The fix is in- Intel microcode update to prevent stability issues/degradation
ฝัง
- เผยแพร่เมื่อ 12 ก.ย. 2024
- The ongoing Intel 13th and 14th gen stability issue saga has now resulted in an new microcode update 0x129 being released to prevent further degradation by limiting Vmin to 1.55V. If you have a 13th or 14th gen CPU you should check your motherboard BIOS page and apply this update as soon as it is available for your board.
Sources:
community.inte...
community.inte...
community.inte...
www.reddit.com...
www.reddit.com...
www.reddit.com...
www.reddit.com...
www.reddit.com...
• Intel's new Microcode ...
What equipment do I use to make my videos?
Camera: Sony a6100 amzn.to/3wmDtR9
Camera Lens: Sigma 16mm f/1.4 amzn.to/36i0t9t
Camera Capture Card: Elgato CamLink 4K amzn.to/3AEAPcH
PC Capture Card: amzn.to/3jwBjxF
Mic: My actual mic (AT 3035) is out of production but this is a similar mic (AT 2020) amzn.to/3jS6LEB
Portable Mic attached to camera: Rode Video Micro amzn.to/3yrT0R4
Audio Interface: Focusrite Scarlett 2i2 3rd Gen: amzn.to/3wjhlad
Greenscreen: Emart Collapsable amzn.to/3AGjQXx
Lights: Neewar Dimmable USB LED amzn.to/3yw4frD
RGB Strip Backlight on desk: amzn.to/2ZceAwC
Sponsor my channel monthly by clicking the "Join" button:
/ @danielowentech
Donate directly to the channel via PayPal:
www.paypal.com...
Disclaimer: I may earn money on qualifying purchases through affiliate links above.
The problem is that normies don't even know what a BIOS update is.
One of the many things you learn when you are using a PC. UEFI BIOS in no different then any other application installed on the hardware in terms of use.
@@valentinosgsxr It is different thou. BIOS is vital for a PC to run, any other app is optional ! Also, you can damage your PC if you mess with BIOS and don't know what you're doing. Let's not simplify things that should not be simplified !
@@Qs_Internet_Cafe There aren't many things that are more simple than flashing BIOS to a PC. You underestimate yourself and others saying so.
@@valentinosgsxr Except I can install Windows and any other app during a thunderstorm and the worse I get I need to re-install Windows. As far as BIOS is concerned depending on the motherboard you can brick your system. Driving a 75HP car is different than driving a 500HP car. Sure, it's still driving but it's not the same ! Anyways, I believe we are speaking the same language but interpreting some words differently. The cons of typing :)
@@Qs_Internet_Cafe I know what you mean. For me it is just hardware. I always backup everything important regularly and before firmware installations. If it dies, bad luck, I am not going to cry over a dead PC mate. I would claim the warranty and build another.
The person @15:12 - That board (MSI PRO Z790-A WIFI DDR4) doesn't have 0x129 yet, the 'new' bios they are testing is 7E07v1D with 0x125 which was released on the second of August. Also the reason they have such a performance drop is that MSI is being very conservative with the load lines and so they're imposing the PL1/2 limits but still blasting voltage and thermal throttling. Adjusting the Lite Load setting in the MSI BIOS to adjust the load lines down should eliminate that.
Things will get very confusing with posts about 'new' BIOSes. My own board (MSI Pro Z790-P WIFI DDR4) got a 'new' bios with 0x125 today, after the first beta BIOS with 0x129 were released, so I'm not going to update yet.
Edit: Now I know what CVE-2024-3687 is, I've decided to update immediately.
I have the 13600k and MSI b660 A Pro WiFi board. Which update am I meant to be installing? Or should I wait till further iterations are out? I’ve not had any issues that I know of.
@@Raxiel497 Undervolting the CPU is lightyears better than adjusting the load lines.
People are just terrible at giving advice as well, i heared another BS like sync all P-cores to 5.7GHz when the all P-core boost clock of the 13900K is 5GHz.
@@rover1374 Wait and go for a decent "Lite load" setting in the BIOS between 4 and 12. Get a hwmonitor and check it does not exceed 180w at full blast i.e. Chinebench and you are good.
@@rover1374 I think the new BIOS are only available for Z790 chipsets at the moment, but keep an eye out because it should roll out to all chipsets over the next few weeks.
@@Brakballe thanks, will wait and then do those things. I already have HWmonitor for temps but will use it for power now as well.
Well we can ignore the reddit post at 15:05 as this person with the MSI Pro Z790-A DDR4 Wifi is using the older 0x125 microcode, MSI hasn't updated this model yet with 0x129, supposedly all 600 / 700 series MSI boards will be updated by the end of August (per MSI statement).
We shouldn't be ignoring it. It is one more example of false accusations against intel's platform.
ya MSI did release beta bios' on Aug 2nd so maybe they thought it was the new 0x129 update.
@@valentinosgsxr "one more example of false accusations against intel's platform." - I guess all those server owners lie also ? Shill much ?
@@Qs_Internet_Cafe It makes big difference by the way you interpreter what you read. For example, "The failure rate we have observed from our own testing is nearly 100%, indicating it's only a matter of time before affected CPUs fail". He doesn't say what the failure is, he doesn't say what exactly is the failure rate and what do you get from the is a matter of time? Weeks? Months? Years? Eventually every CPU will die. The affected CPUs will fail, means that the CPUs haven't fail yet... What I see in this report is a bunch of not proved accusations and this is what actually is. The hole writing is more like an AMD advertisement. I couldn't take all this serious.
MSI did release bioses for the 0x129 mcode. I am using it atm.
Well rip EVGA motherboard owners.
you know you can just UV your CPU right?
it "basically" works the same as this update.
@@MaxIronsThirdno unless you do a static undervolt. Adaptive UV is just an offset. It doesn't limit the voltage. The bios update is a hard limit of 1.55V
@@pretentious_a_ness static uv yes, the only chips I do a regular uv are the X3D cpu
@pretentious_a_ness You can do voltage offset per cpu thread boosting behavior
So 2x thread voltage offset, 4x threads, 6x threads etc
@@pretentious_a_ness You are definitely wrong, you should try first and see for your self. Actually there are more than one ways to reduce the voltage your CPU receives.
Exactly same thought here , these games with worse 0129 results may have already degraded.
@@skye7690 Poor performance is not a sign of degradation, is a sign of bad silicon quality. If your CPU has a higher VID, your frequencies will be lower after the update.
@@dethskullcrusher Great observation!
I know that everyone is saying the sensible thing is to install this update/microcode NOW but I think I'm gonna sit tight and wait a little while to see the real effect of it on users. Some comments under these videos are somewhat...concerning. To say the least.
yeah, also never buy new cpu atleast a year after release
Sorry, that's kinda like saying I know got steel balls bouncing around in my crankcase but I'm gonna let them keep bouncing, sit tight and see if removing is in my best interests.
@@freedomearthmoon1 That too
Same, i want to wait a week or two scouting reviews, reddit post etc, looking for the experiences of those who update... Let's hope it ends up working in deed in the end
I tried on my i7 14700K and it actually gave me more performance. It only limits the voltage requests up to 1.5V, if your CPU doesn't use that kind of voltage anyway, you won't notice a difference. If your BIOS is older and it does not have the intel profiles as an option, then yes, you will see difference but it will be because of the intel's limits will be applied after the flash.
finally a real fix coming up again! I am so excited to see how the NEXT update going to fix this problem again!
Friend updated bios and now windows won't boot unless in safe mode. Removing mobo drivers from device manager and booting up works fine until those drivers inevitably get reinstalled, either from the mobo itself or from windows automatic update. That's tough.
Flash it again. Might be a bad flash. There's usually 2 ways to flash it, bios flashback and then the utility in the bios setup. However he did the first time do it the other way the second time.
It happened to me that a BIOS meshed up the usb ports and it wouldn't boot whenever I had an external drive in a particular port. Rolling back the BIOS fixed it.
@@pf100andahalf but those things a loaded up on integrity checks aren't they?
@@kittenburger_prime Data corruption can happen whether there are integrity checks or not.
@@pf100andahalf Not really that's the point. If the software knows how to check, and it does the check, what kind of strange magic do you think makes the corruption possible?
Since the day I sold one of my livers (lol) to buy a 14900K, I was so worried about mounting everything right that I went into a JayZ video teaching to turn of mobo overclocks and let it at stock (253W). I did it for trust and heat issues (way before this major intel problem came up) and maybe thats what saved me. My CPU works perfectly so far
Same, my 14700k has been working flawlessly. I didn't turn anything off but I did power limit it to 253W which is Intel's max rated clock speed. Cooled with a Noctua NH-U12A
Same. I have a 13600k and from the day I built the PC I have always enforced Intel's limits and have never had an issue with the processor.
Installed the update today. Other than all my bios settings being reset to default I've had no other issues. My temps are even a little bit lower. It'son 38 39 when idle where before it was 41 42. On i7 13700k
I don't think the microcode can tell if degradation happened, so all CPUs will run the same performance independent of degradation, just the degraded CPUs will be unstable running these speeds.
@@MaxIronsThird That's... not how it works. As long as thermals allow it (and the degradation very likely doesn't affect thermals in a significant way), the CPU will try to boost because it doesn't know that it's degraded... and promptly crash. Both a degraded and non-degraded ones will try to boost the same way, just that the degraded ones, because they physically can't do that, will be unstable when they try anyway.
@@Barteks2x neat explanation
I find it funny that TH-cam decided to trim the video's name on my Sub page and it ended up being "The fix is in- Intel microcode update to prevent stability..."
Basically, what Intel is doing here is delaying the degradation process with this new microcode update. There should be a discussion about what is caused this problem in the silicon and how that occured in the first place. It will help users decide whether or not to consider Intel in the future.
Most likely cause is too high of voltages and/or too much heat. That's what causes Electromigration in IC's and the denser the transistors the more likely it can happen which is why Mission Critical systems (Military, medical devices,aerospace etc.) are still on 28nm and 40 nm nodes
@@longjohn526 That is true. So my question is whether the underlying cause of this issue is material defect or something in these CPU designs itself?
I 100% believe they are delaying the degradation process to exhaust the warranty period on the majority of customers, so they don't lose a ton of money. They knew something was wrong and still decided to release the chip when they could have pushed releasing it back a couple of weeks. We are now seeing them kick the can down the road and hoping that nobody makes a class action lawsuit, sad to watch really
@@Afurai_ Yes, they seem to be deceiving customers by hiding information that should otherwise be known. It seems this is a trend that many major semi-conductors related brands are doing. Things are very unreliable these days.
I wonder if this has something to do with work productivity or something wrong with their manufacturing process itself. Does the situation in China have something to do with all this?
I'm skeptical whether a cpu can detect it is degraded. The behavior should be identical regardless of degradation, except for stability issues.
It is perhaps more plausible that CPUs with bad VF curves more frequently hit the new cap, and thus have a greater perf hit - regardless of degradation.
Are they comparing 125 with 129 or like 104 with 129? 🤷♂️
Hadn't Intel mentioned a tool to understand the CPU's degradation status?
Talked about it but not yet released it
I just updated my asrock pg riptide sporting a 13600k with the 0x129 microcode today. Went smooth. No issues.
what mobo u got? gigabyte didnt release theirs yet
@@dastutw asrock pg riptide
@@dastutwAsrock z790 pg riptide.
No hiccups, huh? Hmm...maybe I'll try it then. I just checked Asrock's BIOS downloads for my specific model (Taichi Z690). They haven't updated it yet, it's still the one with the previous microcode 🤔
Did you notice any differences?
Damn I'm sure glad mine been sitting at 1.23V this whole time.
I did the MSI update an immediately ran SC2 which was crashing on campaign constantly and on multiplayer randomly after 12 hours of playing not one crash so far, to my surprise
it's working fine now then? did you see any difference in power consumption/voltage with HWINFO?
@@MaxIronsThird for my 13700kf vcore went down by like 0.03 at peaks; previously vcore was reaching 1.41 now it reaches only 1.38; but either way bot values are safe; it s mostly i9s that had problems bcs they were reaching like 1.7 or more; that s what was causing degradation; VID also went down by 0.04; temps in gaming and cinebench went down by 4-5 degrees C; and performance wise no difference
Intel is the GOAT.
@@realbeetlejuiceeh?
@@KoItai1that's what am getting around 1.4 xmp 1 on and chip clocked at 5.4 with 280mm aio with the p14 max fans coolimg nice when I run a heavy game up be lucky to get 1.370v under heavy load haven't lost no performance to
i'm glad this showed up in my recommended, thanks for the update
MSI still has not published 0129 for my MSI motherboard (Z690), but I’ll keep checking their website.
same here! Not sure i'll even update to the new microcode. my cpu seems to be fine and it's not pulling crazy high voltages from what i've seen and monitored for hours while gaming and running other things. I also only have a 13700k so i should be less likely to have an issue anyway.
same here :\
They're also still pushing out new 0x125 bios updates for some boards, which will confuse a lot of people into thinking 'this is the one'
ME Firmware ver: ME_16.1.30.2361
This one PRO Z690-A WIFI DDR4
Look what happens when companies tries to do an overclockers job... Lets go back to relaxed CPUs and leave me my headroom to choose if I wanna fuck up my hardware! ;)
Yup, exactly. They burned themselves really bad for their greed this time. But there is a bright side for actual enthusiasts, there will be plenty of monster cpu's on the market at discounted prices that they know how to make them work, don't need microcode updates and premade settings from mobo manufacturers for that.😁
AMD tried that yesterday and everyone lost their sh!t
true
@@douglasmurphy3266 what? Zen5 is an awful upgrade over Zen4, their new lower TDP is to pretend they chose efficiency over performance, guess what, they got neither.
Minor caviat is that major oem's may have build-in mechanisms to update the bios, I know HP has that at least.
As for performance impact, could be that some users just updated the bios now, so combining the intel default settings which has a performance impact with this microcode update.
Lenovo also pushes firmware/bios updates to at least some devices thru windows update.
My two cents on this topic is that because of the oxidation caused by mistakes during manufacturing, the processors have started degrading and atarted requesting higher voltages because of that, in order to maintain performance. And because they requested higher and higher voltages, they degraded further and fuerther to a point where this isssue couldn't go further unnoticed by the public and that is where we are. I think that the update could prevent stability problems on CPUs that haven't had any problems yet, but because of the oxidation problem, even those CPUs will be impacted on the performance-side once they reach a certain age. And that certain age is the age at which the CPUs that have stability problems NOW have started to have those problems. So in the end in my opinion there's no solution to chips that have the oxidation problem. As for those that don't, those don't even need the microcode update. A side note at the end: I wanted to buy an Intel laptop in early July but since these problems started getting more and more common, I now have no other choice than AMD. And now I'm in a position where in my country I could get an Intel laptop which fits my budget, that absolutely crushes AMD's every option within my budget. But I still need to go for AMD. Intel have put themselves in a position so bad that their own shareholders are suing them right now. And the fact that they made a mistake with these processors in and of itself is not a big deal, it can and will happen. But the fact that they basically ignore their customers and hide every single information avalible to them is how they are currently digging their company's own grave.
Well they hid the oxidation issue for like what a year? And they denied RMAs. A lot of people had a dead chip in their hands. Mistakes can happen but if you ignore them and do everything in your power to hide it and avoid try to the consequences now that causes issues for other people which can't be ignored.
Exactly. And those who have dead cpus have just lost potentially $200-300 because Intel didn't want to take responsibility.
That sounds like good thinking.
The title is a real piece of art. In my case on a smaller monitor it shows: "Te fix is in- Intel microcode update to prevent stability" 😆
This has pretty much been Intel's business model over the last few years 🤣
Damn! That White flash 😱
i have noticed for the asus mobo, it states that this update is not for the k processors. and other areas on the internet it says that it is for the k processors, i am just wondering which is true and which isn't before i update
Also curious about this
I'm a patient man when I can afford to be, and with a 5700X backup gaming PC, I can. Though I have observed no instability issues to date, my 13700K (in use since late 2022 but never hammered), is going to remain out of service until we see a fair amount of independent testing to ensure the 1.55V limit actually works. As of this video's release, the new BIOS w/ 0x129 for my ASRock Z690 Taichi DDR5 board isn't out yet anyway, and even the 0x125 BIOS is still in Beta. Their Z790 version IS out, so looks like they're doing 7-series first.
you can just undervolt your 13700K and keep it under 200W, you won't have any degradation problems.
@@MaxIronsThird unless the chip has already degraded. I had a 13700K that started having stability issues after 13 months of pretty heavy gaming use. No matter what I change BIOS settings to, update to, etc., the ship has sailed, and it crashes almost every time I try to play a game.
I got fed up a couple of months ago and swapped to a 7800X3D on my main PC, couldn't be happier now. This entire thing has been just a giant shitshow
Hey Daniel, former overclocker here. "Vmin" can be defined as the minimum operating voltage of the CPU - what intel is investigating when a CPU degrades is a shift in Vmin (going up) in regards to the minimum operating voltage of the CPU (ie. needing more voltage for stability). What this microcode update is limiting the maximum operating voltage.
I installed the latest bios and firmware update to my msi tomahawk motherboard as I have 14700k. I immediately regretted it and rolled back to the old stable bios which was released back in January. The reason for that was the temperatures with the new bios skyrocketed. in cinebench it hit immediately to 100c with all p cores and I want to note that I also chose the intel recommended settings in the bios. Also I checked the voltage during the tests, the cpu was hitting 1.5V and the cinebench score went down to 30k from 33k previously. I also played some cyberpunk and the result was the same. Fps dropped and cpu was around 90c constantly. So with the old bios and cpu cooler setting to box cooler, my cpu only hits 1.3V max and during the same tests no thermal throttling. So I have no idea what was fixed here?
I updated to the latest bios on MSI Z790 GAMING PRO WIFI for my i7-14700KF hoping it will resolve the crashing and bluescreens and unexpected shut downs. I just gave up and returned back to my MSI B550 Gaming Plus paired with AMD R9 5900x no issues at all. I just wasted a lot of money sadly on Z790 GAMING PRO WIFI and Intel CPU
It's sad to see that your 14700KF already degraded.
My gen4 blown up one day and put my house on fire. I returned to my good old 14900K and it's working like a charm... in my new apartment. True story. PS AMD refused RMA because there was nothing left of my CPU to return to them.
@@valentinosgsxr Yeah, doubtful. AMD allowed RMA as the issue was ASUS motherboards, not their CPU.
You should RMA your intel CPU because the microcode update will not fix already degraded CPUs.
I'm running same mobo and i9 13900kf , no issues after nearly a year of constant use. I won the lottery I guess!
For Asus motherboards, it is labelled as a Beta BIOS that supposedly has the fix. I don't want to flash a beta BIOS that may or may not void the warranty on all of my PC components.
Edit: also, in the description of the beta BIOS, it is stated that this BIOS update addresses the instability problems happening on the 13th and 14th non-K SKU CPUs. Meaning, by their own admission, this "fix" is not meant to be the fix for all of the K-SKU line-up from 14900K/KS all the down to the 13600K. It only addresses the non-K SKU parts and I don't know why none of the tech TH-camrs are catching on to this. The actual fix is still not here yet.
Classic. Here is the fix which voids your warranty, also never overclock your K series CPU over the base line...warranty etc.
@@martinxyz also, in the description of the beta BIOS, it is stated that this BIOS update addresses the instability problems happening on the 13th and 14th non-K SKU CPUs. Meaning, by their own admission, this "fix" is not meant to be the fix for all of the K-SKU line-up from 14900K/KS all the down to the 13600K. It only addresses the non-K SKU parts and I don't know why none of the tech TH-camrs are catching on to this. The actual fix is still not here yet.
That's not what it says. It says it contains the 0x129 microcode *and* adjusts default settings for non-K-series processors. It's doing two things... updating microcode on all 13th and 14th gen processors, and it's also adjusting defaults for the non-K-series.
not true, micro code asus bios update was for my 14900k. So far so good!
I have an i5-13600K, i have not had issues and Ive been using it since release day. Im not going to apply patches at this time, im not seeing issues. If its not broke, then dont fix it.
Ill hold on to the i5-13600K to see what happens down the road. Im also going to continue to use 13th/14th gen, prices will drop when Arrow Lake arrives.
Ive also been using Arc gpus for awhile now, Intel tech has enormous potential with XeSS, oneApi and Deep Link.
Also this microcode update may slow things down so I'd rather not mess with things. Is Intel stating that all 13/14th gen users should upgrade?
@@xlr555usa Just people with i7/i9s.
Had no idea this was an issue until 3 days ago when my motherboard died. I've been using the 13900k since last july. Is there anyway to tell if my cpu is trash now
11:57 13900K Cinebench R23 performance drop from 40K to 37K drop matches to profile change from PB2: 350W to PB2: 253W. This performance drop is in performance profile choice.
How is single thread performance?
@@Mike-kr5dnST performance should stay the same. Those are not power limited.
I've got a 12th gen cpu so i'm not in a hurry to update, but is curious to me that nobody mentions the fact that the available bios are almost all BETA BIOS not final release.
Do not update bios unless u wanna kill yo PC ❤
@@lil_per_vert How so? Can you please elaborate?
@@dethskullcrusher if you have a 12th gen why would you even want to update the Bios? Alder Lake is not presenting any issue.
@@jeantechnoir7702 Plenty of reasons, improvements in security, improvement in ram compatibility, is allways a good idea (unless you have an avx 512 12th gen cpu).
I update bios every 3-4 iterations since 8th gen. I usually wait to see other users experiences aswell.
Hi, hope everyone has a good day.
I just updated my bios so i can put i5 14600k cpu on my motherboard...lets hope its working
Thanks for the breakdown of these news.
I still don't trust the new eTVB algorithm for example as it still looks like it's drawing crazy voltages imo as i also did not trust the whole "stock" settings situation from the start, hence fine tuning my own voltage settings which saved me from any stability issues so far. Therefore i will not update the bios and continue with my settings because gaming wise the performance hit is minimal to none and i'm not that heavy into productivity apps to affect me on that front. I prefer stability over 5% extra gaming performance that will definetly affect longevity with those 1.5V+ imo, also with the need of crazy cooling to keep those high power limits viable. So thanks for nothing intel 😅
Still waiting for the MSI Z690-A PRO DDR4 bios update :(
The BIOS with the new microcode is still in Beta. For those concerned about issues with the update, I'd tell them to wait till it comes out of Beta at least. Intel announcement makes it sound like they'll be coming out with another update by end-August anyway.
Don't think they mean a microcode update by end of August. I think they meant they will report back to us about whether or not the vmin issues in degraded chips is fixable by microcode.
it is possible that the new asus bios may be somewhat faulty. tests with the same cpu on MSI new bios and Asus new bios show what appears to be a turbo boost variation on the 2 preferred cores, that are boosting as they are supposed to on MSI but below expected with asus (about 100 to 200 mhz) so it is possible that those 6% jay notice may be tied with asus only as on msi the difference appears to be within margin of error . Gaming performance in the vast majority of the cases (accordingly to different sources) appears to not have been affected, with variations being within margin of error at least for the 14900 and 14700.
Dont update bios
There are obviously biased testers and we shouldn't believe everything we hear. Even if the tester is legitimate, there are so many variables that all numbers should be considered as guidelines rather than raw values.
Throttling may also happen due to max current limit in intel defaults. It doesn't have to be temp throttling. If you enforce max current 200A on a 13600k any heavy multicore load will cause instant throttling to keep the 200A limit.
Im a regular person and I tried updating my bios and it black screened my PC 🤣 send help?
its all a shit show, look before u leap dont just follow advice
Updated my 14900k with the msi Z790 carbon motherboard bios and no issues whatsoever. One thing I would suggest to 13th and 14th gen users is to use an adaptive or fixed vcore voltage which is what I have always done as I noticed a long time ago especially with Asus motherboards that on auto settings they were pushing way too much voltage. 1.3 to 1.350v will give max performance and keep temps in check and some CPU's can do this with even lower voltage at 1.3v or even below...
I agree. I have done very extensive testing undervolting with offsets against undervolting with load line settings and always the offset works better. In my experience, AC/DC load line is not supposed to be used the way people is using it.
updated to 2503 on my ROG STRIX Z790-H with a 13900k in it. everything looks ok, though the BIOS screen hangs for a bit on my end. I am currently using the intel default EXTREME profile right now and everything is working well. No crashes and no real issues so far. I don't have time to game right now so I will test games later.
@@DustyCruz Do you have multicore-enhancement off with your i9-13900K and undervolting?
The all P-core turbo should be 5GHz.
Can you check the power voltages profile with the new bios? Someone reporting this new bios is setting power profiles back to the “unlimited” values (501a-4049w)
Thanks
@@saricubra2867 Multi-core is indeed disabled on my computer, and I am not using any undervolts. I did however, enable CEP and enhanced C-States.
The All P-Core boost on my end is 58, but I syncd them all to 55.
@@ALM_Relaxed On my Asus Board, it is set 400a, and 253w. as for the voltage? it is set to 1.259v I think.... and it isn't sending insane voltages to my cpu anymore. The highest I've seen my per core VID go to is 1.4v
The problem of my BIOS taking like... 30 seconds to get in to windows is my other worry right now.
And if you're on the same board as I am, this is a BETA release, so might be best to just stick to the 0x125 microcode BIOS, before going to this one.
TL;DR, I actually gained a smiiiiidge of performance, and no instability so far.
EDIT: I didn't even answer your question but no, they did not reset the power profile back to 501a-4095w. Out of the box, without any other changes, it remains 400a-253w for the extreme profile, and 307a-253w for the performance profile.
Fantastic video bub. In my opinion you deserve to hit 1mil subs by the end of 2025. Congrats on the sweet new pad too! Keep up the hard work, it will pay off
As for how one could verify that limit of 1.55V is safe - you could try to run the CPU continuously at that voltage and see whether it degrades. If it does degrade eventually, that to me would mean they only really slowed it down to a level that hopefully lasts past extended warranty. If it doesn't... it should be safe. If the CPU can't run under that voltage continously, short bursts will also eventually accumulate, it's just a matter of how long it will take (5 years? 10 years? 20+ years?). Also it's not Vmin they are limiting, it's the voltage requests made by the CPU. "Vmin" is the name for the minimum voltage needed for the CPU to be stable, which goes up as the CPU degrades. And as the voltage request patterns normally don't really change, as Vmin goes up, you see the CPU become less stable.
Mine never goes over 1.444V even without 0X129 update, but a very recent bios version is used both on 12700K and 13900KF on MSI boards Using Cinebench 2024 measuring with HWinfo 8.2
i have an i9-14900HX and I've still yet to see anything regarding my processor. My laptop is ungodly unstable and literally every game and application is crashing on it. I've been dealing with a lot of "workarounds" and can only minimize crashes, but not stop it outright. The fix needs to hurry the hell up, I didn't pay all that money to get a paperweight.
Your laptop is faulty. You should claim the warranty instead of waiting.
@@valentinosgsxr I'm most likely going to refund it and find something else.
Are laptop processors also experiencing elevated voltage issues? I have an i5-13420H with max TDP of 45W. Should I be concerned about a microcode update?
I've been away on vacation since the 14th of August. There was a windows update on my PC, but did the Bios update do anything? Should I download it and update?
I don't believe this can possibly be a fix.
it just seems unlikely that limiting to 1.55v instead of the 1.58-1.59 they sometimes request that caused it.
if they knocked it down to 1.50 or 1.48 that's a lot. that's 8% or something.
Thing is, the degraded cpu's will need more voltage over time not less, so it's a vicious cycle they entered that will probably kill all degraded i9's and i7's in a not distant future. I suspect that's the reason they still kept a very high ceiling imo of 1.55v. This voltage was considered dangerous until like 8 months ago, lets be real...
@@adriancioroianu1704 that makes sense.
@@adriancioroianu1704 They had to keep the 1.55V for the KSs to clock at 6200GHz. That's what it takes to get there. For the CPUs that can not clock that high, that is not an issue anyway.
@@valentinosgsxr I think hitting 6.2ghz is literally the last of the last care rn, both for intel and for people with KS's.😂
my 14900k was fine before the microcode update. It is crashing constantly after the latest microcode update. rolled back to the previous bios and it seems to be holding so far.
Actually no, if your processor already have the defect, you need to let it die and claim for warranty before the extended warrany end for a new actual good processor instead.
Intel is replacing cpu's now no questions asked, so if it crashes at all rma it.
Every processor is bad 😂
@@lil_per_vert I hear there are a small number that are okay but they'll probably go bad too.
@@lil_per_vert Good, pay back time, all buyers do not update it and send all the faulty processors back to Intel, if they do not recall them.
Bios updates via OS is very unreliable. For example on my Z690 Asus Strix D4, asus has actually tried this approach, the bios will update successfuly, however, this update from windows never goes away and will always reflash it back to this version, even after 2yrs. Asus/windows has not fixed this.
Great news, my 13600KF will survive, lol.
Not sure i will get this BIOS update though, seems to ruin cpu frequencies/temps etc what the hell... i'll wait for more testing from Gamer Nexus or Hardware Unboxed.
You will learn nothing valuable from them because they are AMD biased and they are not even try to help intel users.
learn to undervolt
MSI PRO GAMING Z790 motherboard with 13600K with 0x125. I have voltage offset -0.125 and it never goes over 1.13V and 60º, with 5.2/4.2Ghz OC, 24100 CineBench 23. Noctua D12. When running CB the most it goes is 73º
I regret getting a 12700k rather than a 5800X3D or an AM5 setup when I built my computer back in 2022, the hope that 13th or 14th gen Intel being a huge upgrade didn't pan out.
Changed my CPU via RMA process of Intel. (I had time to time BSODs, sometimes 2 per day, sometimes 2 per week). Intel sent me a new 13900K cpu in less than 2 days. Old one's SP score in BIOS was 108 (Asus Z790 Proart) and new one's SP is only 97. So, I guess, i got a bit worse unit, but at least new and with no degradation. I red a looot of reddit etc about right bios settings for 13900k, so the main idea is do not let CPU go over 1,4V in any case, so I just set max V cap to 1400 in BIOS, and played a little with Loadlines. My AC/DC is 0,9/0,2, LL set to 4.
My Vcore on idle is about 1.3, when runing C23 - drops to ~1,18. I also ignored 253/253W cap and set a bit higher. Because the main problem for Intel was overvolting, my voltage is set correctly, do not spike, so I decided to increase P1/P2 settings is 270/253. Everything runs good already for two days. Will avoid new BIOS update and wait a bit for more information and experiences.
My C23 score now is ~37000. Before everything i had 38500 on old CPU, but P1/P2 was set to 280/280. I mostly work with PS and LR, so stability is my priority, that's why I lowered it to 270.
P.S. I after 125 microcode I also had an issue with safe boot. I had to pick other OS, instead of UEFI WIndows (or smth), and then it boots up. For me this doesnt look OK to disable default settings in bios boot, to make it work.
Your 13900k CPU should not go anywhere close to 1.5v period, let alone 1.55v
Intel is diverting lawsuits, but sadly...it's inevitable
Are the laptop cpus affected? Should I be concerned about this as a peasant with a measly i5-13450HX?
Intel should recall the CPU`s period. This nonsense with microcode's and BS is just kicking this down the road. Shame on intel
they don't have the cash for a recall. they just have to wait for this to blew over.
I would have to agree with this. You have to stand by your product, and this is just kicking the can down the road.
Asus has it labeled as "beta update" so the wait is not over apparently.
For what I see so far all the related BIOs would be BETA because of the rush to release them due to the spread panic. So far what I experienced and see on the comments, BIOS is remain the same except for the microcode. The microcode itself is not BETA though. So I would say it is safe to update to BETA considering the conditions.
Yeah, it's labeled as beta for my MSI Z790 GAMING PRO WIFI
Just did the bios flash for my Gigabyte AD z790 board. i had some crashes inside ghost recon and jedi survivor i hope my cpu isnt stuffed. I had already swapped to intel from a bad 7700x so unhappy this generation.
I updated the MSI Z790 for my 13700k, don’t think I have degradation , but have had to undervolT to control temps in cinebench. Without changing any other variables other than the microcode changes and keeping the new bios profile to intel default, my core VID’s were at 1.35-1.37 , unnecessarily high and thermally throttled after 1-2 runs in cinebench r23 despite a 240 Lian Li 2 AIO.
So I had to go to Intel XTU and apply an adaptive undervolt to 1.125 ish and now core vid is back to 1.23-1.25 core temps are normal and don’t thermal during cinebench r23
The annoying thing is: Every Bios update i have to tune my fan settings and especially if you run a "quiet" pc its pretty hard to find the right values.
What’s your opinion on what to run? In my 14900k dark hero z790 newest update with intel default settings on r23 runs only around 36k which is a severe downgrade from asus Oc profile but could will constantly hit 100c intel limits barely hits 70c now
Have not gotten the update yet for my Asus 790 MB, but have been playing more with the settings. Reducing the max multiplier has worked well, but now I've increased the multiplier back to 50x, along with setting the IA VR voltage maximum to 1550mV. My core voltages have dropped as reported by i7z to 1.4V from 1.5-1.6V....and just as stable or more stable. Time will tell of course.
Why is reducing my voltage seemingly increasing my stability? Could overvoltage lead to leakage between adjacent traces/components in the CPU given the very small scales involved?
i7-13700k, Asus Prime Z790-P Wifi
How is this not in windows updates? Windows updates from windows xp days were so bad but they still have issues coordinating with companies. AMD driver conflicts are another one over and over.
I love how they just change the TOS and warranty and the terms it was sold on. If you are taking massive performance hits, I'd just get a refund on it.
Well, this whole thing with Intel sucks. I know AMD had their shit show many years ago and bounced back. I'm an AMD guy and want the best for everyone involved. Hope Intel honors all of the RMA's coming their way and takes full ownership of this debacle. This is a good start. Keep it going Intel.
I'm one of those people who doesn't update their bios... frantically researching how to do this now
I updated the bios for my asus motherboard with the new microcode. Still crashes when opening a web broswer.
You will get lower scores and not much temp drop if you keep CEP enabled. If you keep CEP enabled and try of undervolt , CEP will fight the undervolt and you WILL lose performance. I have a 14900k that did not have any issues. After downloading the new mircocode, i got higher temps and high voltages. Here is what to do....Bring CPU lite load down to mode 12, disable CEP. I then did a core offset undervolt of 0.090, I also lowered PL1 to 200 and PL2 to 220 and amps to 307. With these settings my VCore is 1.310v, and temps dont exceed 80 degrees running Cinebench R23 with a arctic v3 360 cooler. If you suffer from instability at this point, increase CPU lite load up a notch til you get stable. With my settings, im stable and still get 38k in cinebench R23. Im totally happy.
Am I only the one seeing the FUTURE date in the blog ? 🤪
I installed the update on my Aorus 790 and it’s all good . I note that I did not have any problems before .
is a non k cpu affected too ? have 13700f do i need update the bios too ? i have msi b660m mortar max wifi
but i dont have problem since im buying on early 2023
Still waiting for that 0x129 code update for my board - MSI only published the 0x125 BIOS for my Z690 last Monday. With that BIOS installed, I already had to up the "Lite Load"-setting (LLC related) from 4 to now 6. Default used to be 9, now it's at 12 for some reason. 4 had been running stable for months and then I started getting a couple of weird crashes, but under "no to light" loads (like watching a video on YT), so probably some sort of problem with drops in voltage. So I upped Lite Load to 6 which seems to have made things stable again. Performance-wise (14700KF on a Z690 DDR4 mobo), most of my benchmark scores are unchanged (within margin of error) - sometimes slightly higher, sometimes slightly lower than in the before-times. Timespy seems to have generally dropped a couple of hundred points of overall, CPU and GPU score for some reason, but I really don't care if that score is at just over 19000 or at 18800. Note: I had been running near "Intel Default/Performance" settings well before any of this happened, simply because I didn't see the need to let the CPU run with no limits, get crazy hot and draw scary amounts of power for only marginal gains in certain benchmarks.
Re the whole "how do you confirm that loss in performance isn't just already degraded CPUs"? I guess Intel would have access to older examples of 13th/14th gen CPUs so they could at least check all that internally. Independent reviewers might also have access to those chips I would think (reviewers should have some sort of collection of reference chips?). Of course, in an ideal world, none of this would matter since Intel would simply RMA degraded chips and users with problems would be provided with new CPUs that will, hopefully, not degrade. I wonder how that will play out, since some folks have already started to complain about Intel stalling customers who have applied for RMA.
So the 1.55v limited is related to eTVB thing which only available for i9 huh? What about the t7 14700k?
2 different issues. There was an eTVB problem AND an over volt issue.
As per Intel regards the eTVB issue: “The microcode patch (0125) referenced in press reports fixes an eTVB bug discovered by Intel while investigating the instability reports. While this issue is potentially contributing to instability, it is not the root cause.” Confirms eTVB and over volt are 2 separate issues.
The over volt issue is fixed in the 0129 update.
And to clarify, ALL Intel 13 and 14 gen CPUs with a tdp of 65 and upwards are susceptible to damage from the over voltage issue.
We'll see in future if this really fixed the problem.
So far Gigabyte Z790 Master Aorus UEFI has not been updated yet. Gigabyte removed the F14e BIOS but the latest one has not been released yet. Also when updating the new UEFI also make sure you update the Microcode which can be found on the Drivers Site for that board in the Menu called Chipset. It is not out yet.
can i ask if i5 13500 and 14500 not include on this issue please I loved to get a response from you.,
Uh, I think I'm going to let this one marinate for a bit before I sign up
My Perfomance not affected after the bios microcode update! Works the same 31k points cinebench and gaming same
I wonder how it will impact performance in games.
What about companies that regularly push BIOS updates through Windows Update. Will
the Intel microcode come packed into these? While not ideal to update your BIOS in this way it could be beneficial for people who are no computer savvy
Not all PCS are running windows and not all windows versions are supported by microsoft anymore.
I updated to 1402 and set PL1 and pl2 to 320 and set an all core of 59 p and 45 e. With a slight under volt Get over 4000 in cinebench r23 and just updated to 1503 and setting same performance and don't go over 85 C
according to framechaser, things still not working, and it only bring more aggresive low voltage idle optimizations.
If I use Intel Default in the motherboard settings, my CPU (13600) only hits 9000 in Cinebench r23. If I use Asus Profile I get 22500-23000. The voltages etc are the same in the bios. And in hwinfo. But with intel default the CPU wont boost properly anymore, it only goes to 4.8 on pcores an 1.8-2.2 on ecores
I went with the 14700k in my 4090 build, I was debating between that and the 7800X3D and I don't fully regret it since I haven't had any issues but if Intel keeps going in this direction, AMD will get my money next time.
Nope. No difference in XTU benchmark after BIOS update. Around 7600+ marks without any overclock. Gigabyte Z790 D DDR4 (f11d BIOS) and 13600KF (2022). Was stable before update.
I would still recommend to undervolt your CPU, my i7 13700K runs a lot smoother (and scores higher in Cinebench) after I offset the core voltage by -100, thus going from the default 1.350v to 1.250v and this is both before and after the new bios update.
I also still believe that the problem got exacerbated by the likes of ASUS and especially their board settings (ASUS OC Profile) just deferring to the CPU, so if the CPU, as Intel says, requests the motherboard for extremely high voltages, then those boards/profile will just allow it and thus frying the silicon.
So I just updated to 029 version. But now I have an extra option either Intel default settings or ASUS Advanced OC Profile. When I select Intel Default Settings it give me another option after that with a window and a drop down menu buttttt the only option is performance. When this is selected (only option) it give a warning about any use of unlocked features bla bla bla may result in reduced stability. But there is no other option so is this ok ? Is it safe like this???? Someone help me. Can I now use my pc again?
Does this issue only affects the Dekstop Cpu version ?
They way I look at this : "It's not gonna fail in 3 months... Now I have entire 12 months! Yay"
I have an Asrock B760 Pro RS and installed the update. No problems so far.
Do I need to update bios for micro code 0x125 before 0x129? Or the 0x129 will automatically update the previous micro code(0x125) update alongside?
The latest one includes all the previous patches
@@imBenjamin741 thanks
Intel default settings update is still in beta for my board for over 2 months so I guess I'm going to wait unitll 2025 for stable non-beta microcode update.
I have a cheap Gigabyte B760M DDR4 Gaming motherboard paired with an i5 13600K. The board is completely underpowered for the CPU. In sythetic tests, but also when rendering, the VRM MOSFET temperature goes up to 108°C within 30 seconds, then power and performance are slashed in half, but during gaming it used to be fine. That was until I did this latest BIOS update which causes it to overheat in when gaming.