I recently came across this and a whole slew of more in-depth, related videos that would seem, on the surface, to explain a lot of my random Cubase crashes, of which I experienced two of the “a serious problem” variety in a row yesterday: https://www.youtube.com/watch?v=H8LrwI-I_fY
Is Steinberg aware of this? Can they rightly point the finger squarely at Intel, and to what extent? As the title would imply, I am running an i9-13900K and I have had nothing but trouble with this build when it comes to trying to get work done with Cubase 12 and 13 under it.
A statement from Steinberg would be most welcome. I’d do damn near anything to alleviate this instability.
By virtue of the vid you posted, going back to gen 12 would be your quickest fix. I’m here on the forum because I was getting ready to upgrade to a 14th gen build but with this update report, I’ll sit back and watch the story unfold first. Good luck on finding a more suitable fix for you problem!
I’m not related to Steinberg or Intel, but I do know quite a bit about Intel and AMD products! (it’s been my day job for over 20yrs)
Without opening another long and boring debate, I can assure you this video is 99% b.s. It’s pure Intel-bashing in its ugliest form.
I also can add my personal experience, as I personally own a 13th gen CPU and it runs just fine, just like millions of other happy users.
A 13th gen+ CPU will be more efficient running Win11, and that’s normal as it introduced architectural features (e-cores mainly) that work best with proper OS support.
It’s not entirely impossible that you happen to have a bad CPU unit, but it’s 10x more probable that you have a bad motherboard, and 10000x more probable that you have a bad SW installation / configuration.
Start by opening Task Manager and cleaning up your machine of useless bloatware and processes lurking in the background.
13700k here has been fantastic, performance wise. I don’t seem to see any E-core related issues.
I get my fair share of crashes but they’re plugin related and don’t seem performance related at all.
I do know that the 13900k’s run very very hot (why I opted for the 13700k), so I’d start by checking your cooling. But more likely it’s a problematic plugin.
I don’t think so: there are many other videos on the same topic and Intel has acknowledged the problem. Apparently, it only really severely manifests in the 13900 and 14900. Note the *900. All the others are supposed to be not nearly as afflicted with this.
And I’ve had two different motherboards in this system with similar results, so that’s not it.
You may note from the many, many other threads I’ve started that I’m a bit of a power user, so bloatware & the like is almost certainly not the issue.
I think Intel may really have a problem on their hand with just the highest end models alone, and they’re trying to find out what’s going on themselves.
Thanks, but my temps are fantastic: I’m using a high end liquid cooling system.
I’m not saying my issues with Cubase come down entirely to this, but I’m genuinely curious if it could be a factor with certain types of crashes I’ve seen.
I would lean more towards an issue with the CPU. There is simply too much online bad press for this to be ignored, from various reputable tech companies and from many consumers. It is also effecting many Server boards.
A key takeaway from the XDA website is this:
" While Intel acknowledged the issue and released a fix for the eTVB bug, the root cause of the stability problems remains unresolved."
Still, this is no guarantee that you have a defective CPU. But if all else fails.
Appreciate it. Yeah, it’s been a really, really bumpy ride with this build. On many levels. The P-Core vs. E-Core thing alone has been driving me mad: I can open the same project and the load distribution can be different each time. If I get unlucky and something ends up on an E-Core just 'cause that probably shouldn’t, I get spiking till I roll the dice and reload the same project.
I’ll reiterate that I strongly believe that an industry lobbying arm is important at this point. Genuine expertise on what real time audio applications need to deliver audio in real time on modern computers needs to be pushed and pushed hard so Intel, Microsoft, NVIDIA, and others get the message and stop engaging in practices that screw everything up for creative professionals because they either don’t know better or nobody’s putting sufficient pressure on them to care enough.
It’s a major drag that so many of the same issues from the early 2000s haven’t gotten any better over twenty years later, and in some ways some of them have even gotten worse as processor architectures have changed so radically (like the different core types issues).
But again, at least DSP quality has gotten phenomenal. Now, if I could only use it reliably!
Interesting article here also, specifically regarding the 13900K and 14900K.
In summary: “This discovery confirms that Intel’s stability problems with Raptor Lake and Raptor Lake Refresh are more complicated than ever. The server-based motherboards used by these 13900K and 14900K servers are focused entirely on stability and running the chips within specifications, with no way to overclock these chips.”
It seems to be related to high loads causing thermal problems. I do not think Cubase is able to utilize all cpus to that extent that it will trigger that problem, it will have big amount of audio drop outs and be unusable for that reason. However it also seem to be some claims that it will be permanent damages with the temperature issue so a computer that is also used for gaming might have this damage.
If intel can’t fix this with firmware it will be very costly so even if the know the root problem they might not confirm the issue until they have solution.
They say “we have determined that elevated operating voltage is causing instability issues in some 13th/14th Gen desktop processors […] Intel is delivering a microcode patch which addresses the root cause […] currently targeting mid-August for patch release”.
Yeah, and there’s even mention of via oxidation in some of the early production waves. Great. Looks like they’ve got a serious, serious mess on their hands here. Good grief.
Intel Core 13th and 14th Generation desktop processors with 65W or higher base power, including K/KF/KS and 65W non-K variants could be affected and if you have a CPU already showing instability symptoms, the damage is permanent and the CPU will have to be replaced.
I just updated to the latest BIOS supposedly with the new microcode from Intel and Cubase seems even crashier now.
If this turns out to be all Intel’s fault from the get-go, I’m going to blow a gasket. This system has been nothing but trouble for audio since I built it, and I’m starting to strongly suspect I’m about to find out exactly why.
This is an epic, historical zoinc up from Intel, already. Let’s see how much worse it can get.
Hello,
Intel chips are having stability issues.
You would be better off with the newer Amd ryzen 7900x or 7700x.
I’m using the 7900x with a Asrock B650E Taichi. C12.0.70 runs great and I’m now using C13.0.40 pro runs good, having small intermittent graphic lag when switching to the mixer and realtime graphic on eq. Not sure the workaround but zero crash.
Well, I would like to, but I spent over $1200 for this CPU + motherboard alone, so I want to get my money’s worth and have Intel make this right. I can’t just go and blow another grand or so on a competitor’s platform just because someone on the Internet told me I’d be better off, no offense.
Maybe I should’ve just switched to Mac from the start. What a nightmare this has all been. And I don’t like Macs all that much, personally: they have their own set of issues.
I think the best way to save your 12/13/14th gen Intel is to monitor VID requests and actual Vcore values… As for now there is unofficial internal info that Intel will limit CPU VID requests to 1.55v max in upcoming microcode patch next month… So the best way for now is to check BIOS have PL1, PL2 and ICCMAX values set within Intel spec and maybe undervolt CPU - make some negative offset to VID request table and check if CPU is stable on those settings… There are a lot 14900K which have 1.5v at 6GHz boost fused VIDs and even 14900KS with 1.55v at 6.2GHz stock boosts… So if you have such CPU and for example your load line calibration with SVID behavior are exaggerated by motherboards VRM settings - you will see voltages especially on single core workloads above 1.55v which can lead to degradation… So for example I have 13900KS with Asus Strix Z790E mobo and I undervolted it and I never see VID requests or actual Vcore spiking more then 1.45v which could be I guess safe region for this CPU… Other variant is to limit ratio and boost multipliers to around 5.2GHz (P) 4.2Ghz (E) and set power limit to around 200w with ICCMAX at around 300A - this deforced profile pulls around 70C and 37K points in Cinebench R23 multi-core - so its not a huge decrease from stock performance…
I think the main issue is that not a lot of people are monitoring VIDs and Vcore on their PCs so at default BIOS settings it could actually be very high voltages straight from the ‘‘defaults’’… This is Intel and motherboard vendor mess and they should be responsible for this huge fail… At least they should RMA and exchange CPUs which are degraded already… And overall this all is not a proper and pleasant experience for the end users who spent thousands of euros for their PC builds… Total failure…
I know there are a lot of bad things told about how awful is Intel with their hybrid architecture and how awful is Windows 11 for music production… I spend quite a lot of time to optimize my 13900KS and Win 11 for music production, and I guess now it works fine in Cubase 13. Interesting why such a thing could not be straight out of the box experience… But we need to mess around with bunch of OS and BIOS settings to get such a low DPC latencies and stable audio performance…
can’t just go and blow another grand or so on a competitor’s platform just because someone on the Internet told me I’d be better off, no offense.
If I were you, I’d buy a 13700kf for $299 on Amazon (Free returns), swap out your 13900k, and see if your issues persist. Then at least you be sure its the CPU and not a bad RAM stick or something.
