Old 05-23-2020, 01:51 PM   #1
David Carlyon
Human being with feelings
 
Join Date: Feb 2019
Posts: 182
Default Reaper Sample Rate

Hi all, I wonder if there any experienced users who can answer this...

I noticed that in preferences i have my interface set to 96k, the project is set to 96k and yet, in Audio/Midi settings (on mac) my interface is set to 44.1.
When i render all track and take FX into a new take, i get the 'i' symbol, when i click on it i can see the source audio is at 96k and t is being resampled.

My question is this - can i actually run my session at 96k like this, with the interface at 44.1 and have all of the audio from start to finish (all track renders, freezes, renders including final mixdown) at 96k?

Surely if i render each track like this, the audio will be at 96k, then it is being resampled to playback in realtime at 44.1
but when if i then do a mixdown (which happens at 96k) surely those audio tracks will then be played back normally (not resampled) during mixdown?

Is this viable? my point is, i don't want to create mixdowns where all the audio is resampled, but if every time i render, it happens at 96k, surely it won't be resampled?

This could be a great way of working at 96k without actually working at 96k.
My plugins use significantly less resources, and i can run them without internal oversampling (less latency too)

Is anyone else doing this? Is it a bad idea in any way?
David Carlyon is offline   Reply With Quote
Old 05-23-2020, 03:17 PM   #2
valy
Human being with feelings
 
Join Date: Jan 2020
Posts: 1,927
Default

This has been discussed quite a bit recently...a forum search could be illuminating.

My two cents: In line with the concept of "no free lunch," I posit that there isn't an effective way to mix at 96k without everything being at 96k for the duration.
valy is offline   Reply With Quote
Old 05-23-2020, 03:57 PM   #3
serr
Human being with feelings
 
Join Date: Sep 2010
Posts: 12,562
Default

It will take more resources to sample rate convert on the fly.

An example where you might want to run a 96k project at 48k (for example) would be if you're working on the project on a laptop on the go where your built-in audio interface might only go up to 48k. You can set the hardware connection in preferences to 48k but keep the project set to 96k. This is something other DAWs will not let you do! But there's no free lunch as valy said. It WILL take more CPU power. Might be a moot point now. Built-in audio on laptops has done 96k for well over 10 years. But that's the premise behind the feature.

Reaper's feature set around this also gives you ammunition with interfaces that don't play well with others. (Or OS's that don't play well with others!) The ability to disable control from Reaper (untick the boxes on the Audio/Device page) and use other control panels, for just one example.
serr is offline   Reply With Quote
Old 05-23-2020, 05:51 PM   #4
David Carlyon
Human being with feelings
 
Join Date: Feb 2019
Posts: 182
Default

Thanks for the replies guys, i will have another search also. Saw one thread so far but that was from a while back.

I understand that it will take resources to convert the audio - but when i have everything in 96k my plugins tale up twice the CPU and thus ends up taking way more resources.

I suppose one workaround might be - do this trick and run the interface at 44.1/session at 96k whilst i am in the midi programming phase - lots of VST instruments and FX running. Then render all the channels, once everything is rendered to audio, flip the interface to 96k...
That could work pretty well actually!

I love Reapers flexibility. It is ultimately why i am here and why i can't go back to S1 or Ableton - as much as i love them both and they both have unique things about the workflow, Reaper just does so uch that they cannot touch!
David Carlyon is offline   Reply With Quote
Old 05-23-2020, 05:52 PM   #5
David Carlyon
Human being with feelings
 
Join Date: Feb 2019
Posts: 182
Default

One more question actually, does the audio still consume resources if the audio is muted?
David Carlyon is offline   Reply With Quote
Old 05-23-2020, 07:16 PM   #6
valy
Human being with feelings
 
Join Date: Jan 2020
Posts: 1,927
Default

Quote:
Originally Posted by David Carlyon View Post
I understand that it will take resources to convert the audio - but when i have everything in 96k my plugins tale up twice the CPU and thus ends up taking way more resources.
Quite right. But if you mix while listening to everything in 44.1k, then render everything at 96k...don't you think the end product will be different than what you heard while mixing?
valy is offline   Reply With Quote
Old 05-24-2020, 03:45 AM   #7
David Carlyon
Human being with feelings
 
Join Date: Feb 2019
Posts: 182
Default

Quote:
Originally Posted by valy View Post
Quite right. But if you mix while listening to everything in 44.1k, then render everything at 96k...don't you think the end product will be different than what you heard while mixing?
Yes it would a little bit. The main differences i notice (other than less aliasing on distortions etc) are from the oversampling filters in various plugins, so i would def need to check everything at 96k first.

Thats why i am thinking of doing this - working with my synths at 44.1 and rendering all single channels (using 'render track and take FX to new take')
and then switching to 96k, putting an instance of neutron/2c audio precedence and Breeze on each channel, along with my buss plugins and doing a mixdown.

Hopefully my computer will handle that. I will have done most of the CPU heavy processing already and rendering down will mean i dont have to run any of my synths in realtime at 96k either.

This could be a great half way compromise. Going to give it a try.
What do you think?
David Carlyon is offline   Reply With Quote
Old 05-24-2020, 07:30 AM   #8
valy
Human being with feelings
 
Join Date: Jan 2020
Posts: 1,927
Default

Quote:
Originally Posted by David Carlyon View Post
Yes it would a little bit. The main differences i notice (other than less aliasing on distortions etc) are from the oversampling filters in various plugins, so i would def need to check everything at 96k first.

Thats why i am thinking of doing this - working with my synths at 44.1 and rendering all single channels (using 'render track and take FX to new take')
and then switching to 96k, putting an instance of neutron/2c audio precedence and Breeze on each channel, along with my buss plugins and doing a mixdown.

Hopefully my computer will handle that. I will have done most of the CPU heavy processing already and rendering down will mean i dont have to run any of my synths in realtime at 96k either.

This could be a great half way compromise. Going to give it a try.
What do you think?
Well, as long as you A/B the mix and the render and are satisfied with both, I see no reason not to, I suppose.

Your case is a bit the opposite of mine: I could probably run my VSTi tracks at 96k initially and develop the track, but there's no way I could do the mixing stage at 96k due to the CPU-heavy plugins I use.
valy is offline   Reply With Quote
Old 05-24-2020, 08:47 AM   #9
David Carlyon
Human being with feelings
 
Join Date: Feb 2019
Posts: 182
Default

Quote:
Originally Posted by valy View Post
Well, as long as you A/B the mix and the render and are satisfied with both, I see no reason not to, I suppose.

Your case is a bit the opposite of mine: I could probably run my VSTi tracks at 96k initially and develop the track, but there's no way I could do the mixing stage at 96k due to the CPU-heavy plugins I use.
I should say i have a kind of three tier approach - Folders with tracks inside them. A folder for each 'instrument' (Kick, Bass etc) The tracks contain the 'layers' for each sound. Each layer has VST insts, followed by a few things - Slate VMR, SSL X-EQ, DMG Track comp and one of the bx consoles.
I do all my synth automation along with any eq/channel fx automation.

Then when i have rendered those down i do a second stage - this involves the most CPU heavy and oversampled stuff - mostly consists of various distortions and enhancers/spectral FX. It includes stuff like IK Tapes, MIA Spark, high latency plugs like soothe etc.
I generally do this as item FX. So i add these to the rendered items, render into a new take lane.

Then when all of my layers are mixed, i record them onto the folder tracks as audio.

For final mixdown i mostly just have 2cAudio Precedence and Breeze, Material Comp, and Neutron.

I know this sounds like a long way round, but i am in the flow of doing this and getting great results now.

I plan to build a very powerful computer this year and maybe then i can re asses.
One thing i love about doing this is that my first stage is 0 latency, which is great for writing, recording, using hardware synths etc.

One feature i haven't really used yet is freeze. Maybe that would help me quite a bit. Generally when i need to render i have been using render to take (made an action to render to take and take FX offline) but freeze could be something for me to consider.
David Carlyon is offline   Reply With Quote
Old 05-24-2020, 09:00 AM   #10
valy
Human being with feelings
 
Join Date: Jan 2020
Posts: 1,927
Default

The thing I dislike the most about Reaper's freeze implementation is that it removes the frozen plugins from view on the track. Makes it a lot more annoying to remember what you have frozen. You can use the Track Manager to help somewhat, but still. I wish it just left them in place and changed their color, similar to how it does when you offline an FX.

Better than not having freeze at all, though.
valy is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -7. The time now is 05:23 AM.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.