I’m still looking for definite recommendations, but as far as I currently understand it I’m not sure that a simple OS patch solves the vulnerability. A firmware/microcode update may be necessary via a BIOS update or whatever, and that could make additional performance losses happen.
However I don’t know the result of installing the Win 10 update manually from the Microsoft Catalog page. My Cumulative Win 10 KB4056892 was installed automatically on my Win 10 Creators Edition with automatic updates. If you’ve turned off updates completely, you’d probably want the Delta edition from the Microsoft Catalog page but I’m not really sure, maybe someone else does.
MS has issued an official statement on it. It affects older AMD CPUs as far as I know, not the latest generation. MS shouldn’t patch for AMD CPUs to begin with until it’s known they’ve been “hacked” as well.
Well, so far I haven’t really seen any reports of decreased performance by DAW users, with the exception of someone essentially getting a broken system after an update. But that’s a bit of a different issue I think.
Anyway, I doubt they feel it necessary to spend the time to test it right now, and I’d be inclined to agree with that. It’d be more valuable if a company such as Scan audio ran benchmarks again since they’ve run through DAWbench on a bunch of different CPUs and could then provide a more specific picture of any degradation.
It’s to early to predict, all I can say is I have not run into performance problems after the patch, and the cpu utilization looks the same. But none of the projects I work on right now are that big that it would matter.
Time will tell, I’m holding off buying a new computer right now till there are more facts/benchmarks on the table.
Reading the latest details in the German Heise.de website, performance degradation is already mentioned to be @20% !
Where forum users report even with Xeon v6 CPU -40%!
The performance degradation is all about the specific task you are using your computer for. Many Xeon CPUs sits in servers that have to read large databases and such, those will be affected quite a bit. Normal desktop applications such as gaming, DAW, office and such will not suffer any noticeable difference, at least with a newer CPU generation from Skylake and forward.
The DAW workload with high track counts is similar to the database workload, which is one with a high number of IO syscalls. The mitigating factor is that, in 2018, a SSD “crippled” by the Intel chip workarounds can still easily read data at a rate required by the DAW for even the largest track counts. I think at this point we can safely say that the workarounds’ performance degradation, in audio environments where its seen, is more of an annoyance than anything else. No one wants to see SSD performance cut by 30%, even if it’s still more than enough for any DAW workload.
I have to disagree with you on that. DAW workload is no near the workload of accessing a database. This is also why most users in this forum says they cannot see any noticeable performance degradation.
For the SSD, yes you are partially correct, but not with a typically DAW workload. Microsoft's 'Meltdown' Patch Has Little Impact On Storage Application Performance | Tom's Hardware
As they also say, the largest performance degradation is with random write performance and that’s not a typically DAW workload, as when you record it’s sequential write, not random write.
When you scroll down a little more, you will see some real world bench with gaming, office and media creation all showing no difference at all.
I agree it’s annoying to know that your hardware performe worse with the update, but for normal usage the difference is too small to notice.
As someone who works in both fields, I can tell you categorically this is not the case. If you have any evidence to support your claim, I’d love to see it.