Intel Ships 0x129 Microcode Update for 13th and 14th Generation Processors with Stability Issues

[Steevo]

The fact that Intel said they used the “default settings” when they have and had such a broad definition of what default is means a lot of users are going to see double digit performance drops.

 

[vMax65]

Either god sample or it's ignoring your voltage setting. More likely it's the latter.

Been monitoring voltages closely with HardwareInfo and as importantly temps will tell the story immidiatly if voltage hits to high. Have you actually tried a fixed or adaptive with a 1.350v limit? Also on MSI dropped Lite Load from a high to 3....Also now running the latest microde and again voltages beiing ignored would have shown up.

 

[Dr. Dro]

Been monitoring voltages closely with HardwareInfo and as importantly temps will tell the story immidiatly if voltage hits to high. Have you actually tried a fixed or adaptive with a 1.350v? Also on MSI dropped Lite Load from a high to 3....

My CPU becomes unstable in applications like Cinebench with anything below 1.375 to 1.40 range. Are you positively sure that your effective clock is 6000 MHz @ 1.35 V while running an AVX workload? I guess it's possible - you just need a CPU with a sky high quality rating. Do you have an ASUS ROG motherboard? (Maximus series, Strix doesn't have the silicon quality predictor functionality)

 

[vMax65]

1.55v is insanity on a consumer chip for the better part of the last dozen process nodes lol

Heck even 1.4v is quite the push already. Its crazy that these chips can even do this on stock. Conservative... lmao

Feels like people have been those frogs in a slowly warmed pot of water. Oblivious to the fact they're about to melt as the limits slowly got pushed up, and up.


Because that's also where Intel sets its limits, or used to. I remember them saying 1.5v and beyond is no go territory. And it always used to align quite well with the practical outcomes too.


Fine, but not great, and that's where they need to be, because their competition is.

Totally agree, 1.4v or more is absurd and some motherboard manufacturers on auto/enhanced settings were pushing even more but I would have thought that those buyiing anything with a 'K' CPU would know a little bit about these CPU's as they are for tuning though I can also understand those that just wanted the best/fastest or perbuilt and leave things at auto and Intel should have forced motherboards to have a default profile that actually stuck to the Intel defaults.

My CPU becomes unstable in applications like Cinebench with anything below 1.375 to 1.40 range. Are you positively sure that your effective clock is 6000 MHz @ 1.35 V while running an AVX workload? I guess it's possible - you just need a CPU with a sky high quality rating. Do you have an ASUS ROG motherboard? (Maximus series, Strix doesn't have the silicon quality predictor functionality)

Screenshots attached

 

[Wirko]

Intel is fine.

Well, they've replaced the rotten cherry on the rotten cake. At least that's the impression of Intel right now.

 

[vMax65]

Attached the screenshots in the post before this: Running 14900K, 4 stick of 16GB Trident Z5 6800 RAM so 64gb running at 6800Mhz and MSI Z790 Carbon Wifi with a Arctic Liquid Freezer III 360 AIO. The only changes are the latest 129 microde BIOS and leave everything in auto except vcore whcih is fixed to 1.350v, CPU Lit Load at 3 and 320w...when normal use I drop it down to 1.325v and a 300w limit.

 

[john_]

Fine, but not great, and that's where they need to be, because their competition is.

Even if their competition is in great position, some tech channels and probably tech sites, will convince their audiences/readers that's in fact, the other way around

 

[Visible Noise]

wonder how many 14700k and 14900k are gonna hit advertised 5.6hz and 6ghz now hahahahaha

All of them?

 

[vMax65]

Even if their competition is in great position, some tech channels and probably tech sites, will convince their audiences/readers that's in fact, the other way around
View attachment 358399

Just remember that all the manufacturers have crudded up in the past and none should get a pass and it his how they deasl with the mistake that will mater. AMD with the absurd benchmarks claims, initial Ryzen CPU's not hitting advertised boost clocks etc, Nvidia selling cut down GPU's with the same name as a good GPU to con buyers, AMD following suit and then to rub salt in the wounds charging inflated prices with AMD doing the same and on and on...Intel now crudding up big time and deservedly getting battered as they should....None of these manufacturers are our friends but rather beholden to shareholders and if they can get away with it, they will but to have one of the big two drop out will only mean worse for us as we will be stiched up not only on pricing but products that only very incremenatly improve if at all, so compition no matter how cruddy the manufactures is key...We needed AMD to keep Intel honest and it is the other way around...

 

[Dr. Dro]

Totally agree, 1.4v or more is absurd and some motherboard manufacturers on auto/enhanced settings were pushing even more but I would have thought that those buyiing anything with a 'K' CPU would know a little bit about these CPU's as they are for tuning though I can also understand those that just wanted the best/fastest or perbuilt and leave things at auto and Intel should have forced motherboards to have a default profile that actually stuck to the Intel defaults.


Screenshots attached

None of your cores are reaching 6 GHz under this workload, this is why you get away with 1.35. This is to be expected, though. It doesn't (and shouldn't) run high-current workloads at eTVB target. Still an excellent result, by the way. Your chip must be a pretty good sample. 6 GHz will require these high voltages in the mid 1.4s to low 1.5s range.

Even if their competition is in great position, some tech channels and probably tech sites, will convince their audiences/readers that's in fact, the other way around
View attachment 358399

Everything that HUB has said on all of these videos is objectively, demonstrably and provably true - the Chinese motherboard is trash, AMD marketing lies regarding these 4-year-old "new" CPUs, about Zen 5 being underwhelming and/or overpriced, etc.

 

[RogueSix]

I definitely suffered quite a bit of performance degradation (in addition to the hardware degradation so thx a lot Intel ) in Cinebench 2024. Single core is fine. My 13900K scores 130 points. But multicore is way down from 2132 to low 2000s. The first result was actually in the mid to high 1900s but I did three more runs that all ended up in the low/mid 2000s. Never got even close to hitting 2100 or above again.
One small caveat: It is pretty hot here in SW Germany at the moment so the CPU may be hitting some thermal limits sooner. I will have to try again when there is a non-beta BIOS and when it is cooler again.

System specs: ASUS ROG STRIX Z790-E Beta BIOS 2503 / 64GB DDR5-6600@C32 XMP / 13900K (Intel Default Extreme Settings 253W/253W PL1/PL2).

 

[Dr. Dro]

I definitely suffered quite a bit of performance degradation (in addition to the hardware degradation so thx a lot Intel ) in Cinebench 2024. Single core is fine. My 13900K scores 130 points. But multicore is way down from 2132 to low 2000s. The first result was actually in the mid to high 1900s but I did three more runs that all ended up in the low/mid 2000s. Never got even close to hitting 2100 or above again.

ASUS ROG STRIX Z790-E Beta BIOS 2503 / 64GB DDR5-6600@C32 XMP / 13900K (Intel Default Extreme Settings 253W/253W PL1/PL2).

253 W is the "performance" profile on the KS, and it's noticeably slower than the 320 W profile. Has always been. Something I've noticed when I moved from my MSI Z690 Ace to the ROG Apex Encore is that ASUS adheres to spec much closer, the MSI board runs wild in comparison. Consistently higher power allowance (and CPU benchmark scores) with the old motherboard.

Something to account is that the new profile has a lower ICCMax cap and it's very careful not to exceed spec here. It was one of the changes of 0x125

 

[iameatingjam]

"Intel's current analysis finds there is a significant increase to the minimum operating voltage (Vmin) across multiple cores on affected processors due to elevated voltages. Elevated voltage events can accumulate over time and contribute to the increase in Vmin for the processor."

This is just confusing wording to say "high voltage causes something to happen, which, over time, makes the processor unstable at default voltage, so it would only work if we gave it higher voltage"

I was more thinking about the wording root cause.... I just don't recall seeing them ever say elevated voltage was the root cause of stability issues.' But again maybe they did and I missed it.

Because if they did say that, it would mean there's no other problems, and if they didn't it would mean maybe there is.

 

[RogueSix]

253 W is the "performance" profile on the KS

As I stated in my post, I have a 13900K, not KS, and the Extreme profile for the K is definitely 253W/253W (PL1 = PL2). It says so in a huge box with an info text when you select the Extreme profile in BIOS.

 

[iameatingjam]

wonder how many 14700k and 14900k are gonna hit advertised 5.6hz and 6ghz now hahahahaha

Well neither of my 14700kfs ever seemed to like staying at 5.6ghz for more than like a second. I dunno if thats power limits or because it wants only for single core/dual core loads. Either way, I don't care, I'll be keeping all my pcores at 5.5. As long as they can hit that SAFELY I'll be happy. I'm downclocked to 5ghz right now just cause, well you know, there were failure reports everywhere. I'll do the update. But I think I'll wait a little while longer before I bring the clocks back up....

 

[Philaphlous]

How sure are we that this wasn't intentional by Intel? I mean it kinda has Office Space theme to it anyway... they just got caught before the place burned down...

 

[iameatingjam]

How sure are we that this wasn't intentional by Intel? I mean it kinda has Office Space theme to it anyway... they just got caught before the place burned down...

What was intentional? Dealing with risky voltages? Yeah I think that was intentional.
Purposefully degrading chips before their warranty is up? No I don't think they'd want to do that

 

[vMax65]

None of your cores are reaching 6 GHz under this workload, this is why you get away with 1.35. This is to be expected, though. It doesn't (and shouldn't) run high-current workloads at eTVB target. Still an excellent result, by the way. Your chip must be a pretty good sample. 6 GHz will require these high voltages in the mid 1.4s to low 1.5s range.



Everything that HUB has said on all of these videos is objectively, demonstrably and provably true - the Chinese motherboard is trash, AMD marketing lies regarding these 4-year-old "new" CPUs, about Zen 5 being underwhelming and/or overpriced, etc.

Thanks, maybe for the first time ever I have recieved a better than normal CPU. Just for clarity, I never said all core 6Ghz, what I did say was 2 cores hit 6GHz and all core is 5.6Ghz. For a lot of people out there with instability or even temprature issues, maybe running a fixed or adaptive with a max vcore limit at under 1.4v (1.350v for me) is the safer way to go as you still get great performance without worrying about vcore spikes into the mid 1.4v to 1.5v and much lower temps. Also, stick a max power limit that you are good with and be done....Anyway fingers crossed for the future as at least warranties have been extended by two years and they are now replacing CPU's under RMA pretty fast though not having a CPU for a week or two is a pain....

 

[ShrimpBrime]

How sure are we that this wasn't intentional by Intel? I mean it kinda has Office Space theme to it anyway... they just got caught before the place burned down...

Ahh... the wonderment of it all.

Like at one time could BCLK 13th gen and they took it away. (I have screen shots)

So they very well can control the cpu through Management Engine and with ME pushed through windows updates, Bam, got what they wanted.

Ah yes, that speculation that got the company in trouble in the first place. But will anyone ever really know the truth? Not likely.

 

[thesmokingman]

Congratulations,you broke it. Too late?
What did Intel expect from turbo boosting the thermal velocity boost the turbo boost in the first place to achieve?

I dunno, to beat AMD??

 

[FoulOnWhite]

The updated bios (beta) appeared for my board, Rog strix z690-a WiFi d4, yesterday. Not much point me updating it though as I will never have a 13th gen in it.

 

[W1zzard]

I was more thinking about the wording root cause.... I just don't recall seeing them ever say elevated voltage was the root cause of stability issues.' But again maybe they did and I missed it.

Because if they did say that, it would mean there's no other problems, and if they didn't it would mean maybe there is.

Oh, I get it now. Elevated voltages must have a reason, in some meeting someone said, let's use x.xx V. Probably based on some data from an engineer, who based his testing on something else, using parameters provided by someone else .. at some end of the reason chain is a human who made a mistake, certainly not intentional. This is probably not what everyone wants to hear though

 

[FoulOnWhite]

The voltages should always be the lowest the CPU/system will run at reliably, boards always see to exaggerate them if left on auto.

 

[chrcoluk]

Is it just me or could the title be misinterpreted to sound like 129 is going to fix instability issues on cpus already experiencing them, rather than just preventing it on those that aren't ( as intel has said) ?

Also noticed this line

"Intel's analysis shows that the root cause of stability problems is caused by too high voltage during operation of the processor."

Did they say that? I remember them saying this in the July press statement:

"Intel is delivering a microcode patch which addresses the root cause of exposure to elevated voltages."

Which is not quite the same thing. But I could be mistaken, maybe they said that at a different time.

This is the first statement they have made (at least that I have read) where they said they have identified and fixed the root cause.

The eTVB bug was something they discovered whilst investigating but they clearly stated it wasnt the root cause, the earliest news about board vendors shipping whacky defaults was also never claimed by Intel to be the root cause.

Not sure if there is a point for me to install the new microcode, as I have no eTVB (i7) and my voltage requests are getting nowhere near 1.55v, they not even hitting 1.4v. Fairly happy with how I have the chip configured. I now think after seeing igor's binning data I have a better then average i7.

The fact that Intel said they used the “default settings” when they have and had such a broad definition of what default is means a lot of users are going to see double digit performance drops.

The bios has already been benched, its barely any difference in performance. Those users you talk about, if they want they can put their system back out of spec 4095w power limit, whacked AC/DC, ICCMAX etc. Microcode should still cap their voltage requests.

 

[Ruru]

Probably now the CPUs hit their advertised turbos for just a blink of an eye so a monitoring software can report that it has been running at those clocks.