-
Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
Realtemp ideal temp?
- Thread starter JDMStanced
- Start date
- Joined
- Dec 27, 2008
yes, Uncore is another term for cache. No sense in having it increase along with core ratio and become potential source of instability when it won't have any bearing on peformance. LLC/cache voltage can also be lowered and set to manual if the cache ratio is being limited to 40x. The bios may increase the LLC/cache automatically along with the core increase unless you set the cache/LLC voltage to a manual set value. Check out my pic of HWMonitor for the values I assigned to these things to maybe get a rough idea of what you might get by with if you set the LLC/Cache voltage manually. Keeping those secondary voltages as low as possible can help contain temps.
OP
- Joined
- Nov 22, 2014
- Thread Starter
- #63
ive been changing around the value for hours now.
4.4 OC is what i'm trying to achieve. It passes the intel Burner test. Max temp. is at 81. ( i read that finely OCed 4.4 is better than turbo boost to 4.4)
CPU Vcore is 1.140V.
CPU VRIN Loadline Calibration is Medium from AUTO.
Ring Voltage: Normal
Ring voltage offset: -.025v
The thing is... when i run OCCT test, it fails after about 20 mins. PC restarts saying it will try to solve the problem and restart.
Is it just that i have to raise the VCORE temp. little by little until PC doesn't restart? do i have to disable turbo?
4.4 OC is what i'm trying to achieve. It passes the intel Burner test. Max temp. is at 81. ( i read that finely OCed 4.4 is better than turbo boost to 4.4)
CPU Vcore is 1.140V.
CPU VRIN Loadline Calibration is Medium from AUTO.
Ring Voltage: Normal
Ring voltage offset: -.025v
The thing is... when i run OCCT test, it fails after about 20 mins. PC restarts saying it will try to solve the problem and restart.
Is it just that i have to raise the VCORE temp. little by little until PC doesn't restart? do i have to disable turbo?
- Joined
- Dec 27, 2008
OCCT is very similar to Prime95 and IBT in how it stresses the CPU and none of them may do as good a job at giving a well-rounded stress test as does AIDA Extreme or XTU while driving up temps unnecessarily high in the process. Real world computing isn't like that. That's why a lot of folks are moving over to AIDA and XTU for stress testing.
JDM, let me suggest using the XTU stress test instead. Run it for 4-8 hr. It will utilize 97-98% of the CPU but leave you enough processor slices to be able to use the computer for other things while you are stress testing. I think you will find that if you can pass the XTU stress test for several hours then your computer will be stable for everyday use.
VRIN and Ring refer to the CPU cache. In this case, Load Line Calibration refers to a supplementation of the cache voltage, especially under load. You might try disabling it and just going with a fixed value for the cache since you have also fixed the cache ratio.
JDM, let me suggest using the XTU stress test instead. Run it for 4-8 hr. It will utilize 97-98% of the CPU but leave you enough processor slices to be able to use the computer for other things while you are stress testing. I think you will find that if you can pass the XTU stress test for several hours then your computer will be stable for everyday use.
VRIN and Ring refer to the CPU cache. In this case, Load Line Calibration refers to a supplementation of the cache voltage, especially under load. You might try disabling it and just going with a fixed value for the cache since you have also fixed the cache ratio.
Confused again... why are we touching the cache ratio in the first place? With manual overclocking, it doens't move, even when the CPU ratio goes up, right (that has been my experience as I NEVER touch cache ratio except when benchmarking)? I always left that on auto way over 4.5Ghz.
I'm just wondering if we are making things more confusing...(or if I am forgetting how things go with this platform already, LOL).
I'm just wondering if we are making things more confusing...(or if I am forgetting how things go with this platform already, LOL).
OP
- Joined
- Nov 22, 2014
- Thread Starter
- #68
DISABLE is not listed under Loadline Calibration. It's like normal, medium and up to extreme.
I, again, read from somewhere that setting it to extreme is same as disabling it. Would this cause the CPU or whatever to shorten it's life?
should i disable Turbo since i'm at 4.4?
Although BIOS shows that i have OCed to 4.4, CPU-Z shows under specification, 4790K [email protected] GHZ. Is this suppose to change to 4.4 GHZ?
Core Voltage on CPU-Z is .632V idle.
I, again, read from somewhere that setting it to extreme is same as disabling it. Would this cause the CPU or whatever to shorten it's life?
should i disable Turbo since i'm at 4.4?
Although BIOS shows that i have OCed to 4.4, CPU-Z shows under specification, 4790K [email protected] GHZ. Is this suppose to change to 4.4 GHZ?
Core Voltage on CPU-Z is .632V idle.
Last edited:
ok...
So, does cache ratio move if you are only touching cpu multi and voltage?
(Maybe we should pm, lol)
I just see these mentions but the reality is it's cpu multi, voltage and done. I've never touched cache anything to 4.8ghz...just trying to KISS for the op...
- Joined
- Dec 27, 2008
- Joined
- Dec 27, 2008
- Joined
- Dec 27, 2008
Look at the voltage section of HWMonitor listed (Processor voltage section, not the motherboard stuff at the top). See what effect changing the LLC/Ring voltage setting in bios from normal to medium to extreme has in what is reported by HWMonitor. It's not the exact voltages that count as the utility may not be reporting values with precision but it should show relative changes.
OP
- Joined
- Nov 22, 2014
- Thread Starter
- #74
1) Turbo is enabled again. (All turbo cores are set to 44 manually.)
I changed LLC to High from Standard.
2) On CPU-Z, i'm noticing Core voltage is bouncing around from .617v to 1.--.
Also under Clocks (core#0), Core speed bounces alot and quick from 799.81 mhz to 4399 mhz. I right clicked Cored Speed and not all cores are bouncing equally.
Before i messed around with values, Core volt and Core speed were pretty solid, IIRC. They didn't bounce around much.
This kinda tells me that it's not stable even though it passes IBT test.
3) After booting and desktop background shows, it takes me a long time for the system to open a file/folder.
Previously, i was able to open a file pretty quickly right when background shows.
Idle temp is low 40 or high 30 most of the time. But sometimes i see one or two cores spiking to 45.
It passed IBT test little quicker than factory setting. I'll do Intel Extreme test while i'm in bed.
did i do something wrong? lol..
so tired now
**Did benchmarking on Intel Extreme Tuning. (whatever it is lol)
Score is 1062 marks. Highest CPU temp : 71 degrees. CPU 4.4 ghz used.
Here are my current settings:
I changed LLC to High from Standard.
2) On CPU-Z, i'm noticing Core voltage is bouncing around from .617v to 1.--.
Also under Clocks (core#0), Core speed bounces alot and quick from 799.81 mhz to 4399 mhz. I right clicked Cored Speed and not all cores are bouncing equally.
Before i messed around with values, Core volt and Core speed were pretty solid, IIRC. They didn't bounce around much.
This kinda tells me that it's not stable even though it passes IBT test.
3) After booting and desktop background shows, it takes me a long time for the system to open a file/folder.
Previously, i was able to open a file pretty quickly right when background shows.
Idle temp is low 40 or high 30 most of the time. But sometimes i see one or two cores spiking to 45.
It passed IBT test little quicker than factory setting. I'll do Intel Extreme test while i'm in bed.
did i do something wrong? lol..
so tired now**Did benchmarking on Intel Extreme Tuning. (whatever it is lol)
Score is 1062 marks. Highest CPU temp : 71 degrees. CPU 4.4 ghz used.
Here are my current settings:
Attachments
Last edited:
- Joined
- Dec 27, 2008
It's normal for frequencies and voltages to jump around like that under load. The cores are taking turns under the load. Not all are working at full capacity at any given time. This is normal. You can also see this in HWMonitor as you observe how the temps dynamically change among the different cores.
OP
- Joined
- Nov 22, 2014
- Thread Starter
- #77
i'm getting this notice.."Display driver has stopped working" Then it says it successfully recovered. Screen freezes for like a sec.
This is keep happening.
Could this be because i lowered the CPU VCORE too low? i don't think it's too low though.. it passed 6 hour Intel Extreme test and PC doesn't restart on its own.
I also noticed that Core #0 and 3 are about 4 degrees higher than other cores. And sometimes they spike to Mid 50s degrees from 39-40 at idle.
This is keep happening.
Could this be because i lowered the CPU VCORE too low? i don't think it's too low though.. it passed 6 hour Intel Extreme test and PC doesn't restart on its own.
I also noticed that Core #0 and 3 are about 4 degrees higher than other cores. And sometimes they spike to Mid 50s degrees from 39-40 at idle.
- Joined
- Dec 27, 2008
If your core temp variance is only 3-4c that's actually pretty good. I'm getting 7c differential between hottest and coolest cores under load. The variance can be/is due to a number of factors such as variance in the calibration of the core temp sensors, variance in how close each one is to the cores they are measuring, variance in how the underside of the CPU lid is making contact with the die, variance in the contact between the heat sink bottom and the top of the CPU, etc. You're fine there.
I would think the "display stopped working" message would not be due to the vcore being too low but if you want to test it, then raise it a little. More likely a video driver issue. What are you doing when you get this message?
I would think the "display stopped working" message would not be due to the vcore being too low but if you want to test it, then raise it a little. More likely a video driver issue. What are you doing when you get this message?
OP
- Joined
- Nov 22, 2014
- Thread Starter
- #79
I never got this message before i did overclocking or changing values. Everything is only like 6 months old
In the same message box, it says display stopped working and says it resolved the issue. One sec of freezing screen bothers me ..
no idea what happened and what it did to resolve it...lol..
I wonder what i can change around to see if the message doesnt appear again.
In the same message box, it says display stopped working and says it resolved the issue. One sec of freezing screen bothers me ..
no idea what happened and what it did to resolve it...lol..
I wonder what i can change around to see if the message doesnt appear again.
