News Raptor Lake instability hits streamer Asmongold — Black Myth: Wukong shader compilation fails on stream

[Admin]

Well-known Twitch streamer Asmongold experiences Intel Raptor Lake instability issues as Black Myth: Wukong shader compilation process fails on stream.

Raptor Lake instability hits streamer Asmongold — Black Myth: Wukong shader compilation fails on stream : Read more

 

[usertests]

https://www.google.com/search?q=Asmongold+roach

 

[-Fran-]

One thing is reading reports pointing at there being a problem (I've read two already, plus the Steam forums message from the Monkey Game Devs), but another is watching this on one of the streamers with the most amount of viewers across all platforms. Oof Intel, oof.

Keep in mind he uses one of this own company's PCs. Starforge PCs. So he is using a pre-built, technically speaking. No idea if he mentioned he's using latest BIOS and all that.

Regards.

 

[Mattzun]

Curious about the claim that in the article that the new Intel microcode has patched the instability issues.

I thought that Intel is just claiming that the new microcode prevents further damage and that the extended warranty is available to help people RMA CPUs that have already suffered enough damage to cause crashes.

The workarounds for a PC with the stability issue seem to be the ones documented by the gaming companies - downclock memory and CPU etc.

We are likely to run into another batch of weird failures every time a new game comes out and does shader compilation. Its too bad that this may cause companies to ignore actual bugs until someone on AMD or 12th gen reports the same issue.

 

[bill001g]

I suspect he partially did this for his viewers. He owns a large part of starforge so there many senior techs that would likely run over to his house to help him. Would turn out to be good advertising if he got it fixed under the starforge warranty.

 

[-Fran-]

Curious about the claim that in the article that the new Intel microcode has patched the instability issues.

I thought that Intel is just claiming that the new microcode prevents further damage and that the extended warranty is available to help people RMA CPUs that have already suffered enough damage to cause crashes.

The workarounds for a PC with the stability issue seem to be the ones documented by the gaming companies - downclock memory and CPU etc.

We are likely to run into another batch of weird failures every time a new game comes out and does shader compilation. Its too bad that this may cause companies to ignore actual bugs until someone on AMD or 12th gen reports the same issue.

According to this: https://community.intel.com/t5/Proc...rts-on-Intel-Core-13th-and/m-p/1617113#M74792

It very explicitly says: "we found the root cause and we're patching it". Is there a second reading I'm missing?

Regards.

 

[Mattzun]

According to this: https://community.intel.com/t5/Proc...rts-on-Intel-Core-13th-and/m-p/1617113#M74792

It very explicitly says: "we found the root cause and we're patching it". Is there a second reading I'm missing?

That would prevent further damage, but doesn't help with CPUs that are already experiencing instability or are close to failure. People buying a new 13th/14th gen or getting an RMA replacement and updating their BIOS now are likely to be OK.

The next line and the warranty extension is the fix for people who already have stability issues or where they show up in the next two years (i.e. RMA the CPU)

Intel is committed to making sure all customers who have or are currently experiencing instability symptoms on their 13th and/or 14th Gen desktop processors are supported in the exchange process.

To help streamline the support process, Intel's guidance is as follows:

• For users who purchased 13th/14th Gen-powered desktop systems from OEM/System Integrator - please reach out to your system vendor's customer support team for further assistance.
• For users who purchased boxed/tray 13th/14th Gen desktop processors - please reach out to Intel Customer Support for further assistance

 

[Eximo]

According to this: https://community.intel.com/t5/Proc...rts-on-Intel-Core-13th-and/m-p/1617113#M74792

It very explicitly says: "we found the root cause and we're patching it". Is there a second reading I'm missing?

Regards.

He has it correct. Unless that is a brand new CPU and had the microcode applied to it before much use, it may already have damage that this game exacerbates.

The fix also has to be applied, and there will be people out there who miss it. And they will likely need a CPU replacement, but may not know they can get one.

Already seen a few posts where people have this problem, already took the computer to a shop and were told they had bad memory and other things without having knowledge of the ongoing issue.

 

[-Fran-]

That would prevent further damage, but doesn't help with CPUs that are already experiencing instability or are close to failure. People buying a new 13th/14th gen or getting an RMA replacement and updating their BIOS now are likely to be OK.

The next line and the warranty extension is the fix for people who already have stability issues or where they show up in the next two years (i.e. RMA the CPU)

Intel is committed to making sure all customers who have or are currently experiencing instability symptoms on their 13th and/or 14th Gen desktop processors are supported in the exchange process.

To help streamline the support process, Intel's guidance is as follows:

• For users who purchased 13th/14th Gen-powered desktop systems from OEM/System Integrator - please reach out to your system vendor's customer support team for further assistance.
• For users who purchased boxed/tray 13th/14th Gen desktop processors - please reach out to Intel Customer Support for further assistance

So it's a matter of interpretation.

If your car brakes fail and you bump into things or have tire issues due to it, you still need to fix the brakes, but the other damage will still be there.

Call it whatever you want and I agree with the "if it's already borked, you better replace it" for the record, but Intel's wording makes it so that anyone who has a "problem free" RaptorLake after this patch (0x129), it should be ok going forward, which is what the article is refering to. At least that is how I interpert the wording.

Thing is, it seems like it is not.

Regards.

 

[rluker5]

I can replicate that problem on UE5 games. I just decrease voltage to an unstable one under load. To fix it I increase my voltage under load. The First Descendant took 5 More mv than SOTTR for stability.

There is really a lot of vdroop in most stock motherboard settings with i9s. Trying to fix it by increasing the voltage leads to too high of idle voltage, but fixing it by reducing vdroop via LLC settings and undervolting back to a bit higher full load voltage results in lower low use/single core voltages and more stability in high use scenarios.

 

[UnforcedERROR]

I'd have rather a truck had hit him, but this is a tech site so the article tracks I suppose.

 

[jlake3]

Intel has already patched Raptor Lake's instability issues through a new microcode update, which not only rectifies CPU instability but also addresses CPU degradation issues caused by excessively high voltages.

Sadly, however, Intel's microcode update is unable to fix Raptor Lake CPUs that have already degraded. These chips will still see instability issues even with the new microcode installed.

Is it just me, or do these two sentences effectively contradict each other?

The first one seems to imply that the microcode update rectifies CPU instability in addition to preventing degradation, while the second one states that the microcode does NOT fix CPUs that have already degraded to the point of having instability issues.

 

[ThisIsMe]

I still say it’s weird that this only happens with nVidia cards in games that use the latest Unreal Engine update that released a few months back when compiling DX12 shaders. These devs like to point fingers and wine and preemptively give notice about the errors, but none have done any troubleshooting at all. Or, at least none that they have posted about. Maybe start by disabling shaders in groups then test to narrow it down to see if there is a specific one or two that cause this issue?

Look, at this point (if it’s truly so rampant and widespread) I don’t see why they haven’t done this in an attempt to at least let their games run on these systems without errors or less errors. Or, maybe just give players the chance by adding options to disable specific shaders on their own and test for themselves.

This whole thing seems so weird. People are just going along with being okay that something doesn’t work right because it’s the other guy’s fault.

 

[Mattzun]

I still say it’s weird that this only happens with nVidia cards in games that use the latest Unreal Engine update that released a few months back when compiling DX12 shaders. These devs like to point fingers and wine and preemptively give notice about the errors, but none have done any troubleshooting at all. Or, at least none that they have posted about. Maybe start by disabling shaders in groups then test to narrow it down to see if there is a specific one or two that cause this issue?

Look, at this point (if it’s truly so rampant and widespread) I don’t see why they haven’t done this in an attempt to at least let their games run on these systems without errors or less errors. Or, maybe just give players the chance by adding options to disable specific shaders on their own and test for themselves.

This whole thing seems so weird. People are just going along with being okay that something does work right because it’s the other guy’s fault.

I don't know how widespread the issue is - there are enough people with issues for the dev to pin a post on the problem, but there are a LOT of people playing. They may just be trying to avoid support incident for something they can't fix.

The symptoms of the Intel degradation problem are unusual - its impossible random changes to data that can crash shader compilation or cause you to be identified as a hacker in anti cheat code.

The compilation code for shaders on NVidia cards isn't bad, it is just different enough from AMD's approach to trigger the crash. Its why NVidia posted a blame Intel article of its own - people thought that there were driver bugs or failing GPUs.

The problem doesn't show up when a system is first built and it gets worse over time.

When it first appears on a CPU, the system will work fine in most stress tests (Cinebench etc) and most games but causes odd crashes in shader compilation or NVidia driver installation.
As the degradation continues , the system is more and more unstable until it eventually fails to POST.

The workaround of reducing the CPU and memory frequency have a similar effect to backing off on an unstable overclock. Things start working again, at least until the system degrade more.

 

[tamalero]

I can replicate that problem on UE5 games. I just decrease voltage to an unstable one under load. To fix it I increase my voltage under load. The First Descendant took 5 More mv than SOTTR for stability.

There is really a lot of vdroop in most stock motherboard settings with i9s. Trying to fix it by increasing the voltage leads to too high of idle voltage, but fixing it by reducing vdroop via LLC settings and undervolting back to a bit higher full load voltage results in lower low use/single core voltages and more stability in high use scenarios.

You just can't keep increasing the voltage. IF the chip was degraded for excessive voltage.. you're just making things worse.

 

[Guardians Bane]

S

I suspect he partially did this for his viewers. He owns a large part of starforge so there many senior techs that would likely run over to his house to help him. Would turn out to be good advertising if he got it fixed under the starforge warranty.

Am I understanding you right in that you're suggesting this was basically a publicity stunt to gain more sales and better brand recognition?

 

[Guardians Bane]

Is it just me, or do these two sentences effectively contradict each other?

The first one seems to imply that the microcode update rectifies CPU instability in addition to preventing degradation, while the second one states that the microcode does NOT fix CPUs that have already degraded to the point of having instability issues.

There are CPUs out there that already have degradation issues. But not necessarily to the extent that it's noticable yet. The code attempts to stop that degradation from happening further (at the accelerated rate the the main issue caused)

Thing is, all CPUs slooooowly degrade and loose performance.

Scenario:

gamer has degrading CPU but hasn't had any problems yet. But little does he know it's right on the edge of failures.

Gamer applies microcode. Thinks everything will be fine! Game on!!!
The whe time he feels safe thinking it's fixed! But since it was already at the edge of failing, and all CPUs slowly get worse, the CPU goes over the the edge.

Now they use the extended warranty to RMA the CPU and now he's once again a happy gamer!

I don't know if that helps or not.

 

[Guardians Bane]

I still say it’s weird that this only happens with nVidia cards in games that use the latest Unreal Engine update that released a few months back when compiling DX12 shaders. These devs like to point fingers and wine and preemptively give notice about the errors, but none have done any troubleshooting at all. Or, at least none that they have posted about. Maybe start by disabling shaders in groups then test to narrow it down to see if there is a specific one or two that cause this issue?

Look, at this point (if it’s truly so rampant and widespread) I don’t see why they haven’t done this in an attempt to at least let their games run on these systems without errors or less errors. Or, maybe just give players the chance by adding options to disable specific shaders on their own and test for themselves.

This whole thing seems so weird. People are just going along with being okay that something does work right because it’s the other guy’s fault.

I hear what you're saying. But for the vast majority of people with these CPUs, they don't have the tech expertise to know what's going on or how to fix it.

As consumers the CPU should be able to work out of the box with little to no tweaking.

I'm taking about pre-builds. If someone is DIYing a PC for the first time, they also might not have a lot of tech-spertise as well. But anyone diving into this hobby will need to learn how to do basic BIOS work. But all the other stuff will come with time. They may not understand what you're talking about about with shaders and such. That being a GPU thing.

As consumers, we should be able to troubleshoot and such. But the more complicated it gets with the CPU issues Intel's been having, makes it a lot harder for regular people to be able to find or learn what the issues are and how to rectify them.

 

[Mattzun]

Am I understanding you right in that you're suggesting this was basically a publicity stunt to gain more sales and better brand recognition?

Setting this up does seem like something Asmongold might do for views and he could easily get a system with a known bad CPU. Seeing the issue live in a stream provides more visibility than a dozen tech channels covering it.

That said, the CPU is definitely bad. He responded to Steve from Gamers Nexus asking if he could test the chip and said he was getting a replacement CPU and that he would send the chip for testing

That thread on Gamers Nexus channel on X has other people noting that they had the same issue and that Intel was accepting RMAs based on the error.

 

[Guardians Bane]

Setting this up does seem like something Asmongold might do for views and he could easily get a system with a known bad CPU. Seeing the issue live in a stream provides more visibility than a dozen tech channels covering it.

That said, the CPU is definitely bad. He responded to Steve from Gamers Nexus asking if he could test the chip and said he was getting a replacement CPU and that he would send the chip for testing

That thread on Gamers Nexus channel on X has other people noting that they had the same issue and that Intel was accepting RMAs based on the error

I agree, I wasn't sure I understood the comment correctly. He really is a very peculiar person. And I wouldn't put it past him to do that. He is very strange indeed.

 

[adamboy64]

I'm guessing it didn't sour his perception of the game much, he still gave it like a 9.5 / 10 from his early impressions.

 

[MXM0]

Callin' BS. Overvoltage damaged CPUs don't wait to conveniently fail on shader compilation in a brand new game, and also are so consistent about it.

Not the CPU. Try again. I might believe it if it was a thing on other game installs, on the same hardware, that aren't brand new.

 

[baboma]

>As consumers the CPU should be able to work out of the box with little to no tweaking.

That's true for pre-builts. But DIY PC, per its name, is all about tweaking.

The whole premise of Intel's K series is overclocking (ie, tweaking). Sites like THW relentlessly promotes pushing the power envelope with unlimited power defaults. Using it OOB means overclocking.

No doubt that RPL instability issue is a black eye for Intel. But OC'ing always has its downsides in system crashing & instability. Every OC'er knows this, or should know this.

CPU degradation (via OC'ing) is not new. I had OC'ed two of the infamous Celeron 300A. They ran fine for a while, but over time--about a year--I started getting crashes on both. I ended up having to disable OC altogether.


>But the more complicated it gets with the CPU issues Intel's been having, makes it a lot harder for regular people to be able to find or learn what the issues are and how to rectify them.

I always stick to KISS when troubleshooting. It's hard to ascertain whether a Wukong crash is due to the RPL issue, but you can keep it simple and just RMA the CPU (if RPL), and remove that from consideration. It doesn't need to be complicated.

Regardless of RPL issue, the game Wukong itself is buggy and crash-prone, per the Steam thread and from many reviews. Crashes aren't limited to just RPL PCs. This is unfortunately the norm with graphically ambitious games. Just as with CP2077, Wukong will need multiple post-launch patches.

 

[MXM0]

I still say it’s weird that this only happens with nVidia cards in games that use the latest Unreal Engine update that released a few months back when compiling DX12 shaders. These devs like to point fingers and wine and preemptively give notice about the errors, but none have done any troubleshooting at all. Or, at least none that they have posted about. Maybe start by disabling shaders in groups then test to narrow it down to see if there is a specific one or two that cause this issue?

Look, at this point (if it’s truly so rampant and widespread) I don’t see why they haven’t done this in an attempt to at least let their games run on these systems without errors or less errors. Or, maybe just give players the chance by adding options to disable specific shaders on their own and test for themselves.

This whole thing seems so weird. People are just going along with being okay that something does work right because it’s the other guy’s fault.

No, you're right. CPUs don't fail in such a predictable manner, especially when failing from voltage issues.

 

[Mattzun]

No, you're right. CPUs don't fail in such a predictable manner, especially when failing from voltage issues.

You are incorrect - this is a completely predictable failure based on the CPU needing more and more voltage over time.

NVidia and multiple development teams have spent a lot of time investigating the issue and eventually confirmed that it is an Intel problem.
That is really easy to demonstrate if you have a degraded CPU and working CPU of the same model.
If you can reproduce crashes with the degraded CPU and not on the good CPU, you don't have a software or driver bug.

Certain workflows (shader compilation, compression/decompression) seem especially sensitive to a 13th/14th gen CPU that doesn't have enough voltage.

People have claimed to be able to cause the same error when intentionally undervolting their perfectly functional CPU.

Fortunately, Intel seems to have informed their customer support to accept RMAs if someone complains about a shader compilation crash.