Reason 9.5.2 is now available!

This forum is for discussing Reason. Questions, answers, ideas, and opinions... all apply.
User avatar
miscend
Posts: 1955
Joined: 09 Feb 2015

22 Sep 2017

Marco Raaphorst wrote:
22 Sep 2017
Can't open Reason any longer! Getting the "Reason quit unexpectedly" after I start it.

How can I downgrade to previous version?
If you are on a Mac the old version is in the Recycle bin. If you cleared the recycle bin have a look at your Time Machine backup.

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

22 Sep 2017

miscend wrote:
22 Sep 2017
Marco Raaphorst wrote:
22 Sep 2017
Can't open Reason any longer! Getting the "Reason quit unexpectedly" after I start it.

How can I downgrade to previous version?
If you are on a Mac the old version is in the Recycle bin. If you cleared the recycle bin have a look at your Time Machine backup.
WOW! that worked.

Hope this issue gets solved.

User avatar
Dante
Posts: 531
Joined: 06 Jun 2015
Location: Australia
Contact:

22 Sep 2017

Matthias - thanks for the update. From my observation a big opportunity for VST optimization exists for the UAD platform, as there seem to be large discrepancies between host CPU consumption on Reason as compared to other DAWs - and seems to affect irrespective of whether over USB, Firewire (external units) and PCIe (internal cards).

I am not a UAD owner myself so have not done my own tests, but I hope to be one day if this is addressed. Its just that I have noticed quite a few postings about this since VST support and subsequent point releases, mainly these two:

viewtopic.php?f=48&t=7503021&hilit=UAD
viewtopic.php?f=48&t=7502758&hilit=UAD

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

22 Sep 2017

Marco Raaphorst wrote:
22 Sep 2017
miscend wrote:
22 Sep 2017

If you are on a Mac the old version is in the Recycle bin. If you cleared the recycle bin have a look at your Time Machine backup.
WOW! that worked.

Hope this issue gets solved.
I'm not having any issues here on macOS 10.12.6 on my MacBook Air

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

22 Sep 2017

Marco Raaphorst wrote:
22 Sep 2017
Can't open Reason any longer! Getting the "Reason quit unexpectedly" after I start it.

How can I downgrade to previous version?
With 9.5.2 we have made changes to the caches and file system to be compatible with High Sierra which means you'll likely have to re-scan VSTs. When scanning VSTs, a VST plug-in could potentially crash Reason at start-up. The next time you start up Reason, a crashed plug-in will be ignored during scanning. If I had to speculate, this might be why your 9.5.2 would quit and not 9.5.1.

Try moving all your VSTs temporarily out of the VST folder and launch Reason 9.5.2. Does it work? If so, a VST is causing a crash.

per-anders
Posts: 224
Joined: 09 Jul 2015

22 Sep 2017

Please please please please look into the threading. You guys know the drill - no locks/semaphore, volatile or openmp, reduce memory and disk access, use const, aligned well blocked data, atomics, keep threads running etc, get the locks and cache misses down, profile like heck. If the problem is Lua then maybe it's time to build your own interpreter or change schema, it seems like you're mostly using it to just parse resource files anyway and some bare bones interface logic.

User avatar
Exowildebeest
Posts: 1553
Joined: 16 Jan 2015

22 Sep 2017

per-anders wrote:
22 Sep 2017
Please please please please look into the threading. You guys know the drill - no locks/semaphore, volatile or openmp, reduce memory and disk access, use const, aligned well blocked data, atomics, keep threads running etc, get the locks and cache misses down, profile like heck. If the problem is Lua then maybe it's time to build your own interpreter or change schema, it seems like you're mostly using it to just parse resource files anyway and some bare bones interface logic.
They should also look into the tiberium core processing phase equation fuctuation - as we all know, those can be a batch if they interact with the binary vector operator routing plumbus. Quantum mechanically speaking, in n-dimensional R-space, naturally.

User avatar
aeox
Competition Winner
Posts: 3222
Joined: 23 Feb 2017
Location: Oregon

22 Sep 2017

Exowildebeest wrote:
22 Sep 2017
They should also look into the tiberium core processing phase equation fuctuation - as we all know, those can be a batch if they interact with the binary vector operator routing plumbus. Quantum mechanically speaking, in n-dimensional R-space, naturally.
Image

User avatar
normen
Posts: 3431
Joined: 16 Jan 2015

22 Sep 2017

per-anders wrote:
22 Sep 2017
Please please please please look into the threading. You guys know the drill - no locks/semaphore, volatile or openmp, reduce memory and disk access, use const, aligned well blocked data, atomics, keep threads running etc, get the locks and cache misses down, profile like heck. If the problem is Lua then maybe it's time to build your own interpreter or change schema, it seems like you're mostly using it to just parse resource files anyway and some bare bones interface logic.
:lol:
Exowildebeest wrote:
22 Sep 2017
They should also look into the tiberium core processing phase equation fuctuation - as we all know, those can be a batch if they interact with the binary vector operator routing plumbus. Quantum mechanically speaking, in n-dimensional R-space, naturally.
Thats about the same content, yeah ^^

per-anders
Posts: 224
Joined: 09 Jul 2015

22 Sep 2017

normen wrote:
22 Sep 2017
per-anders wrote:
22 Sep 2017
Please please please please look into the threading. You guys know the drill - no locks/semaphore, volatile or openmp, reduce memory and disk access, use const, aligned well blocked data, atomics, keep threads running etc, get the locks and cache misses down, profile like heck. If the problem is Lua then maybe it's time to build your own interpreter or change schema, it seems like you're mostly using it to just parse resource files anyway and some bare bones interface logic.
:lol:
Exowildebeest wrote:
22 Sep 2017
They should also look into the tiberium core processing phase equation fuctuation - as we all know, those can be a batch if they interact with the binary vector operator routing plumbus. Quantum mechanically speaking, in n-dimensional R-space, naturally.
Thats about the same content, yeah ^^
Lets just hope the Props know more about MT coding than you then...

User avatar
normen
Posts: 3431
Joined: 16 Jan 2015

22 Sep 2017

per-anders wrote:
22 Sep 2017
Lets just hope the Props know more about MT coding than you then...
Oh, you were serious? Sorry, I thought you were just throwing around these basic multithreading-related terms as a joke. Most of those would have been decided on long ago or be defined by the circumstances anyway? Fyi, LUA is used for quite a bit more than just "bare bones interface logic".

No hard feelings, I thought you were just doing the equivalent of the boss coming in going "Are you doing your -- whatever you are doing?" ;)

RandyEspoda
Posts: 275
Joined: 14 Mar 2017

22 Sep 2017

Thanks for the update, props !

per-anders
Posts: 224
Joined: 09 Jul 2015

22 Sep 2017

normen wrote:
22 Sep 2017
per-anders wrote:
22 Sep 2017
Lets just hope the Props know more about MT coding than you then...
Oh, you were serious? Sorry, I thought you were just throwing around these basic multithreading-related terms as a joke. Most of those would have been decided on long ago or be defined by the circumstances anyway? Fyi, LUA is used for quite a bit more than just "bare bones interface logic".

No hard feelings, I thought you were just doing the equivalent of the boss coming in going "Are you doing your -- whatever you are doing?" ;)
Ohhkay... I guess you don't realize that those aren't a sequence of random basic mt related terms, but the basics of producing good MT code. A highly concatenated and slightly omissive off the cuff 101 guide/reminder.

It doesn't matter how clever people imagine themselves to be, they're human, they make mistakes, or they just don't yet know the pitfalls or the architecture and of course architectures do change.

Reason scales negatively. That's not an indicator of good things. There are tools and armies of young keen developers willing to prove their mustard if the Prop's dev team isn't overjoyed at the prospect of refactoring millions of lines of code. Of course the Props likely know exactly what the bottleneck is, but never assume and it still doesn't hurt to double check the basics, my post was a gentle prod, because at one point Reason was the lightweight (CPU wise) option, now it's the fat kid on the block.

User avatar
Gorgon
Posts: 1233
Joined: 11 Mar 2016

22 Sep 2017

MattiasHG wrote:
22 Sep 2017
And an important thing to understand about hyper threading: Hyper threading works by distributing load better. However, the way audio processing and Reason works when you put multiple devices in one chain (for example as insert fx or create one monster combinator) they are all processed together as one unit since all things must happen before they can render.
That's really shitty programming if you ask me. That is not why multicore was invented.Real multicore processing should distribute the load of any given millisecond to all the cores. The way you have it now is that some cores have to wait on others for rendering. This is very similar to "wait states" which were a thing back in the 90's with the 386/486 processors.
"This is a block of text that can be added to posts you make. There is a 255 character limit."

User avatar
normen
Posts: 3431
Joined: 16 Jan 2015

22 Sep 2017

Gorgon wrote:
22 Sep 2017
That's really shitty programming if you ask me. That is not why multicore was invented.Real multicore processing should distribute the load of any given millisecond to all the cores. The way you have it now is that some cores have to wait on others for rendering. This is very similar to "wait states" which were a thing back in the 90's with the 386/486 processors.
Uhm.. Its simple logic. To be able to manipulate a sound coming out of an amp with an EQ you FIRST need to put the sound through the amp, THEN you apply the EQ. Its a sequential process, there is simply no way to do that at the same time, you need the result of the previous plugin first.

Edit: If the plugins themselves use MT to do their thing then they can do that no problem. What Mattias is speaking about is the normal load balancing happening from the Reason side. But few plugins really do use MT, many of the problems in audio DSP have the same issue as the signal flow, a lot of stuff simply has to happen sequentially.

User avatar
MikeMcKew
Moderator
Posts: 199
Joined: 16 Aug 2017
Location: Leesburg, VA
Contact:

23 Sep 2017

I appreciate this update!

User avatar
Oquasec
Posts: 2849
Joined: 05 Mar 2017

23 Sep 2017

My ass got reason on layaway half a decade ago lmfao
Producer/Programmer.
Reason, FLS and Cubase NFR user.

Spyderfyre
Posts: 3
Joined: 23 Sep 2017

23 Sep 2017

Does anyone know if this update adds support for more than 4 cores?

seqoi
Posts: 417
Joined: 12 Aug 2017

23 Sep 2017

MattiasHG wrote:
22 Sep 2017
Heater wrote:
22 Sep 2017
Any improvements to VST performance?
This release has no general performance improvements.

For everyone with performance issues I have to stress all the normal things to try when you're experiencing issues:
  • Try with Hyper Threading both on and off
  • If you have VSTs that use a lot of juice, see if they have CPU settings. Many support multi-core/hyper threading where on/off can make a big difference.
  • Try increasing your buffer size
  • Try setting sample rate to 44.1, it's generally the only thing you need sonically and performs better
  • If you're on a Mac, try opening Reason in Low-Resolution Mode. Right click on the application, select Get Info and choose Open in Low-Resolution
  • Close all background applications
  • Don't confuse CPU monitor with what's audibly happening. You can see high CPU use sometimes because Reason reserves some CPU to not have to wake up idle threads.
And an important thing to understand about hyper threading: Hyper threading works by distributing load better. However, the way audio processing and Reason works when you put multiple devices in one chain (for example as insert fx or create one monster combinator) they are all processed together as one unit since all things must happen before they can render. Therefore hyper threading won't necessarily help when you just have a handful of super heavy plug-ins, but can help a lot when you have a lot of devices and tracks.
Mattias thanks for chiming in here. HT makes things worse not better. I hate to stress that i lost countless days trying to find is it my side error or Reason error but after seeing many people reporting exactly same issues across mac and win i don't see how is this my error. And it's not HT either.

As i pointed out in other threads there is drastically worse VST performance when compared to any other existing DAW. See my example here ( viewtopic.php?f=47&t=7502998)

Mattias i am all for upgrade i will buy it same day of release just tell me you are working on VST performance issues (i and many users reported UAD issues ) and they are going to be remedied in Reason 10 sometime - please don't tell me there is misconception here and that you guys at PH think that Reason bad VST performance is on users side?

seqoi
Posts: 417
Joined: 12 Aug 2017

23 Sep 2017

MattiasHG wrote:
22 Sep 2017

Of course we haven't stopped working on improving performance. No need to worry! It's just important that performance reports are actually about worse performance and not about a miss in information on how to optimize. :)
Sorry i responded but did not saw your later response. Thanks man. Keep up with good work there. And yes you are right i am making more music with Reason alone (minus VST) then ever before. I merely wanted to know are you guys aware of problem.

Have a nice weekend.

User avatar
Oquasec
Posts: 2849
Joined: 05 Mar 2017

23 Sep 2017

This is for people who skipped iris & sylenth btw or don't use those. [Or the few of you who got those]
Producer/Programmer.
Reason, FLS and Cubase NFR user.

User avatar
Noise
Competition Winner
Posts: 470
Joined: 16 Jan 2015
Location: Lisbon
Contact:

23 Sep 2017

Reason 9.5.0 (build 8.412) is the perfect build for my setup, before the Hyper Threading newest functionality. Recent versions are loading slower and eat one more precious DSP bar. Totally kills the creative process. I've got a feeling I will be sticking with this version for a long time!
Albums: BandCamp | Youtubz: Noise Channel
Projects: P1 Easy Remote Mapping | Personal Refill Sale Store: https://payhip.com/noisesystems | Title Generator! untitled.noiseshadow.com

avasopht
Competition Winner
Posts: 3931
Joined: 16 Jan 2015

23 Sep 2017

per-anders wrote:
22 Sep 2017
It doesn't matter how clever people imagine themselves to be, they're human, they make mistakes, or they just don't yet know the pitfalls or the architecture and of course architectures do change.
It's less to do with clever people and just a matter of in depth knowledge, intuition and experience. A brilliant DSP engineer is not necessarily a kick ass UI developer who could efficaciously develop GUI frameworks to accelerate the R&D process of advancing the UX.

From the little they've shared about their approach to VSTs and scheduling, they sound like they have their heads screwed on. It's important to remember that we don't know at all what they're working on behind the scenes.

Even the complaints about V10's features. For all we know they could have a complete UI overhaul that they chose not to disclose at this point in time. It wouldn't be the first time!

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

23 Sep 2017

PelTek wrote:
22 Sep 2017
None of the bugs I reported since beta are fixed, great!!
Hello PelTek,
curious what bugs you refer to?

syrokitty
Posts: 107
Joined: 15 Jun 2015
Location: Austin, TX.
Contact:

23 Sep 2017

I'm assuming, by the lack of discussion regarding continued VST GUI lag, I must be experiencing user error. Weird.

Post Reply
  • Information