microcenter SSD / 1TB Samsung F3 Random DVD-RW W7 64 Bit Ultimate Asus VH236H 
KeyboardPowerCase
U9BL Cosair 750 TX CoolerMaster 690 II 
  hide details  
Reply
post #14 of 27
Thread Starter 
Quote:
Originally Posted by IOSEFINI View Post
Drop your CPU/NB Volt @ 1.30V

This is where I get confused, because 1/2 the people say raise it and 1/2 say lower it lol.
post #15 of 27
move your cpu-nb up to 1.4volts and then move the cpu voltage down to 1.45 to 1.475 for 3.9ghz overclock. thats too much voltage. afterward...tighten up your timing a bit. those are pretty horrible timing for an amd setup. see if you can do 7-7-7-20-27...if not then try 7-8-7-21-28, then try 8-8-8-22-30, still no go try 8-9-8-24-32. this is running at 1333mhz. leave your rams voltage at 1.65.
Sandy Bridge E
(14 items)
 
  
CPUMotherboardGraphicsRAM
Intel 3930K @ 4.6ghz HT MSI Big Bang-XPower II LGA 2011 Intel X79  2x5870 CFX 2GB(4GB) Kingston HyperX Predator Series 16gb [11-12-11-... 
Hard DriveHard DriveCoolingOS
Crucial M500 240GB  TOSHIBA 2TB 7200 RPM 64MB Cache Swiftech Apogee™ XT + Arctica 900 Btu WaterChiller Windows 8.1 Ultimate 64 
MonitorKeyboardPowerCase
Samsung syncMaster 23" LED BX2331 (dual) Corsair Vengeance K90  OCZ ZX Series 1000W Fully-Modular 80PLUS Gold Corsair Obsidian Series 800D modified 
MouseAudio
Logitech® G9x Laser Mouse Realtek ALC898 8  
  hide details  
Reply
Sandy Bridge E
(14 items)
 
  
CPUMotherboardGraphicsRAM
Intel 3930K @ 4.6ghz HT MSI Big Bang-XPower II LGA 2011 Intel X79  2x5870 CFX 2GB(4GB) Kingston HyperX Predator Series 16gb [11-12-11-... 
Hard DriveHard DriveCoolingOS
Crucial M500 240GB  TOSHIBA 2TB 7200 RPM 64MB Cache Swiftech Apogee™ XT + Arctica 900 Btu WaterChiller Windows 8.1 Ultimate 64 
MonitorKeyboardPowerCase
Samsung syncMaster 23" LED BX2331 (dual) Corsair Vengeance K90  OCZ ZX Series 1000W Fully-Modular 80PLUS Gold Corsair Obsidian Series 800D modified 
MouseAudio
Logitech® G9x Laser Mouse Realtek ALC898 8  
  hide details  
Reply
post #16 of 27
Run a prime95 custom test with large ftts and see if you're stable. It'll cut the first 8 hrs from your stability testing (if you're sure your CPU is stable).

And up/down voltage as you want. For me I need a 24 hr prime95 stable system and I'm sitting at 1.4vs right now. At 1.3v I would pass prime95 till the 8+ hr mark and than it was a crap shoot.

As I mentioned, I doubt the people saying 1.3v will pass prime95 for 24 hrs (unless they played with the FSB to get the higher freq).

Also finding if I run 1T for ram, I need more voltage on the CPU-NB. Does kinda make sense since the IMC has to communicate every clock instead of every 2 clocks.

I prefer 1T with looser timings over 2T with tighter timings so that could be an option for you.

Don't worry too much about your voltage. It seems fine to me. I had to jump to 1.55 voltage to have 3.8 ghz stable but I those voltages was too high for me so I pulled back and at 3.6 ghz stable with 1.45.

I was "stable" at stock voltages at 3.8ghz but gaming + prime95 started random restarts. prime95 + MSI kombuster caused restarts at the 6 hr mark without the extra voltage. And weirdly it passed prime95 by itself 24 hrs.

I guess it depends on your level of stress testing.
Edited by evilferret - 1/15/11 at 11:37am
Main Rig
(17 items)
 
*Not so* Main Rig
(13 items)
 
 
CPUMotherboardGraphicsRAM
I7-2600k Asus Maximus Gene-Z Zotac 970 32 gig (4x8gig) Cosair Dominator 1866 CL9 
Hard DriveHard DriveHard DriveOptical Drive
Samsung 830 - 128 gig VelociRaptor 300gig Random Toshiba 3TB Liteon 24x DVD-RW OEM  
CoolingOSMonitorKeyboard
Cosair H100 Windows 7 Pro x64 Xstar 1440p Realforce TKL 
PowerCaseMouseAudio
Seasonic x750 Arc Mini RAT7 Fiio E10 
CPUMotherboardGraphicsRAM
Phenom x4 955 BE Gigabyte 870a-UD3 XFX 6870 4 gig G-skills 1600 mhz CL9 
Hard DriveOptical DriveOSMonitor
64 gig Microcenter SSD / 1TB Samsung F3 Random DVD-RW W7 64 Bit Ultimate Asus VH236H 
KeyboardPowerCase
U9BL Cosair 750 TX CoolerMaster 690 II 
  hide details  
Reply
Main Rig
(17 items)
 
*Not so* Main Rig
(13 items)
 
 
CPUMotherboardGraphicsRAM
I7-2600k Asus Maximus Gene-Z Zotac 970 32 gig (4x8gig) Cosair Dominator 1866 CL9 
Hard DriveHard DriveHard DriveOptical Drive
Samsung 830 - 128 gig VelociRaptor 300gig Random Toshiba 3TB Liteon 24x DVD-RW OEM  
CoolingOSMonitorKeyboard
Cosair H100 Windows 7 Pro x64 Xstar 1440p Realforce TKL 
PowerCaseMouseAudio
Seasonic x750 Arc Mini RAT7 Fiio E10 
CPUMotherboardGraphicsRAM
Phenom x4 955 BE Gigabyte 870a-UD3 XFX 6870 4 gig G-skills 1600 mhz CL9 
Hard DriveOptical DriveOSMonitor
64 gig Microcenter SSD / 1TB Samsung F3 Random DVD-RW W7 64 Bit Ultimate Asus VH236H 
KeyboardPowerCase
U9BL Cosair 750 TX CoolerMaster 690 II 
  hide details  
Reply
post #17 of 27
Quote:
Originally Posted by blkdoutgsxr View Post
This is where I get confused, because 1/2 the people say raise it and 1/2 say lower it lol.
You were
11 hours stable @ 1.35V
8 hours stable @ 1.40V

I just figured that dropping it @ 1.30V .... 11+ hours.

My old 955 needed 1.225V for CPU/NB @ 2600(Crosshair III). Unfortunately , Everest did not read the CPU/NB voltage at that time(no proof), so its up to you if you believe it or not
AMD-1
(13 items)
 
AMD-2
(8 items)
 
INTEL-1
(8 items)
 
CPUMotherboardGraphicsRAM
FX-8350 ASUS CrosshairV Formula Power Color AX7970 G.Skill Ripjaws Z 2400 10-12-12-31 
Hard DriveCoolingOSPower
A-Data 256G EK Supreme HF, Water Chiller Windows 8 x64 Seasonic Gold X850 
Case
LianLi PC-A77 
CPUMotherboardGraphicsRAM
FX-8320 Gigabyte 990FXA-UD7 Sapphire HD6970 1600 11-11-11-28 
Hard DriveCoolingPowerCase
OCZ Vertex3 120Gb Heatkiller 3.0, GTX140 + GTX120 Seasonic Gold X650 Lian Li PC-7FN 
CPUMotherboardGraphicsRAM
i7-3770K Maximus V Gene Sapphire HD7950 OCZ Platinum 1600 7-7-7-24 
Hard DriveCoolingPower
SanDisk 256G Enermax ETD-T60-TB 650W 
  hide details  
Reply
AMD-1
(13 items)
 
AMD-2
(8 items)
 
INTEL-1
(8 items)
 
CPUMotherboardGraphicsRAM
FX-8350 ASUS CrosshairV Formula Power Color AX7970 G.Skill Ripjaws Z 2400 10-12-12-31 
Hard DriveCoolingOSPower
A-Data 256G EK Supreme HF, Water Chiller Windows 8 x64 Seasonic Gold X850 
Case
LianLi PC-A77 
CPUMotherboardGraphicsRAM
FX-8320 Gigabyte 990FXA-UD7 Sapphire HD6970 1600 11-11-11-28 
Hard DriveCoolingPowerCase
OCZ Vertex3 120Gb Heatkiller 3.0, GTX140 + GTX120 Seasonic Gold X650 Lian Li PC-7FN 
CPUMotherboardGraphicsRAM
i7-3770K Maximus V Gene Sapphire HD7950 OCZ Platinum 1600 7-7-7-24 
Hard DriveCoolingPower
SanDisk 256G Enermax ETD-T60-TB 650W 
  hide details  
Reply
post #18 of 27
Quote:
Originally Posted by IOSEFINI View Post
You were
11 hours stable @ 1.35V
8 hours stable @ 1.40V

I just figured that dropping it @ 1.30V .... 11+ hours.

My old 955 needed 1.225V for CPU/NB @ 2600(Crosshair III). Unfortunately , Everest did not read the CPU/NB voltage at that time(no proof), so its up to you if you believe it or not
I'm finding with the large FTT test to be a complete crap shoot on the x4 955 BE time-wise. Time of stability seems to have no correlation to if it will pass the 24 hr mark. It seems to be more related to the size of the test. I had lower voltages that passed the large ftt test but wouldn't pass the 512k test (would pass large ftts but would fail about 6 hrs in on prime95 blend).

This is one of the reasons this chip is driving me insane. I would have thought the same as you if my x4 955 BE wasn't be finicky with the IMC voltage.

I thought the same as you between 1.25-1.35 v. I found instability started earlier with more voltage so tried lowering it but couldn't pass a 24 hr run. After going past 1.375 I found stability increased in length again. I'm at 1.4v and so far 11 hrs in.

Also found sometimes increasing CPU voltage helps too oddly.
Edited by evilferret - 1/15/11 at 11:51am
Main Rig
(17 items)
 
*Not so* Main Rig
(13 items)
 
 
CPUMotherboardGraphicsRAM
I7-2600k Asus Maximus Gene-Z Zotac 970 32 gig (4x8gig) Cosair Dominator 1866 CL9 
Hard DriveHard DriveHard DriveOptical Drive
Samsung 830 - 128 gig VelociRaptor 300gig Random Toshiba 3TB Liteon 24x DVD-RW OEM  
CoolingOSMonitorKeyboard
Cosair H100 Windows 7 Pro x64 Xstar 1440p Realforce TKL 
PowerCaseMouseAudio
Seasonic x750 Arc Mini RAT7 Fiio E10 
CPUMotherboardGraphicsRAM
Phenom x4 955 BE Gigabyte 870a-UD3 XFX 6870 4 gig G-skills 1600 mhz CL9 
Hard DriveOptical DriveOSMonitor
64 gig Microcenter SSD / 1TB Samsung F3 Random DVD-RW W7 64 Bit Ultimate Asus VH236H 
KeyboardPowerCase
U9BL Cosair 750 TX CoolerMaster 690 II 
  hide details  
Reply
Main Rig
(17 items)
 
*Not so* Main Rig
(13 items)
 
 
CPUMotherboardGraphicsRAM
I7-2600k Asus Maximus Gene-Z Zotac 970 32 gig (4x8gig) Cosair Dominator 1866 CL9 
Hard DriveHard DriveHard DriveOptical Drive
Samsung 830 - 128 gig VelociRaptor 300gig Random Toshiba 3TB Liteon 24x DVD-RW OEM  
CoolingOSMonitorKeyboard
Cosair H100 Windows 7 Pro x64 Xstar 1440p Realforce TKL 
PowerCaseMouseAudio
Seasonic x750 Arc Mini RAT7 Fiio E10 
CPUMotherboardGraphicsRAM
Phenom x4 955 BE Gigabyte 870a-UD3 XFX 6870 4 gig G-skills 1600 mhz CL9 
Hard DriveOptical DriveOSMonitor
64 gig Microcenter SSD / 1TB Samsung F3 Random DVD-RW W7 64 Bit Ultimate Asus VH236H 
KeyboardPowerCase
U9BL Cosair 750 TX CoolerMaster 690 II 
  hide details  
Reply
post #19 of 27
Quote:
Originally Posted by evilferret View Post
I'm finding with the large FTT test to be a complete crap shoot on the x4 955 BE time-wise. Time of stability seems to have no correlation to if it will pass the 24 hr mark.

This is one of the reasons this chip is driving me insane. I would have thought the same as you if my x4 955 BE wasn't be finicky with the IMC voltage.

I thought the same as you between 1.25-1.35 v. I found instability started earlier with more voltage so tried lowering it but couldn't pass a 24 hr run. After going past 1.375 I found stability increased in length again. I'm at 1.4v and so far 11 hrs in.

Also found sometimes increasing CPU voltage helps too oddly.
Yes, you're right. CPU volt helps(sometimes).
RAM voltage helps too
I run P95-blend. Always.
AMD-1
(13 items)
 
AMD-2
(8 items)
 
INTEL-1
(8 items)
 
CPUMotherboardGraphicsRAM
FX-8350 ASUS CrosshairV Formula Power Color AX7970 G.Skill Ripjaws Z 2400 10-12-12-31 
Hard DriveCoolingOSPower
A-Data 256G EK Supreme HF, Water Chiller Windows 8 x64 Seasonic Gold X850 
Case
LianLi PC-A77 
CPUMotherboardGraphicsRAM
FX-8320 Gigabyte 990FXA-UD7 Sapphire HD6970 1600 11-11-11-28 
Hard DriveCoolingPowerCase
OCZ Vertex3 120Gb Heatkiller 3.0, GTX140 + GTX120 Seasonic Gold X650 Lian Li PC-7FN 
CPUMotherboardGraphicsRAM
i7-3770K Maximus V Gene Sapphire HD7950 OCZ Platinum 1600 7-7-7-24 
Hard DriveCoolingPower
SanDisk 256G Enermax ETD-T60-TB 650W 
  hide details  
Reply
AMD-1
(13 items)
 
AMD-2
(8 items)
 
INTEL-1
(8 items)
 
CPUMotherboardGraphicsRAM
FX-8350 ASUS CrosshairV Formula Power Color AX7970 G.Skill Ripjaws Z 2400 10-12-12-31 
Hard DriveCoolingOSPower
A-Data 256G EK Supreme HF, Water Chiller Windows 8 x64 Seasonic Gold X850 
Case
LianLi PC-A77 
CPUMotherboardGraphicsRAM
FX-8320 Gigabyte 990FXA-UD7 Sapphire HD6970 1600 11-11-11-28 
Hard DriveCoolingPowerCase
OCZ Vertex3 120Gb Heatkiller 3.0, GTX140 + GTX120 Seasonic Gold X650 Lian Li PC-7FN 
CPUMotherboardGraphicsRAM
i7-3770K Maximus V Gene Sapphire HD7950 OCZ Platinum 1600 7-7-7-24 
Hard DriveCoolingPower
SanDisk 256G Enermax ETD-T60-TB 650W 
  hide details  
Reply
post #20 of 27
Thread Starter 
Quote:
Originally Posted by IOSEFINI View Post
You were
11 hours stable @ 1.35V
8 hours stable @ 1.40V

I just figured that dropping it @ 1.30V .... 11+ hours.

My old 955 needed 1.225V for CPU/NB @ 2600(Crosshair III). Unfortunately , Everest did not read the CPU/NB voltage at that time(no proof), so its up to you if you believe it or not

I was going to bed and ran out of time and did a 'not good thing' while overclocking and changed more than one variable, I upped the FSB because my goal is actually 4.0... the 8hr was @ 3.95 and the 11hr was @ 3.9
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: AMD CPUs
Overclock.net › Forums › AMD › AMD CPUs › 11hr 38min stable