Intel Has a Problem: Vast majority of Game Crashes come from 13th/14th gen Intel CPU's during Decompression

Marees said:
If there is no damage then new microcode should impact the single core max boost. That is basically bragging rights. I don't see any real loss in impacting single core max boost
Click to expand...
Came across the term Voltage Droop today

https://semiengineering.com/mitigating-voltage-droop/

Was the earlier microcode increasing voltage in advance anticipating a "voltage droop"

If yes, then does it tend to do more damage to processors in good quality motherboards that already have compensating circuitry to minimize the voltage droop ?
 
TheHig said:
To me they are trying squeeze more life out of the problematic CPUs with the least impact to performance. If they truly run safe voltages then you likely have a performance hit that opens Intel up for some additional law suits for false advertising. If I had a 13th or 14th gen cpu I’d limit it to something like 1.35v like the old days when anything over 1.4v was pushing it for daily use. Doing this you end up with a bunch of i9s only doing 5.2Ghz or so and that is way off compared to the numbers on the box.
Click to expand...
This is exactly what I'm thinking. Some of these reductions aren't even huge. So it looks like the update just takes the voltage down far enough not to kill it so soon.
 
kac77 said:
This is exactly what I'm thinking. Some of these reductions aren't even huge. So it looks like the update just takes the voltage down far enough not to kill it so soon.
Click to expand...
1.65 to 1.5 is a colossal drop and golem saw that on two i9's. Buildzoid also had one go from 1.61 to 1.55.
The chips that dropped small amounts may not even have been in the failing category in the first place.

We would need voltage info from failed chips to know how big the impact is.
 
Last edited:
TaintedSquirrel said:
1.65 to 1.5 is a colossal drop and golem saw that on two i9's. Buildzoid also had one go from 1.61 to 1.55.
The chips that dropped small amounts may not even have been in the failing category in the first place.

We would need voltage info from failed chips to know how big the impact is.
Click to expand...
Was looking at the 14700K it went from 1.55 to 1.52.
 
kac77 said:
Was looking at the 14700K it went from 1.55 to 1.52.
Click to expand...
Assuming the microcode update only caps the voltage to 1.55 and nothing else, it would have to be error margin in the sensor, inconsistency in the benchmark they were running, or the bios update itself changed something.
All the chips they tested were at the time healthy (non-degraded). From our perspective there is no way to know which, if any, were unsafe before or after.
 
TaintedSquirrel said:
Assuming the microcode update only caps the voltage to 1.55 and nothing else, it would have to be error margin in the sensor, inconsistency in the benchmark they were running, or the bios update itself changed something.
All the chips they tested were at the time healthy (non-degraded). From our perspective there is no way to know which, if any, were unsafe before or after.
Click to expand...
Or it could be that excess voltage was compensating for more voltage droop in cheap quality motherboards

So the new fix might impact performance if you cheaped out on the motherboard
 
Just updated my Bios to 1662 on the Asus Drivers website it's a beta Bios but I'm pretty sure that is updated Micro code my PC booted and looped about 12 times for 4-5 minutes. I used EZ Flash to update the Bios. Going to run Intel XTU to see if I get any more throttling then I did a few months back.
 
Last edited:
antok86 said:
anyone on ROG STRIX Z690-E GAMING WIFI BIOS 3802 out from 8/9/24
Click to expand...
Nope, I've got the F. Why does ASUS have so many models with almost the same name? The F's 3802 is now final, but when I checked a few days ago it was still beta. I'll probably get to it by the weekend.
 
1_rick said:
Nope, I've got the F. Why does ASUS have so many models with almost the same name? The F's 3802 is now final, but when I checked a few days ago it was still beta. I'll probably get to it by the weekend.
Click to expand...
Most of the board brands release a bunch of products with slight feature differences, for slightly different prices. I guess hoping to catch someone at every $30 price separation.

Seems stupid to me. I see a lot of complaints from component brands that part costs are higher than ever, squeezing their margins. And then they release way too many products. Fewer products with better feature separation, should be better for them. But they keep doing it, maybe I'm wrong.
 
chameleoneel said:
Most of the board brands release a bunch of products with slight feature differences, for slightly different prices. I guess hoping to catch someone at every $30 price separation.

Seems stupid to me. I see a lot of complaints from component brands that part costs are higher than ever, squeezing their margins. And then they release way too many products. Fewer products with better feature separation, should be better for them. But they keep doing it, maybe I'm wrong.
Click to expand...
It does feel that way with motherboards for sure. Like I get that people are going to want different features, and you'll want to have more than one product for a given CPU as you can charge a premium for some features but not everyone will pay. However they all seem to have like 15+ boards with the same basic chipset, many with only minor differences. I do have to wonder if that unnecessarily pushes costs up.
 
pendragon1 said:
taking the weekend out, you should hear back tomorrow.
Click to expand...
" Tomorrow" came and went. Then the next day came and went and so on and so on. They escalated it yet again and then that deadline also came and went. That is three times they've given a timeframe which meant absolutely nothing. And then as of today they finally admitted that they have no idea when I will actually get my refund as they have not even had time to validate it.
 
jobert said:
" Tomorrow" came and went. Then the next day came and went and so on and so on. They escalated it yet again and then that deadline also came and went. That is three times they've given a timeframe which meant absolutely nothing. And then as of today they finally admitted that they have no idea when I will actually get my refund as they have not even had time to validate it.
Click to expand...
Mine has been in validation hell for over a week
 
Keljian said:
Mine has been in validation hell for over a week
Click to expand...
I am well into my third week now that they have had possession of my CPU. And heck it was almost 2 weeks of nonsense back and forth before I even was able to get the RMA properly approved in the first place. So this ordeal is well over a month so far.
 
If they're fumbling their RMA process then maybe they should rush the new 12th gen all P-core CPUs as replacements.
 
jobert said:
I am well into my third week now that they have had possession of my CPU. And heck it was almost 2 weeks of nonsense back and forth before I even was able to get the RMA properly approved in the first place. So this ordeal is well over a month so far.
Click to expand...
They don’t have my chip yet…
 
1_rick said:
Nope, I've got the F. Why does ASUS have so many models with almost the same name? The F's 3802 is now final, but when I checked a few days ago it was still beta. I'll probably get to it by the weekend.
Click to expand...
Anything change from beta to final? Keeps the same rev number?
 
antok86 said:
Anything change from beta to final? Keeps the same rev number?
Click to expand...
It's still 3802, and I don't think anything is different unless maybe the ME version number changed, but there's no way to know. I think I've seen this with Gigabyte, too: when they update beta bioses to newer beta versions, or to the final, they get rid of the older listings. It just goes from 3701 to 3802 in this case, and the 3802 beta listing is gone.
 
1_rick said:
It's still 3802, and I don't think anything is different unless maybe the ME version number changed, but there's no way to know. I think I've seen this with Gigabyte, too: when they update beta bioses to newer beta versions, or to the final, they get rid of the older listings. It just goes from 3701 to 3802 in this case, and the 3802 beta listing is gone.
Click to expand...
On asus site it’s bios “fix” says for non-k which is weird

updated and verified bios is on the same version. now i wait for bsod. lol

this is basically sitting at idle with browser open after bios update
i9 13900K ASUS ROG Strix Z690-E
 

Attachments

  • Capture.PNG
    Capture.PNG
    115.7 KB · Views: 0
  • Capture.PNG
    Capture.PNG
    41.7 KB · Views: 0
Last edited:
And yet another week goes by and still no refund and actually no communication. But I'm still not getting jacked around according to some people?
 
pendragon1 said:
noones said shit to you since i said "tomorrow", which it should have been, under normal circumstances. yeah it sucks, what ya gonna do? rage at us? go blow up their twatter or something...
Click to expand...
And did I say anyone has said anything since then? I'm just referring back to when you and especially one other person seem to act like things were going like they should. I already knew at that time that it felt like I was getting jacked around and was only going to get worse, which it has. I'm not raging against anybody, just pointing out that my gut feeling was correct from the beginning. I've even tried calling Intel and asking for a call back and sending emails and have absolutely got no communication for the last week. So yes, I'm venting here a little bit and I'm sure some other people will too once they realize they're also getting jacked around. If I would have had a good experience then I would have also been posting about that.
 
jobert said:
And did I say anyone has said anything since then? I'm just referring back to when you and especially one other person seem to act like things were going like they should.
Click to expand...
at the time it was within their normal response time, whether you want to accept that or not. shit also did hit the fan for them....
 
pendragon1 said:
noones said shit to you since i said "tomorrow", which it should have been, under normal circumstances. yeah it sucks, what ya gonna do? rage at us? go blow up their twatter or something...
Click to expand...
If anyone works professionally with Intel products, moving forward point their said enterprise at AMD, ARM, RISC-V, and not Intel.
It's probably going to happen slowly at best, but one can dream.
 
pendragon1 said:
at the time it was within their normal response time, whether you want to accept that or not. shit also did hit the fan for them....
Click to expand...
Well about 3 hours after your post Intel gave me a call with all the information I needed to get my refund at Western Union. And they actually gave me the full refund for the price I paid instead of the prorated amount they originally said they were going to give me. I had pleaded my case saying based on all the circumstances that they should be giving everyone a full refund for having to deal with all this hassle. I mean having to go get another CPU and spend an hour and a half of my time for a slightly slower computer and dealing with the weeks and weeks of back and forth that a full refund is the minimum they should be doing. So to anyone that speaks to Intel make sure you get the full refund as they can certainly do it.
 
I'm just glad that I am all but an AMD fanboy for my home builds. No Intel for me,except in laptops.
 
Intel is sharing a few important updates on the Intel® Core™ 13th and 14th Gen desktop processor Vmin Shift Instability issue investigation, including ongoing guidance for BIOS updates and settings and the status of upcoming next gen product families. Intel will be publishing another update by the end of September.

Future Product Update
Intel confirms that its next generation of processors, codenamed Arrow Lake and Lunar Lake, are not affected by the Vmin Shift Instability issue due to the new architectures powering both product families. Intel will ensure future product families are protected against the Vmin Shift Instability issue as well.

Unaffected Products List
Following the recent warranty extension announcement for affected Intel Core 13th and 14th Gen desktop processors, Intel confirms these currently available processors are not affected by the Vmin Shift Instability issue:
  • 12th Gen Intel Core desktop and mobile processors
  • Intel Core 13th and 14th Gen i5 (non-K) & i3 desktop processors
  • Intel Core 13th and 14th Gen mobile processors – including HX-series processors.
  • Intel Xeon processors – including server and workstation processors.
  • Intel Core Ultra (Series 1) processors

Intel Core 13th and 14th Gen Desktop Processor BIOS Updates
While most Intel Core 13th and 14th Gen desktop processors are not impacted by the Vmin Shift Instability issue, Intel recommends all users continue following guidance:
  • Ensure the system is running with the latest BIOS, which users can look up through Intel’s Compatibility Tool and/or their motherboard manufacturer’s website. Users can also learn more about how to update their BIOS by visiting the following site: How to Update BIOS.
  • Utilizing the Intel Default Settings recommendations for their Intel Core 13th and/or 14th Gen desktop processor – including both Intel Core 13th and 14th Gen consumer, commercial, and entry workstation desktop systems.

https://community.intel.com/t5/Proc...re-Products/m-p/1627440/highlight/true#M77071
 
I have a hard time believing the non-K 13900 and 14900 are safe.
If 13600K's are dying at 5.1 GHz then the i9's at 5.8+ are probably at risk too.
 
If i update my mobo to latest bios it defaults settings right?(presumably safe and "fixed") i have a second hand 13900k does this no longer have warranty?
 
antok86 said:
If i update my mobo to latest bios it defaults settings right?(presumably safe and "fixed") i have a second hand 13900k does this no longer have warranty?
Click to expand...
Update the bios for sure.

The settings, no idea if those will change or not, but the problem was in the microcode supplying too much voltage to a few cores. All of the chasing down "Intel Default" blah blah was I believe more to prevent damage until the microcode fix arrived, and not necessarily in itself the cause of the damage. We already know that CPU's running the default (non OC) profiles also had this issue. Intel was blaming everyone but themselves initially. Intel alone is responsible for the microcode.

After updating, go in and check what the settings are set to. The "Performance" profile should be fine, is what I have gathered from GN videos. But it's probably a bit too early to tell for certain one way or the other. Give it 6 months.
 
TaintedSquirrel said:
I have a hard time believing the non-K 13900 and 14900 are safe.
If 13600K's are dying at 5.1 GHz then the i9's at 5.8+ are probably at risk too.
Click to expand...


I mean, I personally have a difficult time believing anything coming out of Intel at this point.

They aren't exactly the most transparent and honest company to begin with, and now they are in full on "survival mode".

I'll consider everything they say questionable until it has been objectively confirmed by an outside neutral party.

In other words, I'm not touching anything 13th or 14th gen no matter what they say, and I'm probably avoiding 15th as well.
 
You must log in or register to reply here.
Back
Top