Reason Slow Performance - when is optimisation coming?

This forum is for discussing Reason. Questions, answers, ideas, and opinions... all apply.
calebbrennan
Posts: 315
Joined: 16 Aug 2016

11 Aug 2018

Reason performs like a champ on a PC with 8 gigs of ram.
I run no VST

I'm hearing Macs are having issues.

Part of the issue is trying to have every bit of software you can install on your machine installed.

Also always remember to record that great synth line and Delete the instrument and you will be fine.
You can leave the unit in a previous version. Commitment to Audio is challenge for synth geeks because "You can't tweak it anymore!!! Good!
Remember Wendy Carlos with a few Moog's and a tape machine!!!!
When you have every option in the world there is no direction.
We are all like that, and I love endless possibilities
realize confining my laser beam could be helpful



.

User avatar
chimp_spanner
Posts: 2908
Joined: 06 Mar 2015

11 Aug 2018

I work with Reason pretty much every day for my job/s. Performance is generally pretty good. A typical project for me on an i7 4700MQ 12GB is 64-96 samples at 48khz, half a dozen tracks of guitar with POD Farm/Guitar Rig 5/etc., Superior Drummer 2 with mostly rack extensions on multiple mix channels, Kontakt bass, some synth pads like Antidote, Europa for special FX. That kinda thing. Those projects usually get me up in the high 70-80s, although really what I should be doing is writing/tracking and producing as separate processes. It's somewhat of a bad habit of mine to try and do it all at once but, it is what it is. Hard to keep those processes separate when it's all so accessible.

Now that said, I will sometimes work on other stuff that hits that limit *way* quicker, with seemingly less going on. I think some of it is perhaps down to my own lack of understanding as to why certain effects or instruments use more than others. But I do feel like I'm out of control of it a bit sometimes, ya know? Like, in Cubase if I put a VST in and my DSP skyrockets, I just up the buffer size or I freeze. In Reason, it feels like sometimes when you stumble across a particular combination of devices and plugins it really goes downhill quickly (UI activity starts to impact hard) and it's hard to tell why exactly. Could be a VST that's not playing ball (NI dynamics and EQs consume DSP bars with each instance without any signal, for instance). Or it could be...something else. Sometimes HT works, sometimes it doesn't. Sometimes higher buffer helps, sometimes it doesn't.

Again I think some of this is down to perhaps not fully understanding what I'm asking of my computer, or indeed exactly how Reason does what it does. I don't know anything about coding plugins, although I've seen the "64 samples" theory come up quite a few times and it seems to make sense. VCV Rack VST Bridge reports 64 samples at all times, regardless of my sample size. As far as I'm aware it reports correctly in Cubase. Not saying this means anything though. VCV is still in...alpha? Beta? So it could just as easily be a quirk/bug! Or it could be a smoking gun :)

Anyway not complaining - just sharing my experience! It's great to hear that PH are making some positive progress. Don't mind if it takes a while. It's just nice to know it's being addressed! I also think it'd be great if, when we post about performance, we have a sort of...standardised format for posting. System specs, project settings (sample rate, buffer size, etc.), plugins in use, etc. So many possible combinations of software and hardware. Glad I don't have to sort it out haha.

User avatar
NekujaK
Posts: 631
Joined: 09 Oct 2016
Location: USA

11 Aug 2018

Here's a simple way of getting a crude but illustrative performance benchmark: open a completely blank project, create a new audio track, and start adding instances of Waves Abbey Road Plates (stereo) VST as inserts. (I'm using this particular plugin because it's known to be a CPU hog).

Here are my results when I do this:

- In Reason, adding the 6th instance maxes out the CPU meter and the "too slow to play song" dialog appears.

- In Reaper, I can add 56 instances of the plugin before the CPU meter exceeds 100%, and you can still press play without any stoppage or dialogs popping up.

I know this isn't the most sophisticated method of measurement, but even so, the differences are too huge and obvious to be ignored.
wreaking havoc with :reason: since 2.5
:arrow: https://soundcloud.com/nekujak-donnay/sets

User avatar
Oquasec
Posts: 2849
Joined: 05 Mar 2017

11 Aug 2018

I called it a day and put Reason inside fl studio. I'm not waitin for things to change but I'm keepin propellerhead's modules.
Producer/Programmer.
Reason, FLS and Cubase NFR user.

User avatar
selig
RE Developer
Posts: 11685
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

11 Aug 2018

NekujaK wrote:Here's a simple way of getting a crude but illustrative performance benchmark: open a completely blank project, create a new audio track, and start adding instances of Waves Abbey Road Plates (stereo) VST as inserts. (I'm using this particular plugin because it's known to be a CPU hog).

Here are my results when I do this:

- In Reason, adding the 6th instance maxes out the CPU meter and the "too slow to play song" dialog appears.

- In Reaper, I can add 56 instances of the plugin before the CPU meter exceeds 100%, and you can still press play without any stoppage or dialogs popping up.

I know this isn't the most sophisticated method of measurement, but even so, the differences are too huge and obvious to be ignored.
As I understand it, a better simple benchmark would be to create an audio track, add the plugin, then DUPLICATE the track. It would be very uncommon to stack the same plugin on one track, but far more common to add it to multiple tracks. And again, as I understand it, the CPU hit can be different with this test and will represent a more “real world” application.
:)


Sent from some crappy device using Tapatalk
Selig Audio, LLC

User avatar
NekujaK
Posts: 631
Joined: 09 Oct 2016
Location: USA

11 Aug 2018

selig wrote:
11 Aug 2018
NekujaK wrote:Here's a simple way of getting a crude but illustrative performance benchmark: open a completely blank project, create a new audio track, and start adding instances of Waves Abbey Road Plates (stereo) VST as inserts. (I'm using this particular plugin because it's known to be a CPU hog).

Here are my results when I do this:

- In Reason, adding the 6th instance maxes out the CPU meter and the "too slow to play song" dialog appears.

- In Reaper, I can add 56 instances of the plugin before the CPU meter exceeds 100%, and you can still press play without any stoppage or dialogs popping up.

I know this isn't the most sophisticated method of measurement, but even so, the differences are too huge and obvious to be ignored.
As I understand it, a better simple benchmark would be to create an audio track, add the plugin, then DUPLICATE the track. It would be very uncommon to stack the same plugin on one track, but far more common to add it to multiple tracks. And again, as I understand it, the CPU hit can be different with this test and will represent a more “real world” application.
:)
A fair point. Using this suggested approach, my results are:

- Reason: 13 tracks/instances before maxing the CPU meter, and getting the dreaded "too slow" dialog.

- Reaper: 32 tracks/instances before maxing the CPU meter.

Reason fared significantly better using this method, while Reaper suffered a bit, but even so, the disparity remains quite large. Would love to see what kind of results other folks are getting. Even if everyone's machine specs are different, it would be interesting to see the performance ratio between Reason and other DAWs on the same machine, using this method.
wreaking havoc with :reason: since 2.5
:arrow: https://soundcloud.com/nekujak-donnay/sets

User avatar
chimp_spanner
Posts: 2908
Joined: 06 Mar 2015

11 Aug 2018

I’ve done a similar test in the past; playing a chord on a serum patch, although the difference wasn’t too great. I think it was 21 in Cubase and 19 in reason. But that’s just that patch, and that plugin.

User avatar
selig
RE Developer
Posts: 11685
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

11 Aug 2018

NekujaK wrote:
selig wrote:
11 Aug 2018
As I understand it, a better simple benchmark would be to create an audio track, add the plugin, then DUPLICATE the track. It would be very uncommon to stack the same plugin on one track, but far more common to add it to multiple tracks. And again, as I understand it, the CPU hit can be different with this test and will represent a more “real world” application.
:)
A fair point. Using this suggested approach, my results are:

- Reason: 13 tracks/instances before maxing the CPU meter, and getting the dreaded "too slow" dialog.

- Reaper: 32 tracks/instances before maxing the CPU meter.

Reason fared significantly better using this method, while Reaper suffered a bit, but even so, the disparity remains quite large. Would love to see what kind of results other folks are getting. Even if everyone's machine specs are different, it would be interesting to see the performance ratio between Reason and other DAWs on the same machine, using this method.
How did the various settings in Reason’s Prefs affect things - better, worse, the same?


Sent from some crappy device using Tapatalk
Selig Audio, LLC

User avatar
Data_Shrine
Posts: 517
Joined: 23 Jan 2015

11 Aug 2018

Reason has slowed down by almost 50% since version 8 (for me).
9.5 was quite a hit, but with version 10 it got worse (I'm talking even just running old stock synths).

In comparaison, Live 10 is only slightly hungrier than Live 9 (and maybe they will optimize it too over the course of it's lifespan, like they did with version 9).

Anyway I really hope performance will get better soon. I'm sorry to say that I will not update to R11 unless performance and graphic issues are taken care of.

Anyway thanks for the update.

antic604

12 Aug 2018

Data_Shrine wrote:
11 Aug 2018
In comparaison, Live 10 is only slightly hungrier than Live 9 (and maybe they will optimize it too over the course of it's lifespan, like they did with version 9).
Why the double standard? Why won't you assume Props do the same for Reason during 10 lifespan?

Also, I know of several people still on Live 8, because it's faster, so go figure. That's technology for you.

User avatar
chimp_spanner
Posts: 2908
Joined: 06 Mar 2015

12 Aug 2018

Performance degradation with new versions isn’t exclusive to one platform. Cubase 7 was a dream for me. 8 onwards was a LOT more jittery and sensitive to heavy loads. I haven’t checked in a while but I remember on 8.5 I’d get CPU spikes in an empty project just drawing a selection rectangle! 9.5 is perhaps a little better but I’m so sick of fighting with elicenser and hanging on startup that I rarely bother with it anymore!

User avatar
Oquasec
Posts: 2849
Joined: 05 Mar 2017

12 Aug 2018

Honestly I think cubase has one of the worst workflows for me in general from what I've used so far.
Reaper was a much smooth choice for me.
Producer/Programmer.
Reason, FLS and Cubase NFR user.

User avatar
NekujaK
Posts: 631
Joined: 09 Oct 2016
Location: USA

14 Aug 2018

selig wrote:
11 Aug 2018
NekujaK wrote: A fair point. Using this suggested approach, my results are:

- Reason: 13 tracks/instances before maxing the CPU meter, and getting the dreaded "too slow" dialog.

- Reaper: 32 tracks/instances before maxing the CPU meter.

Reason fared significantly better using this method, while Reaper suffered a bit, but even so, the disparity remains quite large. Would love to see what kind of results other folks are getting. Even if everyone's machine specs are different, it would be interesting to see the performance ratio between Reason and other DAWs on the same machine, using this method.
How did the various settings in Reason’s Prefs affect things - better, worse, the same?
This was kind of interesting. With hyperthreading turned on (I normally have it off), I was able to create 16 tracks/instances before Reason choked. The other options (multi-core, buffer size, etc.) didn't seem to have any noticeable effect. Bear in mind, Reason is sitting idle and not being asked to play anything during this test.

Seeing that hyperthreading seemed to have a performance benefit, I wanted to try it on a client mix that's been giving me chronic performance problems. Unfortunately, the result was disappointing. With hyperthreading off, Reason can play through about 4 measures of the song before stopping. With hyperthreading on, Reason could barely play 2 beats.

Oh well, so much for that :cry:
wreaking havoc with :reason: since 2.5
:arrow: https://soundcloud.com/nekujak-donnay/sets

User avatar
mcatalao
Competition Winner
Posts: 1824
Joined: 17 Jan 2015

17 Oct 2018

Guys i still have R8 and R10 installed (because we lost Line 6 on R9x). I run an R8 project on R8 and R10 and the performance is barely the same.

What changed was the introdution of VST's and addition of really DSP hungry RE's. You must compare the projects with the same conditions.

User avatar
MattiasHG
Reason Studios
Posts: 488
Joined: 16 Jan 2015

18 Oct 2018

mcatalao wrote:
17 Oct 2018
Guys i still have R8 and R10 installed (because we lost Line 6 on R9x). I run an R8 project on R8 and R10 and the performance is barely the same.

What changed was the introdution of VST's and addition of really DSP hungry RE's. You must compare the projects with the same conditions.
I was just about to come here to post this! When you compare performance between Reason versions, you need to compare them on the same terms. Same projects, same audio interface, same computer, same settings. Otherwise it's very likely something else has changed.

It'd be very strange to say "My Reason 8 song with mostly native devices and some REs had way better performance than my Reason 10 song with Serum, Ozone and tons of other VSTs". :)

sdst
Competition Winner
Posts: 896
Joined: 14 Jun 2015

18 Oct 2018

just give me Reason 5 with audio track and rack extensions,

forget reason 11

User avatar
Loque
Moderator
Posts: 11170
Joined: 28 Dec 2015

18 Oct 2018

*set sarcasm=on*
I can run much more Subtractors now than the first time i tried Reason, so logically they must have done a massive performance improvement 🤔
*set sarcasm=off*
Reason12, Win10

Undistraction

18 Oct 2018

@MattiasHG Please can you confirm that the 10.3 performance improvements are going to address RE performance degradation as well as fixing VST performance.

User avatar
mcatalao
Competition Winner
Posts: 1824
Joined: 17 Jan 2015

18 Oct 2018

sdst wrote:
18 Oct 2018
just give me Reason 5 with audio track and rack extensions,

forget reason 11
You want Reason 6.5. With the OLDEST rack extensions... :)

User avatar
selig
RE Developer
Posts: 11685
Joined: 15 Jan 2015
Location: The NorthWoods, CT, USA

18 Oct 2018

Undistraction wrote:@MattiasHG Please can you confirm that the 10.3 performance improvements are going to address RE performance degradation as well as fixing VST performance.
I haven’t personally seen any obvious performance degradation across all the releases that support REs - what examples do you have and how have you tested to come to this conclusion?


Sent from some crappy device using Tapatalk
Selig Audio, LLC

User avatar
mcatalao
Competition Winner
Posts: 1824
Joined: 17 Jan 2015

18 Oct 2018

Loque wrote:
18 Oct 2018
*set sarcasm=on*
I can run much more Subtractors now than the first time i tried Reason, so logically they must have done a massive performance improvement 🤔
*set sarcasm=off*
If that sarcasm is directed to me, maybe you should read my post again. Carefully. Because that's not what i was saying AT ALL.

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

18 Oct 2018

MattiasHG wrote:
18 Oct 2018
I was just about to come here to post this! When you compare performance between Reason versions, you need to compare them on the same terms. Same projects, same audio interface, same computer, same settings. Otherwise it's very likely something else has changed.

It'd be very strange to say "My Reason 8 song with mostly native devices and some REs had way better performance than my Reason 10 song with Serum, Ozone and tons of other VSTs". :)
Great point!

That said (and completely related): when the legendary update appears that it supposed to address Reason's VST-performance issues, please PLEASE PLEASE include the option to properly freeze tracks. This additional feature would help a tremendous amount for people who really need to push Reason to its limits.
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
Kalm
Posts: 554
Joined: 03 Jun 2016
Location: Austin
Contact:

18 Oct 2018

My performance lag is based on the graphic interface of a couple devices I use, multichannel output routing, and big toys like ReSpire, Softube TSAR-1 etc. The minute I decide to remove any one of these out of my sessions, I get 1/3 of my power back. I have a little more wiggle room on my mac mini than my MacBook Pro.

But bouncing my sessions as audios to another session and constantly doing that workflow has actually improved things for me. Strange.
Courtesy of The Brew | Watch My Tutorials | Mac Mini Intel i7 Quad-Core | 16 GB RAM | Samsung 850 EVO 250 GB | Reason 11 Suite | Studio One 5 Professional | Presonus Quantum | Komplete Kontrol 49 MK2 | Event Opals | Follow me on Instagram

Michaellos
Posts: 153
Joined: 18 May 2016

18 Oct 2018

selig wrote:
18 Oct 2018
Undistraction wrote:@MattiasHG Please can you confirm that the 10.3 performance improvements are going to address RE performance degradation as well as fixing VST performance.
I haven’t personally seen any obvious performance degradation across all the releases that support REs - what examples do you have and how have you tested to come to this conclusion?


Sent from some crappy device using Tapatalk
I would say that Expanse, for example, is pretty much cpu hungry, much more hungry than Omnisphere 2.5, for example. I would love to see some sort of Re performance optimization as well as VST

User avatar
QVprod
Moderator
Posts: 3488
Joined: 15 Jan 2015
Contact:

19 Oct 2018

Kalm wrote:
18 Oct 2018
My performance lag is based on the graphic interface of a couple devices I use, multichannel output routing, and big toys like ReSpire, Softube TSAR-1 etc. The minute I decide to remove any one of these out of my sessions, I get 1/3 of my power back. I have a little more wiggle room on my mac mini than my MacBook Pro.

But bouncing my sessions as audios to another session and constantly doing that workflow has actually improved things for me. Strange.
Not strange at all. Audio isn't cpu intensive. Plugins are. Which is why many want a freeze function as that essentially allows you to bounce without deleting the plugin in case there's further tweaks or automation one wants to add later on. Basically what you do now, but without the need for a separate session.

Post Reply
  • Information
  • Who is online

    Users browsing this forum: No registered users and 23 guests