Overclock.net banner
1 - 20 of 21 Posts

·
Registered
Joined
·
72 Posts
Discussion Starter · #1 ·
I recently purchased an 11900k for use with a Z590-E ASUS motherboard. The chip was a low bin with SP 50 it needed 1.6+ for 5.3 according to the V/F Curve.

Either way, I found this to be fine. I lost the lottery but that's how it goes. However, the CPU seemed to have an issue that I had never seen before.

Occasionally, the temperature sensors - in multiple programs - would read "0c". I thought that this might be a BIOS issue or maybe a microcode issue - it is a new platform after all. I updated my MB to 0902 - latest available BIOS and the issue persisted.

I did some digging and I thought that if this WAS a widespread issue it would be something stated on numerous forums, multiple sites, etc. To my amazement I found only two posts about it on the HWINFO64 forums and no one here complaining about anything like this.

The first thread is an 11700k user: Rocket Lake; Core Temp Mins 0

The second thread is an 11900k user: Motherboard Nuvoton header Temp2 reading erratic v7.02-4430

One is using a Z490 Gigabyte board and the other has a Z590 maximus.

When I contacted intel they had an elevated technician remotely access my PC. He did several tests, IPDT, looked at sensors, XTU and several other tests.

To my amazement, after the technician searched their systems and contacted upper level management, there were zero reports of this issue. He stated the CPU was defective and even provided me with an email to give to my retailer to ensure a smooth exchange.

So, my hypothesis is that this is an issue impacting SOME 11th gen CPUs given the fact that it isn't widely reported and only a few users are experiencing it. I wonder then if it's something wrong at the silicon level during fab. Either way, I am exchanging it - hopefully I don't get SP 50 or bad temp sensors this time.
 

·
Registered
Joined
·
2,062 Posts
Perhaps the technician concluded that simply to get you a replacement CPU
 

·
Registered
Joined
·
3,822 Posts
I've had defective temp sensors on CPU's before. Like my 2500K. It read 0,0,128,128 all the time. Send it for RMA to Intel directly, got a new one within 2 weeks. Which, funnily enough, was a extremely good chip that did well over 5.9Ghz on LN2, CPU-Z validation at 6.04Ghz and did 5.3Ghz on AIO.

 

·
Registered
Joined
·
72 Posts
Discussion Starter · #5 ·
I've had defective temp sensors on CPU's before. Like my 2500K. It read 0,0,128,128 all the time. Send it for RMA to Intel directly, got a new one within 2 weeks. Which, funnily enough, was a extremely good chip that did well over 5.9Ghz on LN2, CPU-Z validation at 6.04Ghz and did 5.3Ghz on AIO.

Wow that was one hell of an overclocker!
 

·
Registered
Joined
·
2,888 Posts
I've had defective temp sensors on CPU's before. Like my 2500K. It read 0,0,128,128 all the time. Send it for RMA to Intel directly, got a new one within 2 weeks. Which, funnily enough, was a extremely good chip that did well over 5.9Ghz on LN2, CPU-Z validation at 6.04Ghz and did 5.3Ghz on AIO.

I just got an 11900K…. checks for bad sensors
 

·
Registered
Joined
·
23 Posts
I think that's normal for 11 gen, mine 11700k with sp of 78 also shows 0c on certain cores for a split of a second when monitoring temp. Don't affect anything.
 

·
Premium Member
Joined
·
20,477 Posts
@Mumak, might be able to help. Probably HWINFO64 needs some updating.
 

·
Not New to Overclock.net
Joined
·
5,017 Posts
I think that's normal for 11 gen, mine 11700k with sp of 78 also shows 0c on certain cores for a split of a second when monitoring temp. Don't affect anything.
no its not.

it was the later stages if hwinfo that started this
or intel ucode

never had this issue from dec 2020 to april 2021..

so one of it is causing it.
 

·
Registered
Joined
·
952 Posts
I mean, have you heard of this sensor issue before?
Are you collecting evidence so to build a new conspiracy theory?
Just get your replacement and forget about it.

At my CPU, one core reads significantly high in temperature than the other three.
And so what?
 

·
Registered
Joined
·
72 Posts
Discussion Starter · #13 ·
Are you collecting evidence so to build a new conspiracy theory?
Just get your replacement and forget about it.

At my CPU, one core reads significantly high in temperature than the other three.
And so what?
I'm not trying to build a conspiracy theory.
 

·
Not New to Overclock.net
Joined
·
5,017 Posts
This is a known bug observed with several RKL CPUs. I have already submitted this to Intel, but don't know if there's any progress nor haven't seen this listed as an official erratum yet.
its ucode then. cause early ucodes afaik no issue. all this happen later on
 

·
Premium Member
Joined
·
3,729 Posts
Are you collecting evidence so to build a new conspiracy theory?
Just get your replacement and forget about it.

At my CPU, one core reads significantly high in temperature than the other three.
And so what?
Technically, if you're using CPU temperature to actively adjust CPU cooling, to minimise noise for example, then it reading cold when it's actually hot could possibly present a problem?
 

·
Registered
Joined
·
952 Posts
Technically, if you're using CPU temperature to actively adjust CPU cooling, to minimise noise for example, then it reading cold when it's actually hot could possibly present a problem?
At four core CPU, there is temperature information this coming due five sensors.
All motherboards will use the fifth sensor in this example, which is the measurement of temperature at CPU metal surface.
DC fan noise this is audible at stock CPU coolers with 90mm blades, I am using double 140mm ones, them not able to produce any noise even at full 1200 rpm.
My CPU cooler surface this is huge, DC fan nominal air flow this is sufficient even at gaming for several hours.

One of the reasons (of the many) of me not upgrading to fresh CPU, this is because the industry of third party CPU coolers, this lost their talent at the delivery of quality CPU cooler solution.
Materials in use for passive heat transfer they can handle thermal sources up to 240W, and CPU and GPU industry keeps making ovens at 300W or higher.
And all such modern mistakes they end up be talked over and over and over, and they are not out there easy solutions.
 

·
Premium Member
Joined
·
3,729 Posts
At four core CPU, there is temperature information this coming due five sensors.
All motherboards will use the fifth sensor in this example, which is the measurement of temperature at CPU metal surface.
DC fan noise this is audible at stock CPU coolers with 90mm blades, I am using double 140mm ones, them not able to produce any noise even at full 1200 rpm.
My CPU cooler surface this is huge, DC fan nominal air flow this is sufficient even at gaming for several hours.

One of the reasons (of the many) of me not upgrading to fresh CPU, this is because the industry of third party CPU coolers, this lost their talent at the delivery of quality CPU cooler solution.
Materials in use for passive heat transfer they can handle thermal sources up to 240W, and CPU and GPU industry keeps making ovens at 300W or higher.
And all such modern mistakes they end up be talked over and over and over, and they are not out there easy solutions.
Well, sound is subjective - If you use a sound pressure meter you'll find that your double 140mm do make a noise at 1200rpm, you might not hear it, others might, depending on the environment/background noise and the person.

I think things have moved on a bit from "all motherboards will use the fifth sensor". I have a £100 entry level mobo, six core CPU and a £10 software app. I'm able to link any sensor (or combination of sensors) to any fan (or group of fans) via any curve or line I choose. 10 years ago I needed £100 of specialist hardware on a high end mobo to do the same thing, today though £10 software on a entry level board.

I use the core temperatures to drive the CPU cooling, it looks at all six cores and reports the hottest one. The CPU will throttle to protect its self when the core temperature reaches it's limit, not the socket temperature - it doesn't care about the socket temperature at all, and these days they throttle individual cores to protect them, not the whole CPU. In the situation where I'm running a single thread application on a six core processor one core is hot, five are cold and the socket temp stays quite low because only one of six cores is stressed - for that I want the cooling to increase, and reduce that single hot core as much as possible.

In that situation if that single hot core says it's zero when it's hot the cooling reduces and it overheats? The sensors have to function correctly regardless - no excuses :)
 

·
Registered
Joined
·
952 Posts
Well, sound is subjective
You may arrange your next vacations to central Greece.
I will also give you a tour of the city of Volos and of my PC.
 

·
Premium Member
Joined
·
3,729 Posts
You may arrange your next vacations to central Greece.
I will also give you a tour of the city of Volos and of my PC.
I'd probably be more interested in your workshop, I can't justify owning expensive electrical test equipment that I'll never really need but would still like to play with it, all I have is a multi-meter :(
 
1 - 20 of 21 Posts
Top