Any performance degradation in Reason on pre-Haswell Windows machines with new MS patch?

This forum is for discussing Reason. Questions, answers, ideas, and opinions... all apply.
househoppin09
Posts: 536
Joined: 03 Aug 2016

04 Jan 2018

Microsoft has just pushed out a patch (KB4056892) for Windows 10 to protect against the Meltdown vulnerability. This has the potential to impact performance on pre-Haswell CPUs (i.e., mostly Sandy Bridge and Ivy Bridge, check Wikipedia to see which generation yours is). I've got some projects going that won't be manageable with even a slight performance degradation, so I would really like to see some reports from Win10 users who have the patch installed on pre-Haswell machines. Are you seeing any performance impact in Reason? Getting more audio crackles/breakups, more "computer too slow" messages, etc.? Probably the best way to test is to play a song that pushed your system just about to its limits before and see how it plays now. Anyone? :)

EDIT: Reports from Win7/Win8.1 users would also be quite useful, of course! The equivalent patches for those versions of Windows should be coming out on Tuesday, January 9th.
Last edited by househoppin09 on 04 Jan 2018, edited 3 times in total.

User avatar
EnochLight
Moderator
Posts: 8405
Joined: 17 Jan 2015
Location: Imladris

04 Jan 2018

househoppin09 wrote:
04 Jan 2018
Microsoft has just pushed out a patch (KB4056892) for Windows 10 to protect against the Meltdown vulnerability. This has the potential to impact performance on pre-Haswell CPUs (i.e., mostly Sandy Bridge and Ivy Bridge, check Wikipedia to see which generation yours is). I've got some projects going that won't be manageable with even a slight performance degradation, so I would really like to see some reports from Win10 users who have the patch installed on pre-Haswell machines. Are you seeing any performance impact in Reason? Getting more audio crackles/breakups, more "computer too slow" messages, etc.? Probably the best way to test is to play a song that pushed your system just about to its limits before and see how it plays now. Anyone? :)
I'm on a 5+ year old Ivy Bridge chip right now (i7-3770K). You first! ;)
Win 10 | Ableton Live 11 Suite |  Reason 12 | i7 3770k @ 3.5 Ghz | 16 GB RAM | RME Babyface Pro | Akai MPC Live 2 & Akai Force | Roland System 8, MX1, TB3 | Dreadbox Typhon | Korg Minilogue XD

househoppin09
Posts: 536
Joined: 03 Aug 2016

04 Jan 2018

Hah! I would volunteer to be the guinea pig if I didn't currently have a couple of projects nearing deadlines that are already straining my system... :)

User avatar
EnochLight
Moderator
Posts: 8405
Joined: 17 Jan 2015
Location: Imladris

04 Jan 2018

househoppin09 wrote:
04 Jan 2018
Hah! I would volunteer to be the guinea pig if I didn't currently have a couple of projects nearing deadlines that are already straining my system... :)
Yeah I'm not volunteering at all, for pretty much the same reason. I've stretched the limits of what my 3770K can do at this point, and have it on a constant overclock at 4.5 Ghz - and any decrease in performance right now would kill me. I've got Windows update blocking that little fix for that reason precisely. I'll live with the vulnerability.
Win 10 | Ableton Live 11 Suite |  Reason 12 | i7 3770k @ 3.5 Ghz | 16 GB RAM | RME Babyface Pro | Akai MPC Live 2 & Akai Force | Roland System 8, MX1, TB3 | Dreadbox Typhon | Korg Minilogue XD

househoppin09
Posts: 536
Joined: 03 Aug 2016

04 Jan 2018

Well, with any luck you won't have to live with the vulnerability--the conventional wisdom I've been seeing is that audio applications may not really be affected. However I've also seen some other anecdotes that contradict that, hence my anxiety and reluctance to apply the patch. Hopefully we'll get some confirmation one way or the other as soon as some ReasonTalk users in our situation try out the patch and report here on what happens.

BTW, for anyone who does post a report, please make sure to include your OS version, CPU model, what audio interface/driver you're using, and your Reason version. Thanks!

User avatar
jappe
Moderator
Posts: 2437
Joined: 19 Jan 2015

05 Jan 2018

househoppin09 wrote:
04 Jan 2018
Well, with any luck you won't have to live with the vulnerability--the conventional wisdom I've been seeing is that audio applications may not really be affected.

"As a result, context switches between the user space and the kernel will get more expensive."

Does this mean heavy interaction with drivers and applications?

https://www.anandtech.com/show/12214/un ... nd-spectre

househoppin09
Posts: 536
Joined: 03 Aug 2016

05 Jan 2018

That's the question! It theoretically shouldn't be a problem if the code is optimized to keep system calls to a reasonable minimum, which is supposed to be best practice for audio applications. You know what Yogi Berra said about theory and practice, though... ;)

RandyEspoda
Posts: 275
Joined: 14 Mar 2017

06 Jan 2018

EnochLight wrote:
04 Jan 2018
househoppin09 wrote:
04 Jan 2018
Hah! I would volunteer to be the guinea pig if I didn't currently have a couple of projects nearing deadlines that are already straining my system... :)
Yeah I'm not volunteering at all, for pretty much the same reason. I've stretched the limits of what my 3770K can do at this point, and have it on a constant overclock at 4.5 Ghz - and any decrease in performance right now would kill me. I've got Windows update blocking that little fix for that reason precisely. I'll live with the vulnerability.
THIS

User avatar
ilikestargazing
Competition Winner
Posts: 72
Joined: 22 Jun 2016
Location: Copenhagen
Contact:

07 Jan 2018

Generally speaking, not music related, there has been very different reports ranging from no effect to 50% loss. My guess is, due to the nature of the vulnerabilities, i.e. NSA sanctioned, it's gonna be hard to get any real answers until we get to test out the patch for ourselves in this specific environment.

Reason never managed to use more than 50-60% total cpu on my system. Even when Reason seems to max out, my overall system resources are always at least 40% free, given I'm not doing anything else resource heavy. If we're lucky that means we won't even feel the difference. But I'm just guessing here :)

RandyEspoda
Posts: 275
Joined: 14 Mar 2017

07 Jan 2018

From KVR forums :

https://www.kvraudio.com/forum/viewtopi ... 6&start=90
EvilDragon wrote:
Audio drivers, though, might be. We'll need to wait for benchmarks after the fix is rolled out on Windows (for macOS, that's already in 10.3.2.)
you’d have to be talking lots of channels (as on 24+ channels input) at very high sample rates AND small buffer sizes before the interrupt rate of any audio interface became enough to even measure

for DAWs that run each plugin in it’s own process the overhead of the context switching for that is many times greater (especially when multiplied by the number of plugins a project will have vs the number of channels in/out).

lnikj wrote:
We aren't average computer users when we have Diva, B2 etc on multiple tracks.
The real problem isn't the CPU intensive nature of Diva or B2. The problem comes in when we have to switch to the kernel to handle an interrupt. Every time you need to send a sample buffer to your USB audio interface you need to do that. So yeah, this is a big deal for sound design people. It's also a big deal if you stream data to/from disk. So Omnisphere and Kontakt users will feel this.

EvilDragon wrote:

Not when you're streaming hundreds of voices on the lowest DFD buffer size from your SSDs. That's a lot of random disk reads right there. Sure it might not saturate the bandwidth fully, but it IS a lot of streaming.
true - people running vienna ensemble could possibly hit high enough i/o loads for it to matter I'd expect. So potentially an issue for people working in the scoring area

by EvilDragon; Wed Jan 03, 2018 1:57 pm

You don't even need to run VEP. Just have a bunch of Kontakt instances streaming various heavy orchestral and other libraries from multiple SSDs at a time (I have 4 1 TB SSDs just for this)...

by rgarner; Thu Jan 04, 2018 12:27 pm

I'm running MacOS 10.12.6, or Sierra (no High – I'm holding off due to that OS still settling down).

I installed this patch on 23rd December, because I had a bit of time spare in the holidays. I was a bit shocked to see the two-reboot loop that usually indicates something firmware- or kernel-related.

I also bought Softube Volume 1 and got into Modular, which I thought was incredibly CPU heavy – suddenly I was having to boost my Logic Pro X I/O buffer size from 128 to 1024 to even run 3 or 4 of these plugins on a (bought earlier this year) top-of-the-line mid-2015 Macbook Pro with quad-core i7@2.8Ghz/16GB, and I'm frequently getting "couldn't process audio in time".

After hearing about Meltdown this morning, I went back to my previous mixes that have never had CPU problems and don't use Modular. Sure enough, I can't run any of them on the built-in outputs without having to whack the I/O buffer up, and that's often no longer enough.

So for me, the "hysteria" about this is entirely justified.


On the other hand :
by EvilDragon; Fri Jan 05, 2018 7:02 am

So after some encouraging results from people at VI-C and GS, I pulled the patch. I did a DAWbench VI in Reaper before the patch and after, and I'm getting much the same numbers: 820 voices at 128 samples buffer with my RME UFX+, and with the factory library on Samsung 850 EVO. i7-6700K at 4.5 GHz here.

So, it seems that ASIO performance wasn't affected one bit (at least as far as Reaper is concerned). Looks like DAW users fall into "average workload" crowd. :)
But this concerns Reaper, not Reason...


Also, since it is now known that 'Spectre' can be mitigated by patching software,
and both Meltdown and Spectre rely on attackers abusing the exploits through software,
I still believe it is very likely that AV developpers and the likes could be able to equally mitigate BOTH exploits in the (near ?) future.
Meltdown has already been fully documented so they can already get busy on that (?)
Spectre is still highly speculative, and still not fully understood, so we don't know yet what that will bring...

My best advice is still to go ahead and install the patch if you have a 6-7-8th gen cpu, for everyone else I would hold off, unless you run Win10, then you'll have to anyway...

User avatar
ilikestargazing
Competition Winner
Posts: 72
Joined: 22 Jun 2016
Location: Copenhagen
Contact:

07 Jan 2018

Btw I'm on Mac though. So far no patch for OSX but when it arrives I'll do a full benchmark and backup before updating. Will keep you posted.

User avatar
joeyluck
Moderator
Posts: 11029
Joined: 15 Jan 2015

08 Jan 2018

ilikestargazing wrote:
07 Jan 2018
Btw I'm on Mac though. So far no patch for OSX but when it arrives I'll do a full benchmark and backup before updating. Will keep you posted.
Hmmm... I thought the patch was included in 10.13.2? I can't tell a difference in performance with Reason and my benchmarks are actually higher with both Novabench and Geekbench.

User avatar
ilikestargazing
Competition Winner
Posts: 72
Joined: 22 Jun 2016
Location: Copenhagen
Contact:

08 Jan 2018

Oh well that's probably true! I'm still running 10.11 :D

User avatar
Marco Raaphorst
Posts: 2504
Joined: 22 Jan 2015
Location: The Hague, The Netherlands
Contact:

08 Jan 2018

why is Apple so slow? this is all about security. are they insane?!

RandyEspoda
Posts: 275
Joined: 14 Mar 2017

08 Jan 2018

joeyluck wrote:
08 Jan 2018
ilikestargazing wrote:
07 Jan 2018
Btw I'm on Mac though. So far no patch for OSX but when it arrives I'll do a full benchmark and backup before updating. Will keep you posted.
Hmmm... I thought the patch was included in 10.13.2? I can't tell a difference in performance with Reason and my benchmarks are actually higher with both Novabench and Geekbench.
I have also read nothing but good results for MAC users, so you guys should be good.
It's mainly windows in combo with older cpus that are affected the worst.

User avatar
EnochLight
Moderator
Posts: 8405
Joined: 17 Jan 2015
Location: Imladris

08 Jan 2018

Marco Raaphorst wrote:
08 Jan 2018
why is Apple so slow? this is all about security. are they insane?!
Are you kidding? Apple's patch will fix things and be done in one update. Everyone else on an Intel chip will not only need a patch from Windows, but also a firmware update for their hardware - and that can be one of any dozens of manufacturers! Trust me, Apple may be slow at doing a lot of things, but you guys got lucky on this one. ;)
Win 10 | Ableton Live 11 Suite |  Reason 12 | i7 3770k @ 3.5 Ghz | 16 GB RAM | RME Babyface Pro | Akai MPC Live 2 & Akai Force | Roland System 8, MX1, TB3 | Dreadbox Typhon | Korg Minilogue XD

User avatar
theshoemaker
Posts: 595
Joined: 21 Nov 2015
Location: Germany
Contact:

08 Jan 2018

Meltdown is already fixed with 10.13.2 (High Sierra) ... pretty sad they removed the patch from the release notes from Sierra and Maverics, because I'm still on Sierra for several reasons. You can read about that here: https://support.apple.com/en-us/HT208394
:PUF_figure: latest :reason: V12 on MacOS Ventura

User avatar
ilikestargazing
Competition Winner
Posts: 72
Joined: 22 Jun 2016
Location: Copenhagen
Contact:

08 Jan 2018

Marco Raaphorst wrote:
08 Jan 2018
why is Apple so slow? this is all about security. are they insane?!
The insane part is that someone at Intel knowingly let this issue exist for nearly 2 decades.

User avatar
EnochLight
Moderator
Posts: 8405
Joined: 17 Jan 2015
Location: Imladris

08 Jan 2018

ilikestargazing wrote:
08 Jan 2018
Marco Raaphorst wrote:
08 Jan 2018
why is Apple so slow? this is all about security. are they insane?!
The insane part is that someone at Intel knowingly let this issue exist for nearly 2 decades.
And... ARM... and AMD...
Win 10 | Ableton Live 11 Suite |  Reason 12 | i7 3770k @ 3.5 Ghz | 16 GB RAM | RME Babyface Pro | Akai MPC Live 2 & Akai Force | Roland System 8, MX1, TB3 | Dreadbox Typhon | Korg Minilogue XD

User avatar
Marco Raaphorst
Posts: 2504
Joined: 22 Jan 2015
Location: The Hague, The Netherlands
Contact:

08 Jan 2018

EnochLight wrote:
08 Jan 2018
Marco Raaphorst wrote:
08 Jan 2018
why is Apple so slow? this is all about security. are they insane?!
Are you kidding? Apple's patch will fix things and be done in one update. Everyone else on an Intel chip will not only need a patch from Windows, but also a firmware update for their hardware - and that can be one of any dozens of manufacturers! Trust me, Apple may be slow at doing a lot of things, but you guys got lucky on this one. ;)
Yes except for that only High Sierra is patched at the moment.

I guess cloud services are the main issue.

User avatar
EnochLight
Moderator
Posts: 8405
Joined: 17 Jan 2015
Location: Imladris

08 Jan 2018

Marco Raaphorst wrote:
08 Jan 2018
I guess cloud services are the main issue.
And most of the cloud runs on Intel chips. I'm not looking forward to how this pans out. :(
Win 10 | Ableton Live 11 Suite |  Reason 12 | i7 3770k @ 3.5 Ghz | 16 GB RAM | RME Babyface Pro | Akai MPC Live 2 & Akai Force | Roland System 8, MX1, TB3 | Dreadbox Typhon | Korg Minilogue XD

User avatar
ilikestargazing
Competition Winner
Posts: 72
Joined: 22 Jun 2016
Location: Copenhagen
Contact:

08 Jan 2018

theshoemaker wrote:
08 Jan 2018
Meltdown is already fixed with 10.13.2 (High Sierra) ... pretty sad they removed the patch from the release notes from Sierra and Maverics, because I'm still on Sierra for several reasons. You can read about that here: https://support.apple.com/en-us/HT208394
I'm curious, are you actually a shoemaker? Cause that would be pretty cool.

User avatar
theshoemaker
Posts: 595
Joined: 21 Nov 2015
Location: Germany
Contact:

09 Jan 2018

ilikestargazing wrote:
08 Jan 2018
theshoemaker wrote:
08 Jan 2018
Meltdown is already fixed with 10.13.2 (High Sierra) ... pretty sad they removed the patch from the release notes from Sierra and Maverics, because I'm still on Sierra for several reasons. You can read about that here: https://support.apple.com/en-us/HT208394
I'm curious, are you actually a shoemaker? Cause that would be pretty cool.
My german family name is ... Maybe I rename myself to theshowmaker :D
:PUF_figure: latest :reason: V12 on MacOS Ventura

User avatar
EnochLight
Moderator
Posts: 8405
Joined: 17 Jan 2015
Location: Imladris

09 Jan 2018

Warning to AMD users, especially on older machines:

https://www.engadget.com/2018/01/09/mic ... d-patches/
Win 10 | Ableton Live 11 Suite |  Reason 12 | i7 3770k @ 3.5 Ghz | 16 GB RAM | RME Babyface Pro | Akai MPC Live 2 & Akai Force | Roland System 8, MX1, TB3 | Dreadbox Typhon | Korg Minilogue XD

User avatar
Marco Raaphorst
Posts: 2504
Joined: 22 Jan 2015
Location: The Hague, The Netherlands
Contact:

09 Jan 2018

Or get yourself an old 486 since it does not support branch prediction :D

Post Reply
  • Information
  • Who is online

    Users browsing this forum: pushedbutton and 24 guests