• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.
  • The forums have been upgraded with support for dark mode. By default it will follow the setting on your system/browser. You may override it by scrolling to the end of the page and clicking the gears icon.

Your Haswell-E max stable OC

What CPU do you have?


  • Total voters
    59
I think that people with AIO System and Air cooler should keep on Max 1.300V without some voltage spikes under load.
What CPU can with that voltage that's it. 4.3 or 4.6 over that I think it's not safe and up to 1.350V for good watercooling where temps no matter on load stay under 60C.
If someone want to CPU work nice for 5 years.
Example I have i7-3770K capable to work on 4.5GHz on below or max 1.200V. Problem was because I had such motherboard where it was not possible to drop CPU Voltage or CPU Clock. Overclock on max speed or not overclocked with Turbo. After 3 years and little more 24/7 constant 1.200V and constant 4.5GHz in idle CPU didn't show any sign of degradation and need absolutely same voltage for overclocking and that's about 1.300V for Prime95 AVX Stable on 4.8GHz. But I kept for every day only 50mV over fabric voltage. Because even after 15h Prime95 I didn't recognize that I will get benefits from 4.8GHz instead 4.5GHz... Even if Prime95 AVX stable so long time is much better prove that CPU will stay stable in normal applications and gaming than any of now attractive tests.

For 1.4V I'm not sure how much Haswell E could behave on 2-3 years period. That's 250mV over fabric voltage.
On short time probably nothing, but on longer period I don't know... Skylake maybe is even safe on 1.4V.
 
I plan to upgrade to Broadwell E when it releases....so I'm not really all that worried about longevity. Just don't to hear it sizzling under my block. lol
 
I recently sent in a RMA 5820K and got one back from Intel from a new batch. The one I sent in would crash at 4.0GHz @ 1.2V and eventually would cause code AF on my Rampage V Extreme motherboard. So, this one that they sent me is currently (at the time of writing this) running at 4.5GHZ @ 1.24V. I'm currently going lower and lower with the voltage until the system crashes, but this is amazing. The other one crashed almost immediately at 4.5GHZ @ 1.3V...
 
I recently sent in a RMA 5820K and got one back from Intel from a new batch. The one I sent in would crash at 4.0GHz @ 1.2V and eventually would cause code AF on my Rampage V Extreme motherboard. So, this one that they sent me is currently (at the time of writing this) running at 4.5GHZ @ 1.24V. I'm currently going lower and lower with the voltage until the system crashes, but this is amazing. The other one crashed almost immediately at 4.5GHZ @ 1.3V...

Until it crashes...explain... Are you stress testing it? If it lasts on OCCT or AIDA64 for a few hours at 4.5 and 1.24v, you got yourself a keeper.
 
I'm trying to get to a point that it crashes quickly. It seems that 1.22V is the lowest I could get before it would become unstable. I bumped my CPU to 4.5 @ 1.23V and the cache to 4.3 @ 1.2V. I'm going to stress test tomorrow, but for some reason the cache will not go higher than 4.3 even if 4.5 is set as min/max. Weird...
 
If you're crashing quickly without stress testing at 1.22v, there's no way that 1.23v is going to be "stable". I'd bet to get OCCT or AIDA 64 stable on a minimum 6 hour run, you're gonna be looking at probably 1.275v or higher. But, g'luck with your testing.
 
If you're crashing quickly without stress testing at 1.22v, there's no way that 1.23v is going to be "stable". I'd bet to get OCCT or AIDA 64 stable on a minimum 6 hour run, you're gonna be looking at probably 1.275v or higher. But, g'luck with your testing.

It didn't crash immediately, actually it didn't crash at all. I just do a soft test by playing Youtube in the background while gaming and see what happens. The video at one point gave me the "something went wrong" message which usually means something happened with the CPU.
 
I'm trying to get to a point that it crashes quickly. It seems that 1.22V is the lowest I could get before it would become unstable. I bumped my CPU to 4.5 @ 1.23V and the cache to 4.3 @ 1.2V. I'm going to stress test tomorrow, but for some reason the cache will not go higher than 4.3 even if 4.5 is set as min/max. Weird...
Pretty normal for these CPUs. You're better off with 4.0 Cache, 4.6 GHz CPU, if you can muster it under 1.3V (sounds like you can). My retail will do 5.0 GHz @ 1.425V.
 
Yeah, its been running for 2 days straight now at 4.5Ghz @ 1.24v and 4.0Ghz cache @ 1.12v. I might try and see what happens with 5.0Ghz but isn't 1.4+V a little high?
 
Yeah....that's high. I've heard that anything over 1.3v will start to degrade the processor. How fast it degrades is anyone's guess, but......at 1.4v the temps have to be pretty freakin high running any kind of stress test too. Heat and volts = bad.

Have you stability tested that 4.5ghz @ 1.24v yet with AIDA 64 or OCCT? Those are pretty low volts for that clock....unless that processor is GOLDEN, I'm thinking if you get 20 mins out of it, you'll be lucky. /shrug
 
Hello! I am a noob here in the forums and owner of a 5930k.
I manage to get a stable, quick and dirty, OC (3 hours OCCT, 5 Cinebench runs with no crash) of 4.5ghz at 1.25v but I always get a "clock_watchdog_timeout" upon attempting to hit 4.6ghz. I tried up to 1.325v but same negative results. I guess 4.5ghz is my limit. If you can advise, my settings are below. Thank you,

Motherboard: X99-Sabertooth
Turbo Speed: 4500mhz
Multiplier: 36
CPU Strap: 125
BCLK: 125
Core Voltage: 1.25v
CPU System Agent Voltage: 1.25v
CPU Input Voltage: 1.90v
RAM: 2666mhz, 1.35v, 14-16-16-35 (successful downclock to 2333mhz, 1.20v, 12-14-14-31)
 
Got some time to test again. Now I am able to get it run on 4.33GHz(125 strap with 127.3BCLK set by XMP) with 1.25V vCore on offset mode. Pretty sweet,.
 
What do you guys think is the maximum voltage on a 5930k ?
 
What do you guys think is the maximum voltage on a 5930k ?
For 24/7 usage, I would not exceed 1.35.


Even if the temperature of the cpu is under 60°C and the mosfets under 30°C ?
 
I wonder how I missed this thread...@4.6 on mine 125 bootstrap, 127mhz bclk
 
i7 5820K @ 4.4 Ghz 1.3V. Cooled with NZXT Kraken X61. 24/7

Stable good overclock. Temps just under 70C° / 158F in Prime95. seems to be an average for 5820K/4.4Ghz/1.3V users all over. just need a good CPU cooler and that's it.

5820K.jpg
 
I have swapped 125 strap for 100 strap. Could not OC ram on 125 but on 100 I can get 3200mhz up from xmp 3000mhz. My core is now 4.4 up from 4.375, no extra V (also updated bios, this mite have helped vcore). I would like to get 102mhz BCLK but not having any luck. Any tips? What setting should I look at?
I have tried uping dram, vcore and syt agent volts. BCLK amplitude to higher, PCIE amplitude to higher.
 
wT7vAEe.png


Highest I could go, still crashes when used in BOINC and higher voltages makes my face melt cause of all the heat. Any suggestions?
 
Highest I could go, still crashes when used in BOINC and higher voltages makes my face melt cause of all the heat. Any suggestions?

A custom watercooling .
 
My cache seems to be very hungry for V. My CPU is 4.375ghz, 1.265v. Cache is 4,250ghz, 1.320v!!!!! I'm trying to lower it, first I tried cahce V @1.290v and it crashed an hour into BF4 so then I tried 1.300v, it crashed after 2 hours of BF4. Tonight I will try 1.310v.
Does this seem like abnormally high cache volts?
 
Last edited:
Back
Top