Intel Isolates "Raptor Lake" Vmin Shift Instability Root Cause, New Microcode Update Coming

[AleksandarK]

Back in August, Intel started shipping its 0x129 microcode update for 13/14th generation "Raptor Lake" and "Raptor Lake Refresh" processors. This update fixed incorrect voltage requests to the processor that are causing elevated operating voltage. Intel's analysis showed that the root cause of stability problems is voltage levels that are too high during the operation of the processors. These increases in voltage cause degradation that increases the minimum voltage required for stable operation. Intel calls this "Vmin." Today, the company discovered the root cause of this instability issue and informed users that a new microcode patch is underway. As explained by Intel, the Vmin Shift instability problem stems from a clock tree circuit in the IA core. When exposed to high voltage and temperature conditions, this circuit is vulnerable to reliability degradation. Intel's research has shown that these factors can cause a shift in the duty cycle of the clocks, resulting in system instability.

There are four scenarios that can cause Vmin Shift: increased motherboard power delivery, eTVB microcode algorithm running at higher performance operating states even at higher temperatures, microcode SVID algorithm requesting higher voltages at higher frequencies and longer durations, and finally microcode and BIOS requesting elevated core voltages. For motherboard power settings, mitigation is switching back to default settings. For the eTVB issue, the fix is a 0x125 microcode update. The 0x129 patch fixes the SVID algorithm, and the fourth condition, where microcode and BIOS request elevated core voltage, is fixed by the upcoming 0x12B microcode update. Intel is reportedly working with OEMs to start rolling out the 0x12B update with no apparent performance degradation. While the timeframe for shipping this update is unknown, we expect to see it soon. Additionally, Intel once again confirmed that the upcoming "Arrow Lake" CPUs don't have these issues.



View at TechPowerUp Main Site | Source

 

[Hyderz]

i just wonder how many are going to ditch 13/14th gen when arrow lake arrives? i think the damage is already done for that platform

 

[aktpu]

I'm not going to ditch 13th/14th gen, but I'm hoping for rock bottom pricing for 14700k, so I can upgrade to that from 13600k dirt cheap

 

[clopezi]

i just wonder how many are going to ditch 13/14th gen when arrow lake arrives? i think the damage is already done for that platform

Intel it's doing RMA's to damaged processors. I had a 13900K and Intel RMA me a 14900K without asking anything. Nobody, or almost nobody, it's going to throw a 600€ CPU because it's there a new one with almost the same capabilities (and also a new motherboard and new ram in many cases).

However, the damage it's done. Intel had a very good reputation about hardware degradation over the years, and maybe some users will think in going with AMD after this, if AMD it's still competitive in some years.

 

[arbiter]

Hope it also includes lowering the voltage cpu uses to run. i am running -0.150 atm to counter the 129 update that made my cpu run 1.45+ volts when doesn't need it.

 

[BoggledBeagle]

It is funny how they called degradation "reliability aging"

The real reason for quick degradation - too high frequency, which is the underlying cause for the elevated temperature and voltage causing high electric current density, is missing from their list of causes.

I am not convinced that even a brand new CPU running the 12B microcode will reliably work for long years at those extreme frequencies.

 

[PassportBro]

Intel it's doing RMA's to damaged processors. I had a 13900K and Intel RMA me a 14900K without asking anything. Nobody, or almost nobody, it's going to throw a 600€ CPU because it's there a new one with almost the same capabilities (and also a new motherboard and new ram in many cases).

However, the damage it's done. Intel had a very good reputation about hardware degradation over the years, and maybe some users will think in going with AMD after this, if AMD it's still competitive in some years.

What? OF COURSE they will throw (away) a $600 CPU to buy the next gen Intel stuff. They will also gladly throw away their motherboard, ram, storage/ssd's and buy new ones and fairly soon they will also dump their 4090 on ebay and a buy a new 5090 for $3k+ and a new PSU to boot to support that 5090.

You have no idea how many such users exist, they are known as "whales".

During the ETH mining craze, a US crypto-miner mounted an array of 3090's and 3080's at the trunk his $200K car and then posted the picture on twitter just to spite ppl like you. That was back when there were 200 ppl queues at Cali Microcenter stores and ppl were camping out in Microcenter parking lots two days in advance to have a shot at a 3080.

 

[LittleBro]

So there were 4 separate causes for single issue on 13th and 14th present at the same time?

 

[BoggledBeagle]

So there were 4 separate causes for single issue on 13th and 14th present at the same time?

No, there is one cause - too high frequency, and some factors that worsen degradation caused by it, Intel chose to cite those factors and pretend that the frequency is OK.

 

[clopezi]

No, there is one cause - too high frequency, and some factors that worsen degradation caused by it, Intel chose to cite those factors and pretend that the frequency is OK.

It's better to sync all P-cores to 5.7 and call it a day to sleep well about this

 

[BoggledBeagle]

5.7 is way too much. Also why to sync all cores to run at the same frequency? You can individually limit each core to whatever frequency you want.

 

[N/A]

No, there is one cause - too high frequency, and some factors that worsen degradation caused by it, Intel chose to cite those factors and pretend that the frequency is OK.

they're connencted high frequency requires high voltage. So frequency alone does not cause aggravation. Voltage is. anything above 1.25 frankly. just run prime for a few minutes and vmin now shifts by a tiny bit. run it for longer and higher and the cpu will crash at previously stable clocks. Intel higher ups needed desperately to be competive and forgot that the physics wouldn't allow it.

 

[LittleBro]

No, there is one cause - too high frequency, and some factors that worsen degradation caused by it, Intel chose to cite those factors and pretend that the frequency is OK.

Frequency itself does not damage a chip. It raises a temps a bit but if there's enough cooling provided, the chip will not get damaged.

I can ramp up frequency of my CPU to 7GHZ with same voltage I run it at 4.7, maybe it will even boot, but then it will crash due to instability. That does not mean that the chip got damaged.

 

[BoggledBeagle]

they're connencted high frequency requires high voltage. So frequency alone does not cause aggravation. Voltage is.

Degradation is caused by high temperature and high electric current density, which is caused by high voltage, which is required to run the CPUs at too high frequencies, which are specified by Intel execs.

The root cause for the mess are Intel execs who chose to ignore all the good industry practices and precautionary principles, which are in place to deliver customers a long term reliable product.

 

[ZoneDymo]

I feel this is the 4th "root cause" they found for the same issue, and the 4th microcode update...

 

[_roman_]

Will there be any decent INTEL news again about quality or a decent new product which works flawless without hassle?

Quality delivered as my (whataboutism) INTEL WLAN card in gnu linux (which crashes since i bought it // determine the crashes - rule it out -> usb tethering instead as of the intel wlan chip -> no crash since than).

 

[Beginner Macro Device]

I'm not going to ditch 13th/14th gen, but I'm hoping for rock bottom pricing for 14700k, so I can upgrade to that from 13600k dirt cheap

Kinda same. Thinking of "retiring" my 12400F to serve in my bro's PC (he now has an i7-6700-alike Xeon CPU so uplift will be more than substantial) and getting myself a proper 5+ GHz CPU. Will probably use at 5 GHz sharp since it's more a "why not" than an "I desperately need this" kinda purchase.

 

[yzonker]

Degradation is caused by high temperature and high electric current density, which is caused by high voltage, which is required to run the CPUs at too high frequencies, which are specified by Intel execs.

The root cause for the mess are Intel execs who chose to ignore all the good industry practices and precautionary principles, which are in place to deliver customers a long term reliable product.

Yup. Maybe they've locked things down enough in the default profile to keep most of them working through the warranty (and maybe not of course, but Intel probably thinks so). My 14900k will downclock some even during gameplay, so they may be reducing clocks in a more subtle way for PR purposes.

Best bet is to run a p-core clock that allows lower voltages and don't run the crazy high voltage that results from the default profile's high ACLL unless your CPU is just not stable any other way.

 

[docnorth]

An updated microcode was expected anyway, simply because 0x129 does not work as it should. Running CB23 with Intel default settings and dual tau boost enabled, CPU draws 219w for a second or less and then stays at 65w for 10 minutes. With 'enhanced performance' settings CPU reached 27351 MC and 2012 SC despite very high ambient temperatures, it was 23 Aug and my LG heat pump had (another) malfunction. At least after bios update the package power doesn't rise to almost 290w before thermal throttling.

 

[dyonoctis]

It is funny how they called degradation "reliability aging"

The real reason for quick degradation - too high frequency, which is the underlying cause for the elevated temperature and voltage causing high electric current density, is missing from their list of causes.

I am not convinced that even a brand new CPU running the 12B microcode will reliably work for long years at those extreme frequencies.

Seeing How the 13600K is also part of the CPUs affected by degradation, it means that anyone running their CPU above 4.8Ghz is still at risk haha

 

[Nater]

I'm not going to ditch 13th/14th gen, but I'm hoping for rock bottom pricing for 14700k, so I can upgrade to that from 13600k dirt cheap

Same. I had always planned going to a 14900K someday, but they're still too damned expensive.

 

[Daven]

Another confirmation that motherboard makers were not to blame whatsoever. This was an Intel chip problem. It happened. Time to move on.

Edit: Oh and the simple answer presented by other commenters is the best answer…Intel ran their chips at too high of frequency.

 

[Nater]

Intel it's doing RMA's to damaged processors. I had a 13900K and Intel RMA me a 14900K without asking anything. Nobody, or almost nobody, it's going to throw a 600€ CPU because it's there a new one with almost the same capabilities (and also a new motherboard and new ram in many cases).

However, the damage it's done. Intel had a very good reputation about hardware degradation over the years, and maybe some users will think in going with AMD after this, if AMD it's still competitive in some years.

Got me wondering if I should get shady and throw a cheap cooler on my 13600K, burn it up, RMA, and get something hope to get something better in return.

 

[chrcoluk]

i just wonder how many are going to ditch 13/14th gen when arrow lake arrives? i think the damage is already done for that platform

I am in no rush to do a platform swap, plus I think the new Intel chipset might be heading the same way as AMD with inferior connectivity compared to Z690. Also if the day comes I do another upgrade, it may not even be Intel, like last time I will look at AMD options.

 

[mtosev]

Intel is very late with these fixes. Unacceptable.