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

High voltage on AMD Phenom II X4 965 BE?

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

Dieter01

Registered
Joined
Aug 19, 2014
Hi,

I have no experience overclocking but based my attempts on various guides... So far I have only adjusted the multiplier from 17 to 19 and increased CPU voltage from the stock 1.400V to 1.475V. I was hoping for the machine to be stable at 1.45, but failed after 30 mins on Prime 95. Then increased in steps to 1.475V, still not stable. Max CPU temp was 46 deg C.

I understand required voltage will vary, but isn't this a bit high? Still within normal? Will the CPU age and potentially become less stable over time? My machine is from 2010. I rarely use the shut down button, turn off monitor and never put the machine to sleep...

System:
Motherboard: ASUS M4A79T Deluxe, Socket-AM3
CPU: AMD Phenom II X4 965 Black Edition Quad Core
RAM: Kingston DDR3 HyperX 1333MHz 4 x 2GB DDR3, CL7-7-7-20
Case: Fractal Design Define R2 Titanium Grey
PSU: Cooler Master Silent Pro M700 700W
GPU: XFX Radeon HD 5770 1GB GDDR5
HDD: WD VelociRaptor® 300GB SATA, 10000RPM, 16MB, 3 GB/s
Fans: 2 x Fractal Design Silent Pro 120mm (both blowing into case)
CPU cooler: Corsair H100i

I plan to swap all fans (including H100i fans) with Notcua version to improve SPL, the machine is running a bit loud. But I don't think that is necessarily relevant for my question... Is this high voltage to be expected? Is there something I might have overlooked?

I did the OC'ing in BIOS. I only changed the multiplier and CPU voltage, nothing else. Temperature readings are using Core Temp. I also tried the Corsair Link software but for some reason it is mixing the CPU temp and motherboard temp. BIOS CPU readings correspond with Core Temp though.

Thanks for helping :)
 
you say the max CPU temp was 49c. But that's only part of the temp picture. What were maximum core/package temps at that point? If you use HWMonitor to monitor temps and voltages it will give you both and both are important. Many temp monitoring programs use "CPU temp" to mean motherboard socket temp. The exception to that is the program, "CoreTemp".
 
I used CoreTemp. Max temp was 46 deg with fans on low speed. If necessary I coulf have increased fan speed a bit but since the max rated CPU temp for this processor is 62 deg I figured I was OK temperature wise.

Attached is a snapshot of my current setup. I have lowered the multiplier to 18.5. Couldn't run Prime now so the temp is at limited load as you can see. Idle temp in BIOS is about 33 deg C.
 

Attachments

  • Capture.JPG
    Capture.JPG
    40.3 KB · Views: 3,606
Just pulled the trigger on some new fans in preparation of further OC'ing...

2 Noctua NF-F12 PWM (Exhaust top of cabinet, replacing stock fans on H100i)
1 Noctua NF-S12A PWM (Exhaust, top rear of cabinet)
2 Noctua NF-P12 (intake, front of cabinet. Opted for these instead of S12 for better handling of the restricted airflow. I use the H100i for PWM and it only supports 4 fans, these will be voltage controlled through one of the MB headers).
1 Noctua NF-A14 PWM (intake, bottom of cabinet)
 
This is the how and what we need for accurate suggestions...

...This is what we need to see for sure and a real good starting point.

Normally during setup and testing we disable C1/E, C6, Cool N Quiet, APC, TurboCore and in windows performance manager itself we set to "performance" mode. That way there are not "other" settings messing with the overclocking process. Some of those settings are not available on all models of cpu but where in evidence we disable for setup of overclock process.

CPU Tab in CPUz from CPUID com
attachment.php


Memory Tab in CPUz from CPUID com
attachment.php


SPD Tab in CPUz from CPUID com
attachment.php


And this is screen capture of HWMonitor (free version) from CPUID com
HWMonitor has been scrolled enough and large enough to show Min/Max of Voltages and includes the CPU CORE TEMPS / "package" temps fully visible. Latest versions of HWMonitor show the CPU Core Temp as " a Package Temp" and is only shown as a single temp since there were never multple, individual core temp sensors anyway.

This capture is made of HWMonitor after it has been open and running on the desktop logging Min/Max temps and voltages while Prime 95 was running Blend Mode test on all cores for at least 20 mins and then the capture of HWMonitor was made and it shows the Min/Max temps and voltages before P95 Blend was started and while running P95 Blend mode and gives much greater insight into how the system is performing without guessing.

attachment.php


In order to attach screenshots of INDIVIDUAL images as suggested, first crop and capture the images with Snipping Tool found in Windows Accessories or equivalent. Then click on Go Advanced, a button at the bottom of every new post window. Then click on the little paperclip tool at the top of the Advanced post window when it opens. Clicking on the paperclip tool brings up the file browser/upload tool and the rest is fairly obvious.
 
Hi,

I have no experience overclocking but based my attempts on various guides... So far I have only adjusted the multiplier from 17 to 19 and increased CPU voltage from the stock 1.400V to 1.475V. I was hoping for the machine to be stable at 1.45, but failed after 30 mins on Prime 95. Then increased in steps to 1.475V, still not stable. Max CPU temp was 46 deg C.

I understand required voltage will vary, but isn't this a bit high? Still within normal? Will the CPU age and potentially become less stable over time? My machine is from 2010. I rarely use the shut down button, turn off monitor and never put the machine to sleep...

System:
Motherboard: ASUS M4A79T Deluxe, Socket-AM3
CPU: AMD Phenom II X4 965 Black Edition Quad Core
RAM: Kingston DDR3 HyperX 1333MHz 4 x 2GB DDR3, CL7-7-7-20
Case: Fractal Design Define R2 Titanium Grey
PSU: Cooler Master Silent Pro M700 700W
GPU: XFX Radeon HD 5770 1GB GDDR5
HDD: WD VelociRaptor® 300GB SATA, 10000RPM, 16MB, 3 GB/s
Fans: 2 x Fractal Design Silent Pro 120mm (both blowing into case)
CPU cooler: Corsair H100i

I plan to swap all fans (including H100i fans) with Notcua version to improve SPL, the machine is running a bit loud. But I don't think that is necessarily relevant for my question... Is this high voltage to be expected? Is there something I might have overlooked?

I did the OC'ing in BIOS. I only changed the multiplier and CPU voltage, nothing else. Temperature readings are using Core Temp. I also tried the Corsair Link software but for some reason it is mixing the CPU temp and motherboard temp. BIOS CPU readings correspond with Core Temp though.

Thanks for helping :)

This looks pretty good here. Max P-state voltage for this chip is 1.4250v at only 3200mhz, you may have seen it that high while posted on auto settings in bios while under heavy load in windows.

I have lowered the multiplier to 18.5. Couldn't run Prime now so the temp is at limited load as you can see. Idle temp in BIOS is about 33 deg C.

Probably more v-core. Under 1.55v for daily clocks is ok. If you can cool 1.5v or more, try and stabilize the 19x multiplier, but slowly add voltage as needed.

Then increased in steps to 1.475V, still not stable.

This is only a slight over-volt. Try a little more.

The only curious thing from cpu-z I would need for voltage to clock speed guesstimate would be the stepping revision of the Cpu. C2 or C3. The C2 is a 125w Phenom core, generally doesn't clock as well and requires more voltage in most cases. With a C2, 1.6+ volts at 4250mhz wouldn't surprise me.
 
yeah, I wondered if the CPU was a C2 or C3 stepping myself. Dieter, are you only using the multiplier to overclock or are you also using the FSB? The combination of the two will generally improve your overclock.
 

Have a 940BE that was stable 4ghz, just barely lol.

Thermal Design Power ? 125 Watt (Socket AM2+/Socket AM3)
139.6 Watt (Socket AM2)

http://www.cpu-world.com/CPUs/K10/A... Edition - HDZ940XCJ4DGI (HDZ940XCGIBOX).html

Curious this 940BE is 140w on AM2, but on AM2+/AM3 boards it's only 125w?
 
I downloaded CPU-Z and attached a few pictures of my data. I was a bit surprised by the memory tab... Isn't my latency and timing a bit on the conservative side?

I will now go ahead and disable all those power saving features mentioned, increase the multiplier back up to 1.9 and increase voltage by one more increment. Will let Prime95 run through the night and see how it goes.

Thank you so far :)
 

Attachments

  • CPU.JPG
    CPU.JPG
    48.5 KB · Views: 3,579
  • Memory.JPG
    Memory.JPG
    41.2 KB · Views: 3,396
  • SPD.JPG
    SPD.JPG
    43.7 KB · Views: 3,430
Normally during setup and testing we disable C1/E, C6, Cool N Quiet, APC, TurboCore and in windows performance manager itself we set to "performance" mode.

OK so I had a look in the BIOS...
  1. C1/E already disabled.
  2. C6 I couldn't find
  3. Cool N Quiet was already disabled
  4. APC... Tried google but came up short... What is it? Cold only find ACPI SRAT (which was enabled).
  5. Other CPU related stuff that was enabled in the BIOS included SVM and Microcode Updation
  6. TurboCore was apparantly introduced for Phenom's but later not used for a few generations. Only Phenom II's from the Thuban and later have it, not my Deneb.
 
I post from more than one computer and and most of the TYPO = APC changed to APM which is used mostly with FX processors and may or may not be listed in your bios when you don't have FX processor in cpu socket. It is Application Power Management and again mostly current after FX processors as is C6 which allows TurboCore to function.
RGone...
 
I downloaded CPU-Z and attached a few pictures of my data. I was a bit surprised by the memory tab... Isn't my latency and timing a bit on the conservative side?

Yeah, you have to choose the "XMP" setting in bios to get those tight timings at 1333 mhz or set them manually if you prefer. "XMP" is the high performance profile for that RAM. Stands for "extreme memory profile" or "extreme memory performance" of some such thing. Maybe "extra memory performance". You get the idea. I'd also give the RAM a .05 increase in voltage at that setting if I were you.
 
Prime95 crashed after a few hours at 1.4875V and a multiplier of 19. I adjusted the multiplier to 18 but kept the CPU voltage, just to find a setting that would at least be successful... It has been stable now running Prime for95 for about 10 hours. Here are some screenshots (the temp CPU temp was stable at 45 but when I opened the door to the room the ambient temp went down and there was an immedeate reponse on the CPU temp so the screenshot shows 43 deg C, 45 deg max. HWmonitor screenshot is also while Prime95 is still running after 10 hours.

I will try to further increase voltage and see if I can get the multiplier stable at 19 as well.
 

Attachments

  • HWmonitor.JPG
    HWmonitor.JPG
    75.2 KB · Views: 3,342
  • CoreTemp, 3.6Ghz.JPG
    CoreTemp, 3.6Ghz.JPG
    41.6 KB · Views: 3,411
System stable for 5 hours on Prime95 at 1.5V and 3.8 Ghz. Tempearure is about 47-48 deg in Core Temp on balanced fan settings for cooler. I will run it a bit more later but I am happy with this CPU OC, at least for now :)

Next I figured I would have go at RAM. The voltage requirement for my "old-school" Kingston is 1.7V though (according to spec when bought, SPD tab says 1.65V at 7-7-7-20). My new fans are not coming in until the weekend so I am not sure how good the airflow is.

Question:
Is there a readout of RAM temperature in HWMonitor?
Do I still run Prime95 in order to stress test RAM OC?
Should I raise voltage to 1.65 or 1.7?
Are chances higher that you might need to do a hard reboot if the system crashes when changing RAM settings?
 
If it's stable on the Prime95 blend for at least two hours you should never have any issues with instability in normal computing tasks and games.
 
Question:
Is there a readout of RAM temperature in HWMonitor?
Do I still run Prime95 in order to stress test RAM OC?
Should I raise voltage to 1.65 or 1.7?
Are chances higher that you might need to do a hard reboot if the system crashes when changing RAM settings?

No readout for ram temps. Put a fan on the ram.

If you want to stress that Ram good, try OCCT linpack tab.

Chances are higher for clearing CMOS when overclocking/tweaking memory and system crashes. If you go a little at a time, you decrease these chances and OCCT would throw you an error and would be an indication of system instability.
 
I couldn't find a setting in BIOS where it was possible to change to an XMP setting. No help from Google this time... Is it possible that this is changed automatically? The RAM is now running with the 7-7-7-20 timing according to CPU-Z. Voltage increased from stock 1.5 to 1.66.
 
Never seen a mobo select XMP ram mode on its' own. That is an optional setting and unless it is AMP and not XMP, it is not a true AMD extreme ram mode although we all seem to use it to guide our most stringent ram settings.
RGone...
 
Back