Intel breaks silence on 13th and 14th-gen Raptor Lake desktop CPU instability issues

What I feel is that this is being blown out of proportions. Only dead samples or processors having issues were showing this as the root cause.
Don’t think that’s the case at all. Lot of people on reddit and elsewhere with similar issues reported a long time back. It took way too long for Intel itself to ack this as an issue, to come out with a questionable patch with even more confusion based on their findings.

I think this issue got just the type of response and attention it deserved thanks to some of the well-known YTubers and PC builders.

FWIW, I have a connection at Intel who splurged on a new rig having these crashes even before we got to know about this being a bigger problem. Managed to tweak some voltage parameters as per the advise handed out by mobo vendors couple months back and seems to not have that issue anymore but all of this raises even more concerns w.r.t longevity.
 
I wonder if NOT having a massive AIO would reduce damage, since it would have hit thermal limits first, so sustained clocks and voltage would have been lower.
Certainly (probably) the case for me, power limited my CPU as my aio can barely handle 200w, no crashes so far, fingers crossed.
 
  • Like
Reactions: n1r0
What I feel is that this is being blown out of proportions. Only dead samples or processors having issues were showing this as the root cause.

Nah man, it isn't being blown out of proportion:

 
  • Like
Reactions: Stronk
I stopped caring about Intel CPUs when they decide to go P/E-Core non-sense because they're incompetent to expand on 10900K.
The 12400 has better single core and the 13400 has better single + multi core performance than 10900K.
If newer CPUs can do more work with fewer cores, what's wrong with it?
Initially I think there were scheduler issues where the game would run on E cores instead of P, but I thought that was fixed with Windows updates
 
  • Like
Reactions: Stronk
Nah man, it isn't being blown out of proportion:

Even that is old (May). Intel themselves have mentioned that the issue potentially extends to any CPU of 65W TDP and above.
By the way, the worst aspect is that currently there is no way to definitively identify if your CPU is impacted. By the time you find out, it's already too late.
 
amm... what it mean? any link to explain.. sorry for noob query
big.LITTLE technology. Performance cores. Efficient cores. None of these crap had existed before. Personally I think this is to save money on the production side. But the marketing sure is doing its work lol.
What I feel is that this is being blown out of proportions.
Oh not at all. If Intel really cared they could've done a full recall but no! They are barely containing the outrage and criticisms. The least they could do was ask for a thorough checking with some warranty in place till they can get the latest batch cleaned and fixed and then replace them slowly. The fact that many of the owners are still unaware of this proves that the company
I have a 13600K ... 16 months old, on a Z790 mobo .. working fine till now... no oc or ramping up stock values on cpu & ddr5 ram of 64GB ....
If you use it only for basic day to day operations then you probably won't notice it as it won't put a stress on the chip.
 
Last edited:
But that applies to every hardware, isn't it.
1000019295.jpg
 
At any rate unwise to invest in Intel right now. The AMD x3d processors are a much safer bet till this tides over. All indications point that the chips had a mfg defect and Intel is just trying workarounds to avoid a massive recall.
 
At any rate unwise to invest in Intel right now. The AMD x3d processors are a much safer bet till this tides over. All indications point that the chips had a mfg defect and Intel is just trying workarounds to avoid a massive recall.
if at all wish to have intel only, than 12th GEN is there.
 
aamir khan GIF

if at all wish to have intel only, than 12th GEN is there.
Just for fun, I might upgrade from the 14th gen to the 12th gen :facepalm:. It's no surprise that this happened since everyone is overvolting their CPUs now. Intel KS models are like hot furnaces. So far, only the Intel 14700X, 14900X, and 13900X have been affected.

Fixing the micro-code will effectively enable auto-undervolting in these CPUs if they continue running at peak speeds for extended periods.
 
But that applies to every hardware, isn't it.
In a way, sure. But this is clearly a design flaw making the chip degrade way faster than it should - which definitely does not apply to every hardware. Yes, motherboard vendors were running out of spec and feeding them extra voltage, but the oxidization issue is something it came with from the factory and would have shown its face anyway. And regarding motherboard settings, it is also on Intel for not making sure that the "default" power and voltage settings are respected - since I'm pretty sure motherboard vendors work with Intel (including for certification) before releasing a product in the market.

There are still overclocked i7 2600K CPUs running today, then why should one accept anything less from a modern counterpart? Tech is supposed to get better with time, right? Moreover, if one is spending such high prices for a processor, shouldn't they be able to use what they paid for? Not sure why you are trying to defend/justify actions of a mega corporation which clearly doesn't care about its customers (including you). Findings also showed that Intel was already aware of this issue at least a year ago and yet chose not to do anything about it.
 
Last edited:
In a way, sure. But this is clearly a design flaw making the chip degrade way faster than it should - which definitely does not apply to every hardware. Yes, motherboard vendors were running out of spec and feeding them extra voltage, but the oxidization issue is something it came with from the factory and would have shown its face anyway. And regarding motherboard settings, it is also on Intel for not making sure that the "default" power and voltage settings are respected - since I'm pretty sure motherboard vendors work with Intel (including for certification) before releasing a product in the market.

There are still overclocked i7 2600K CPUs running today, then why should one accept anything less from a modern counterpart? Tech is supposed to get better with time, right? Moreover, if one is spending such high prices for a processor, shouldn't they be able to use what they paid for? Not sure why you are trying to defend/justify actions of a mega corporation which clearly doesn't care about its customers (including you). Findings also showed that Intel was already aware of this issue at least a year ago and yet chose not to do anything about it.
I just sold my 2600K not a long while ago and loved it for almost 8 years before upgrading to 14th gen. Even that was running under-volted but I have been keeping an eye on it. If Intel knew that and did not do anything, a lawsuit is in the works already.
 
But that applies to every hardware, isn't it.
There is a difference between overclocking while remaining within safe voltage limits and exceeding it. My delidded OC'd 3770k is still running well after 14 years. 13th/14th gen couldn't even last 14 months

Fixing the micro-code will effectively enable auto-undervolting in these CPUs if they continue running at peak speeds for extended periods.
No, it will limit the maximum voltage that the CPU can request, and thereby the highest frequency it can reach.

Depending on silicon lottery, you can decrease this voltage curve to under volt it and remain stable


TLDR: there should have been 3 variables to consider: power limit + voltage limit + thermal limit
  • Mobo brands ignored sustained power limit
  • Micro code ignored safe voltage limit
  • Only thermal limit was enforced
 
  • Like
Reactions: Stronk
There is a difference between overclocking while remaining within safe voltage limits and exceeding it. My delidded OC'd 3770k is still running well after 14 years. 13th/14th gen couldn't even last 14 months


No, it will limit the maximum voltage that the CPU can request, and thereby the highest frequency it can reach.

Depending on silicon lottery, you can decrease this voltage curve to under volt it and remain stable


TLDR: there should have been 3 variables to consider: power limit + voltage limit + thermal limit
  • Mobo brands ignored sustained power limit
  • Micro code ignored safe voltage limit
  • Only thermal limit was enforced
ROFL
I realize this conversation is diverging a bit, but I wanted to share my experience. When running the P-cores on my 14700KF at anything over 5.7 GHz, games would crash and not accept the higher clock speeds. Modern Warfare 3 was the main culprit. However, everything runs fine at stock speeds. I chose the 14700KF mainly for its cores, which allow me to run many virtual machines.

After comparing it to the 14900K, I found the latter to be overly expensive. I didn't go for AMD because VMware Workstation has its own set of issues with it. Otherwise, I would have considered the 7900X3D. However, at 4K, the CPU's influence diminishes, making performance more GPU-bound.
 
Last edited:
  • Wow
Reactions: terence_fdes