Question DEGRADING Raptor lake CPUs

[Kocicak]

I noticed some reports about degrading i9 13900K and KF processors.

I experienced this problem myself, when I ran it at 6 GHz, light load (3 threads of Cinebench), at acceptable temperature and non extreme voltage. After only few minutes it crashed, and then it could not run even at stock setting without bumping the voltage a bit.

I was thinking about the cause for this and I believe the problem is, that people do not appreciate, how high these frequencies are and that the real comfortable frequency limit of these CPUs is probably at something like 5500 or 5600 MHz. These CPUs are made on a same process (possibly improved somehow) on which Alder lake CPUs were made. See the frequencies 12900KS runs at. The frequency improvement of the new process tweak may not be so high as some people presume.

Those 13900K CPUs are probably highly binned to be able to find those which contain some cores which can reliably run at 5800 MHz. Some of the 13900K probably have little/no OC reserve left and pushing them will cause them to degrade/break.

The conclusion for me is that the best you can do to your 13900K or 13900KF is to disable the 5800 MHz peak, which will allow you to offset the voltage lower, and then set all core maximal frequency to some comfortable level, I guess the maximum level could be 5600 MHz. With lowered voltage this frequency should be gentler to the processor than running it at original 5500 MHz at higher voltage. You can also run it at lower frequencies, allowing for even higher voltage drop, but then the CPU is slowly loosing its sense (unless you want some high efficiency CPU intended for heavy multithread loads).

Running it with some power consumption limit dependent on your cooling solution to keep the CPU at sensible temperature will help too for sure.

 

[Carfax83]

I don't think high frequencies alone can degrade a CPU. What degrades a CPU are high amounts of current and voltage and it occurs over a long period of time. I've had CPUs degrade before. My old Sandy Bridge E 3950K degraded towards the end because of the high voltages I had going through it. It took a long time though.

What do you define as "acceptable temperature and non extreme voltage?"

If your 13900K has degraded that fast, something must be wrong and I would definitely seek a replacement from Intel because that's not normal.

The conclusion for me is that the best you can do to your 13900K or 13900KF is to disable the 5800 MHz peak, which will allow you to offset the voltage lower, and then set all core frequency to some comfortable level, I guess tha maximal level could be 5600 MHz. With lowered voltage this frequency should be gentler to the processor than running it at original 5500 MHz at higher voltage. You can also run it at lower frequencies, allowing for even higher voltage drop, but then the CPU is slowly loosing its sense.

I've had mine under clocked and undervolted since the day I got it. I now have it at 5.2ghz on the P cores, 4.3ghz on the E cores and using a -0.150mv offset at 215w and it runs really cool at these settings, maxing out at 77c. I didn't see any use having a couple of cores hitting 5.8ghz, especially if you're running a 4K monitor. To me, once you're over 5ghz,it's already blazing fast.

 

[Kocicak]

The problem may be that Intel themselves are pushing these top end CPUs too much, not only in total power consumption, but in frequency too. If these CPUs have by chance little OC headroom left, they should not be sold as K models, because people will try to overclock them and break them in the process.

I already returned the "flaky" CPU to the seller. I believe the CPU may have been some weaker piece.

I tried the current one at 5600MHz with -50mV offset, it worked well, now I am trying 5400 MHz with -120mV, I am not sure, whether I am not overdoing the offset too much, I will see. I have no motivation to go for even lower frequency.

 

[Carfax83]

The problem may be that Intel themselves are pushing these top end CPUs too much, not only in total power consumption, but in frequency too. If these CPUs have by chance little OC headroom left, they should not be sold as K models, because people will try to overclock them and break them in the process.

I'm not saying it's impossible for the CPU to degrade, but to do it that quickly is a bit hard to believe unless you're doing some extreme overclocking, ie trying to hit 8ghz or something and pumping tons of volts and current into the chip. For my old CPUs, it took years for them to degrade.

I tried the current one at 5600MHz with -50mV offset, it worked well, now I am trying 5400 MHz with -120mV, I am not sure, whether I am not overdoing the offset too much, I will see. I have no motivation to go for even lower frequency.

These CPUs are massively overvolted if you stick to stock settings that's for sure. I've had mine as low as -165mv and it only became unstable during idling. At loads it was perfectly stable. I have it now at -150mv and it seems very stable.

 

[scineram]

I don't think high frequencies alone can degrade a CPU. What degrades a CPU are high amounts of current and voltage and it occurs over a long period of time. I've had CPUs degrade before. My old Sandy Bridge E 3950K degraded towards the end because of the high voltages I had going through it. It took a long time though.

What do you define as "acceptable temperature and non extreme voltage?"

If your 13900K has degraded that fast, something must be wrong and I would definitely seek a replacement from Intel because that's not normal.



I've had mine under clocked and undervolted since the day I got it. I now have it at 5.2ghz on the P cores, 4.3ghz on the E cores and using a -0.150mv offset at 215w and it runs really cool at these settings, maxing out at 77c. I didn't see any use having a couple of cores hitting 5.8ghz, especially if you're running a 4K monitor. To me, once you're over 5ghz,it's already blazing fast.

Nah, overclockers don't deserve free processors.

 

[VirtualLarry]

Nah, overclockers don't deserve free processors.

But they do deserve woking ones, within warranty restrictions of course.

 

[Thunder 57]

But they do deserve woking ones, within warranty restrictions of course.

Overclocking voids the warranty.

 

[Captante]

There's a bit of Pentium 4 "Prescott" in the RL I9's for sure!

6 ghz sustained seems like pushing your luck a bit without some kind of exotic cooling solution no?

 

[DrMrLordX]

Voltage kills, friends.

 

[Kocicak]

1.45V is not an extreme voltage. It may be a little bit higher, but the CPU should be able to detect that something is "melting" or somehow wrong and throttle down to avoid damage. Such voltage cannot cause instant death, the CPU should have plenty time to detect the problem and react.

If it does not have effective self protection mechanisms it should be sold as a locked processor.

 

[Thunder 57]

1.45V is not an extreme voltage. It may be a little bit higher, but the CPU should be able to detect that something is "melting" or somehow wrong and throttle down to avoid damage. Such voltage cannot cause instant death, the CPU should have plenty time to detect the problem and react.

If it does not have effective self protection mechanisms it should be sold as a locked processor.

Well you did get that CPU before the official launch date it smelt it burning, did you not? I can quote you if need be. Also, it has happened before. Ever hear of "Sudden Northwood Death Syndrome"? Overvolting Northwoods caused them to die super fast. At least at first.

 

[DrMrLordX]

1.45V is not an extreme voltage. It may be a little bit higher, but the CPU should be able to detect that something is "melting" or somehow wrong and throttle down to avoid damage. Such voltage cannot cause instant death, the CPU should have plenty time to detect the problem and react.

If it does not have effective self protection mechanisms it should be sold as a locked processor.

The self-protections it has are the preprogrammed voltage/clockspeed tables built into the CPU. If you start manually adjusting the voltage, you're effectively disabling everything the CPU can do to protect itself. It's much harder to dial in a "safe" voltage now since CPUs have dynamic boost tables, making it difficult to know what is the "stock" vcore so you can use the old 10% rule. Also some voltages the CPU might apply to itself will only apply in low-current/low-temp scenarios, for limited periods of time.

 

[coercitiv]

If it does not have effective self protection mechanisms it should be sold as a locked processor.

Bold words coming from someone who ordered multiple Raptor Lake SKUs, overclocked all of them and then decided to keep only one. It almost looks like other honest consumers are paying for your overclocking pleasure.

 

[Kocicak]

I believe that the CPU has A LOT of temperature sensors in it and it changes frequency if needed (something reaches max allowed temperature). I did not lock the frequency, I just limited maximal frequency it could run at. It chose to run at that frequency by itself!

I believe that some of the cores was not up to 500 MHz overclock and despite it could run OK at first, some weaker bit in it just got a little more wear on it.

The question is, if these "weaker bits" are a general characteristic of the particular manufacturing process these CPUs are made with, or if they are just normal random defects affecting just small percentage of the CPUs.

BTW if that "weak bit" was the reason the core it was on was limited to 5500 MHz, the problem may have manifested anyway even at 1.3V and 5500 MHz in the timeframe of dozens or hundreds of hours, instead of few minutes of overall very light load (only 3 cores with heavy load) at 85°C, 1.45V and 6000 MHz.

 

[coercitiv]

It chose to run at that frequency by itself!

This escalated quickly.

 

[therealmongo]

The only "weak bit" is you and your shenanigans, be it either Intel or AMD.

Just another "man child" shrieking responsibility of his actions, a scurge on all honest consumers out there ..

 

[Rigg]

Voltage kills, friends.

Voltage is just a patsy in this crime. The current is the real murderer.

 

[lopri]

It feels like a industrial scandal is just around the corner. Both Intel and AMD are pushing too hard to uncomfortable degree. It can take just one motherboard vendor or system integrator to play fast and loose to have a mass system failure from these factory-overclocked CPUs.

 

[Kocicak]

I do not believe that the "K" CPUs will be used for any professional type of use.

13600K and 13700K may still have some safe OC headroom left, but 13900K may be sold close to what the silicone is capable of.

The stress I subjected the CPU to was not extreme at all, and if some of the CPUs are really weak, we will see some increased failure rate in upcoming months. It only matters, what is the percentage of these weak CPUs, and if the problem are just random infrequent defects or if the problem is deeper and the manufacturing process itself is not suited for these extreme frequencies.

 

[DrMrLordX]

Voltage is just a patsy in this crime. The current is the real murderer.

End users have fine grain control over voltage (if they want it), while current draw varies with workload. You can safely operate the CPU in high-current workloads if you let the voltage and clockspeed drop low enough.

 

[eek2121]

Reset CMOS settings. Do a hard reset (use jumper or remove battery after unplugging the system and pushing the power button to discharge)

CPU degradation does not happen in hours. It happens in months/years.There is likely a faulty stuck CMOS setting stuck related to your overclock.

 

[Kocicak]

I monitored voltage and temperature in HWinfo. The values were acceptable, load was light, I left the room confident that it is just a light 10 minutes test which will probably pass without any problem, and minutes later I found the computer restarted and the CPU unable to run any benchmark with default settings without positive voltage offset (20mV was enough). The CPU must have some voltage reserve for default use, so it degraded by that reserve + 20mV.

The weakness in the CPU may be of the kind it will allow short term 100% functionality, but it will manifest itself fully only after slight stress. I do not believe that CPU could have been long term reliable, because the process at lower voltage would be happening too, just slower. I already wrote that what failed at 1.45V and 6000MHz in minutes would have failed anyway at 1.3V and 5500 MHz in tens or hundreads of hours (or even earlier).

One member of overclock.net describes quick degradation too.

 

[NTMBK]

I do not believe that the "K" CPUs will be used for any professional type of use.

Yeah no kidding. Squeezing an extra 3% of performance out of a CPU isn't worth risking your career with system instability and silent data corruption.

 

[VirtualLarry]

I do not believe that the "K" CPUs will be used for any professional type of use.

Is Intel still disabling VT-d on "K" SKUs?

 

[Rigg]

End users have fine grain control over voltage (if they want it), while current draw varies with workload. You can safely operate the CPU in high-current workloads if you let the voltage and clockspeed drop low enough.

Yes, and the opposite is also true. You can safely run the CPU at higher voltage/clock speeds in lower current loads. In modern CPU's the voltage and clock speed varies right along with the current demand of the workload. The bottom line is too much voltage doesn't kill/degrade CPU's. Too much current does.