Goodbye Hydlide?
I've started work on a new track. At the moment I've lined up about 28 different devices, mostly RE synths, including 1 Kong, 1 Echo, 3 audio tracks with various samples on them. 2 Utility RE's and 2 vst effects. I'll probably remove and change some of the synths as at the moment as I'm testing things. So far only 5 of the tracks have sequence data and one audio track plays various samples.
The point being: In Preferences, if 'Use multi-core audio rendering' and 'Use hyper-threading audio rendering' are unticked I get two DSP bars. If they're both ticked I get one bar.
If I tick 'hyper-threading' only I get two bars, if I tick 'multi-core' only I get one bar.
It jumps up one bar on both settings slightly now and then. There doesn't seem much noticeable difference with having both boxes ticked or only 'multi-core' ticked, therefore I don't know if the hyper-threading setting is doing much in this case. However, this confirms what Mattius said earlier for sure. I'm using an i5 quad core laptop, I think it's about 7 years old now.
It would be really helpful by the way, if the preferences window was always on top or perhaps had it's own taskbar tab.
It was time-consuming to keep minimizing the 3 separated windows all the time when testing.
This would also help when checking / unchecking the boxes to see what works best for individual songs!
Here's a thought: The tool window always stays on top, so could the code that does that be put with the preferences window code?
The point being: In Preferences, if 'Use multi-core audio rendering' and 'Use hyper-threading audio rendering' are unticked I get two DSP bars. If they're both ticked I get one bar.
If I tick 'hyper-threading' only I get two bars, if I tick 'multi-core' only I get one bar.
It jumps up one bar on both settings slightly now and then. There doesn't seem much noticeable difference with having both boxes ticked or only 'multi-core' ticked, therefore I don't know if the hyper-threading setting is doing much in this case. However, this confirms what Mattius said earlier for sure. I'm using an i5 quad core laptop, I think it's about 7 years old now.
It would be really helpful by the way, if the preferences window was always on top or perhaps had it's own taskbar tab.
It was time-consuming to keep minimizing the 3 separated windows all the time when testing.
This would also help when checking / unchecking the boxes to see what works best for individual songs!
Here's a thought: The tool window always stays on top, so could the code that does that be put with the preferences window code?
- Boombastix
- Competition Winner
- Posts: 1929
- Joined: 18 May 2018
- Location: Bay Area, CA
I have seen other YouTube videos showing very similar numbers. The difference between Cubase/Studio 1/etc was pretty small but Reason managed less than half of the VSTs - so it'll be nice once we get it fixed.Arrant wrote: ↑31 Oct 2018I don't watch Hydlide's videos and don't really care what all the fuss is about.
However I did some testing of my own, as mentioned in another CPU thread that died off, and there is no doubt that CPU efficiency in Reason MUST improve. Here are my results again (on a PC):
Using the default patch in the Europa VST («FACT lead») I just put down some notes in a track and then duplicated the track until the CPU melted down. In Reason I managed 70 instances before the computer too slow message showed up.
In Reaper I got a whooping 172 (!).
That means (judging by this limited test) that Reason is 40% as efficient at running VST instuments as Reaper. Or, putting it differently, Reaper is almost 2.5 times more efficient than Reason. Incredible!
Try it yourself if you don't believe me.
Now I don't want to switch from Reason to Reaper, but it's obvious that this situation makes it very hard for Reason to compete in the DAW market.
10% off at Waves with link: https://www.waves.com/r/6gh2b0
Disclaimer - I get 10% as well.
Disclaimer - I get 10% as well.
Hi.Namyo85 wrote:I've started work on a new track. At the moment I've lined up about 28 different devices, mostly RE synths, including 1 Kong, 1 Echo, 3 audio tracks with various samples on them. 2 Utility RE's and 2 vst effects. I'll probably remove and change some of the synths as at the moment as I'm testing things. So far only 5 of the tracks have sequence data and one audio track plays various samples.
The point being: In Preferences, if 'Use multi-core audio rendering' and 'Use hyper-threading audio rendering' are unticked I get two DSP bars. If they're both ticked I get one bar.
If I tick 'hyper-threading' only I get two bars, if I tick 'multi-core' only I get one bar.
It jumps up one bar on both settings slightly now and then. There doesn't seem much noticeable difference with having both boxes ticked or only 'multi-core' ticked, therefore I don't know if the hyper-threading setting is doing much in this case. However, this confirms what Mattius said earlier for sure. I'm using an i5 quad core laptop, I think it's about 7 years old now.
It would be really helpful by the way, if the preferences window was always on top or perhaps had it's own taskbar tab.
It was time-consuming to keep minimizing the 3 separated windows all the time when testing.
This would also help when checking / unchecking the boxes to see what works best for individual songs!
Here's a thought: The tool window always stays on top, so could the code that does that be put with the preferences window code?
A 7 year i5 might not have hyperthreadin. The amount of cpus between laptops and desktops is absurd and configurations are very disparate. For example a lot of desktop i5s do not have ht. But low energy i5s for example the ones in Microsoft surface and other híbrids are dual core ht. So to get better conclusions you should check your cpu model and check on Intel or cpu benchmark if it has ht.
Cheers.
Sent from my WAS-LX1A using Tapatalk
Thank you for the reply and clearing it up, I later posted that I also found out that it indeed is the filter switch ! It's night and day using the 1 or 2 mode performance-wise.Blamsoft wrote: ↑31 Oct 2018It turns out that Epic Plucker has the F-16 filter quality on setting 2. This setting is more useful for getting a warm tone on a stereo pair. It generally shouldn't be used in a polyphonic patch.Cultor wrote: ↑29 Oct 2018
Do you by any chance own the Polymodular System ?
If you do curious what happends when you try the "Epic Plucker" Preset.
Factory bank can be downloaded here: http://blamsoft.com/rack-extensions/polymodular-system/
That preset alone absolutely destroys my CPU i7- 6700k
Great to see someone from PH involved in discussions like this one.
As you mentioned the CPU load information we get from the taskmanager doesn't actually tell us much about the actual load of a project inside of Reason.
Maybe something like a Performance Meter (this one is from Reaper ) could be useful tool to have.
For die hard nerds and RE developers in order to monitor different loads between devices.
I'm no Reaper user myself but from what I understand it shows the load on a track by track basis.
We have seen from different example that heavy load can come from devices we least expect it. (for example using a wrong setting in the F-16 filter)
A tool like this could improve the way we talk about performance and narrow down the origin of possible issues.
Yep, this would be interesting thought reason free routing could be hard to interpret but maybe we could have it just with device info?Cultor wrote: ↑01 Nov 2018
Great to see someone from PH involved in discussions like this one.
As you mentioned the CPU load information we get from the taskmanager doesn't actually tell us much about the actual load of a project inside of Reason.
Maybe something like a Performance Meter (this one is from Reaper ) could be useful tool to have.
For die hard nerds and RE developers in order to monitor different loads between devices.
I'm no Reaper user myself but from what I understand it shows the load on a track by track basis.
We have seen from different example that heavy load can come from devices we least expect it. (for example using a wrong setting in the F-16 filter)
A tool like this could improve the way we talk about performance and narrow down the origin of possible issues.
It could help a lot (specially if it came along with VST performance and a complete freeze process).
That huge difference comes as it's a somewhat unfair comparison. Reason is running all those VST in "real time" with a very small buffer (iirc batches of 64 samples) while Reaper is "cheating" by prerendering each VST in the background with much larger buffers. So a fair comparison would be to check if Reason can do 172 audiotracks.Arrant wrote: ↑31 Oct 2018Using the default patch in the Europa VST («FACT lead») I just put down some notes in a track and then duplicated the track until the CPU melted down. In Reason I managed 70 instances before the computer too slow message showed up.
In Reaper I got a whooping 172 (!).
That means (judging by this limited test) that Reason is 40% as efficient at running VST instuments as Reaper. Or, putting it differently, Reaper is almost 2.5 times more efficient than Reason. Incredible!
If the Props would implement an intelligent Autofreeze function into Reason the performance would be much better. But as there is CV implementing it is not as simple as in other DAWs.
- EnochLight
- Moderator
- Posts: 8412
- Joined: 17 Jan 2015
- Location: Imladris
This is something that most people aren't aware of. That said, it's little consolation for those who use Reason. It would be cool if Props could implement a similar approach, though.
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
They need to rewrite the CV-thingy, which may be the only way. CV did not need to be different from the frontside from in the beginning...
That's not cheating. That's smart programming.jam-s wrote: ↑01 Nov 2018That huge difference comes as it's a somewhat unfair comparison. Reason is running all those VST in "real time" with a very small buffer (iirc batches of 64 samples) while Reaper is "cheating" by prerendering each VST in the background with much larger buffers. So a fair comparison would be to check if Reason can do 172 audiotracks.
This is a block of text that can be added to posts you make. There is a 255 character limit.
Agreed. But thanks for pointing this out, I wasn’t aware. However turning that option(«anticipative FX processing») off in Reaper makes no difference to my test.fullforce wrote: ↑01 Nov 2018That's not cheating. That's smart programming.jam-s wrote: ↑01 Nov 2018That huge difference comes as it's a somewhat unfair comparison. Reason is running all those VST in "real time" with a very small buffer (iirc batches of 64 samples) while Reaper is "cheating" by prerendering each VST in the background with much larger buffers. So a fair comparison would be to check if Reason can do 172 audiotracks.
The impressive amount of buffering and thread usage options available makes me suspect the Reaper devs have a really solid knowledge about how to squeeze performance out of a multicore cpu though.
This is exactly a "trick" that I've already hinted at when I referenced the book from ID software in a previous post. It's what John Carmack did to make a smooth scrolling game on an ancient PC. By smart programming and optimize the shit out of it. That's the only proper way.Arrant wrote: ↑01 Nov 2018Agreed. But thanks for pointing this out, I wasn’t aware. However turning that option(«anticipative FX processing») off in Reaper makes no difference to my test.
The impressive amount of buffering and thread usage options available makes me suspect the Reaper devs have a really solid knowledge about how to squeeze performance out of a multicore cpu though.
This is a block of text that can be added to posts you make. There is a 255 character limit.
This depends highly on what your goal is: If you want to build a virtual Rack with very flexible and highly dynamic routing then the smart programming done in Reaper could be seen as "cheating". If you goal is to be just a DAW, then Reason has chosen a poor implementation of its modulation (=CV) aspect.fullforce wrote: ↑01 Nov 2018That's not cheating. That's smart programming.jam-s wrote: ↑01 Nov 2018That huge difference comes as it's a somewhat unfair comparison. Reason is running all those VST in "real time" with a very small buffer (iirc batches of 64 samples) while Reaper is "cheating" by prerendering each VST in the background with much larger buffers. So a fair comparison would be to check if Reason can do 172 audiotracks.
I hope, you can see why things are the way they are right now.
I call bullshit on that because CV has been a feature of Reason since version 1.jam-s wrote: ↑01 Nov 2018This depends highly on what your goal is: If you want to build a virtual Rack with very flexible and highly dynamic routing then the smart programming done in Reaper could be seen as "cheating". If you goal is to be just a DAW, then Reason has chosen a poor implementation of its modulation (=CV) aspect.
I hope, you can see why things are the way they are right now.
This is a block of text that can be added to posts you make. There is a 255 character limit.
- EnochLight
- Moderator
- Posts: 8412
- Joined: 17 Jan 2015
- Location: Imladris
You can call all the bullshit you want, but is it really shitty programming? Nope. It works.fullforce wrote: ↑01 Nov 2018I call bullshit on that because CV has been a feature of Reason since version 1.jam-s wrote: ↑01 Nov 2018This depends highly on what your goal is: If you want to build a virtual Rack with very flexible and highly dynamic routing then the smart programming done in Reaper could be seen as "cheating". If you goal is to be just a DAW, then Reason has chosen a poor implementation of its modulation (=CV) aspect.
I hope, you can see why things are the way they are right now.
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
Was Reason 1 a virtual rack or a DAW? Your turn.fullforce wrote: ↑01 Nov 2018I call bullshit on that because CV has been a feature of Reason since version 1.jam-s wrote: ↑01 Nov 2018This depends highly on what your goal is: If you want to build a virtual Rack with very flexible and highly dynamic routing then the smart programming done in Reaper could be seen as "cheating". If you goal is to be just a DAW, then Reason has chosen a poor implementation of its modulation (=CV) aspect.
I hope, you can see why things are the way they are right now.
Completely irrelevant question. CV is not a huge factor on performance (or lack thereof). If Reason was a DAW or not depends on what your definition of a DAW is. And it doesn't change anything. So because Reason is a DAW now it's OK to be slapped together like some unefficient Frankenstein monster?
This is a block of text that can be added to posts you make. There is a 255 character limit.
Well, I hope you're not a programmer, if "it works" is your only criterium.EnochLight wrote: ↑01 Nov 2018You can call all the bullshit you want, but is it really shitty programming? Nope. It works.
This is a block of text that can be added to posts you make. There is a 255 character limit.
- Faastwalker
- Posts: 2290
- Joined: 15 Jan 2015
- Location: NSW, Australia
Yes, it’s been directed into ‘The Kitchen’, a place for discussing ‘non Reason related stuff’. How’d that happen?! Anyway the whole world is going in a weird direction. I guess I can say that now we’re in the kitchen. Reason performance issues? First world problems. It doesnt really matter.
-
- Information
-
Who is online
Users browsing this forum: No registered users and 6 guests