Sample rate conversion

WaveLab
1

I am used to mastering in a mixing DAW and switched to Wavelab recently because it looked like it would be faster to render multiple file formats and other time saving features. But I can’t figure out how to change the sample rate when rendering from the montage. I’m used to being able to specify the sample rate and bit depth when bouncing form a DAW (Cubase, Pro Tools, Logic). I can change the bit depth in the render tab in Wavelab but I can’t see a way to change the sample rate. How do I do that?

Thanks.

2

image

1 Like
3

Thanks for this. I realise why it wasn’t working. I was using a preset that bypasses the master section for rendering.

4

And just remember that Resampling in the Master Section means that (to follow best practice) you’d also want to Dither in the Master Section after the Resampler, and then potentially add a true peak limiter between the Resampler and the Dither to catch any overs from the peak level changes that Resampling can/does create and then all of the Master Section settings must be manually saved and loaded for each montage and are not saved as part of the Audio Montage automatically like Clip Effects, Montage Track Effects, and Montage Output Effects.

For these reasons and more, I am not a fan of using the Master Section.

5

Ah I see, this is interesting @Justin_Perkins. I have been using the Output section to true peak limit dither. As you point out that won’t work for resampling. I noticed you resample using RX and bring it back into Wave Lab to change bit depth. Does that mean you are dithering twice (also in RX)?

6

It would be great if there was the option to resample in the output section. Then it could be saved as a preset and all the different file outputs could be done in the Render tab. It would save a lot of time and less to go wrong.

7

No, for me, everything stays floating point until I actually render 24 and 16-bit WAVs from my downstream montages. I only dither one time.

In other words, all the processing aside from dither gets locked in at 64-bit float and native sample rate by rendering a full WAV of the entire montage in one pass. This also helps mitigate any plugin rendering issues, often found at the very first and very last samples and cold starts and hard endings if you render track by track with lots of plugins going.

WaveLab then creates a new montage upon rendering with all markers and data carried over for quick work. Nothing needs to be redone. I then insert a 24-bit dither in the Montage Output section for if and when I want to render 24-bit/96k WAVs of each track. But to get 24-bit or 16-bit WAVs at 44.1k (or DDP)…

I downsample that 64-bit float and native Sample Rate WAV of the full montage with RX (I could also use WaveLab batch processor) to floating point 44.1k WAV and then use Custom Montage Duplicate to have WaveLab again recreate the montage at 44.1k sample rate.

Since I already inserted a 24-bit dither in the Montage Output, it’s already there and ready for 24-bit/44.1k rendering, or I could Save As… the newly created 44.1k version and change the dither to 16-bit and render 16-bit/44.1k WAVs or a DDP.

This way everything is recallable, and safely stored in the .mon file. For me, the Master Section is too dangerous to use and requires too much extra thinking and work. I don’t like it.

The other bonus is that I can analyze the 44.1k downsample before rendering the 44.1k master WAV files and decide if I want add something like Tokyo Dawn Limiter 6 GE using JUST the true peak mode set to -0.1 ceiling before the dither. In this mode, it only catches new peaks from the SRC process that go higher than -0.1dB or whatever you decide you want to do.

It looks like a lot in writing but I have things dialed in to do very fast and because of the things mentioned above, I don’t think I’d ever do it another way.

8

Yes that makes total sense. Thanks so much!

9

It’s not the first request for this but for total recall you’d still end up having multiple montages and doing multiple renders which would take time. With the method I use, only the first render takes some time (to lock in the plugin processing) and then from there, all other renders go VERY fast since dither is really the only plugin needed.

If I wanted to render 24-bit/96k WAV and also 16-bit/44.1k WAV, it would have to do all the plugin processing twice which would take double the time.

10

There was another similar thread about this here:

There is certainly room for some improvement in WaveLab but due the Master Section behavior, operation, and other things, I really feel that I have found the fastest and most flexible approach that doesn’t paint you into any corners and lets you render and any all formats while double checking the rendering accuracy and deciding if you want to manage peak level changes after SRC or not.

With a little practice and usage of shortcuts, it can go very fast and be an overall time saver with more control, despite being more actual steps.

11

Anyway, what PG showed is how to change the sample rate. Whether that solves more problems than it creates is a personal preference or opinion.

12

Hi!

As @Justin_Perkins describes it would be the best way to operate Wavelab since you have total control and total recall.

For me, I’m used to do all my creative work in a multitrack DAW so the audio I import into Wavelab already has all my processing locked in and what I’m looking for is basically just editing, PQ coding and to create all the deliverables I need while doing SRC and dithering in the same process… at the same time I’m not as control oriented as Justin and have not found any issues with my workflow of not using a secondary limiter after SRC in most cases.

After further experimenting I’ve settled on using render presets together with a few master section presets for all my rendering. That takes care of everything I need very quick and easy while not doubling or tripling my project sizes which would be the case if I used Justin’s approach. That thread explains more in detail what would be optimal for me but it comes at a sacrifice of less control but to me it’s a small sacrifice for not having to run a 30TB archive to a 10TB archive.

1 Like
13

Thanks. Yes I am rendering the whole montage as 64 float before I output different file formats and dither. If I send a DDP to a client or even MP3s from this montage and they want one song louder or a different gap, I guess I need to go back to the montage with separate clips, make the change and render the montage at 64 float again. It seems a bit long winded when making all those final adjustments. Or is there a quicker way?

14

There might be a shortcut but I’ve never been interested in one because in my view, it only creates potential issues and sort of misses the point.

The point of rendering the entire montage as one long file is to help mitigate any rendering issues from plugins, and to lock in the entire project as one long file.

Any shortcuts would involve throwing that first part out the window and then you still have to insert the newly render part somewhere right? That seems like more work and opens up room for error in a number of ways.

Plus, at least with the way my brain and workflow works, I want EVERYTHING needed to create the master in one master montage. Once you start parting it out for this little change, and that little change it seems way too easy to get confused about what’s going on, especially if you have to visit the project later.

No matter what change I make, big or small, one or many, the master montage gets a new version number (V2, V3, etc.) so that I always know the approved version number of the full project, and I can always open previous versions to see what might have been done if we need to revert to a previous version.

Also, I do a lot of stuff that goes to vinyl and cassette and once the digital master is approved, I don’t want to do any detective work about all these little shortcuts I did to create the final approved digital master so I can create any alternate formats. I just want to open the master montage of the final approved digital master, Save As… and then make the vinyl and/or cassette version.

I know a shortcut seems tempting even if the change is minor but my preferred method is to render the full project again. There are always emails to write, messages to send, and other tasks to do while the rendering is occurring. I also know some engineers that use a dedicated rendering computer so they can be doing other things while something renders if it will take time.

Maybe there could be a shortcut but for the reasons I mentioned and more, I think a shortcut in this area isn’t very helpful at all and only complicates the matter.

15

That all makes a lot of sense to me as a workflow. I can see it avoids so many potential mistakes. Thanks so much for explaining that!

16

Question, Do you dither when you capture analog ? I mean when recording from a R track , its 6496, and when you record most of the convertors are 24bit, which I’m thinking do I need to dither when capturing ?

17

This isn’t really a thing you can or should do. Dithering is for when you render/bounce/export your audio to a lower bit-depth than the original source audio or after any digital processing which increases what was 24 (or 16-bit) audio to 32-bit or 64-bit floating point (depending on the DAW and settings/options).

With a few modern exceptions (Sound Devices units), nearly all analog to digital converters are 24-bit so it makes no sense to record an analog source at 32-bit or 64-bit float. It will just be empty data. The bit-depth meter of WaveLab can help show you this.

I record back from analog at 24-bit but since I always have some additional digital processing after the capture from analog such as a digital limiter, or perhaps subtle EQ before the digital limiter, or whatever, I first render to 64-bit floating point (and native sample rate) to lock in all the digital processing and then I insert a dither plugin with the correct dither setting (24-bit or 16-bit) when I render my final master files such as WAV files of each album track.

These files are files that require ZERO additional processing. No fades, no sample rate conversion, no level adjustments. I only apply dithering and render to 24-bit or 16-bit as the VERY VERY VERY VERY last step. Very last.

I do apply dither as the very last plug-in before feeding my analog chain because even if the source audio is 24-bit, my pre-analog digital processing will increase that audio to 64-bit float and in theory, one could argue it is worth dithering to 24-bit before it goes analog via your 24-bit digital to analog converter.

But to answer your question, I don’t dither when I record/capture to digital from analog because the converter is 24-bit and I capture at 24-bit so it would be a waste of effort and not helpful to the audio.

WaveLab has a bit-depth meter that can be helpful in showing you what happens to 24 and 16-bit audio when digital processing is applied, and what is actually going on in your WaveLab session so yo can determine if dithering should be considered or not.

1 Like
18

Thank you Justin. Yes I understand, I follow exactly your video on analog and that kind of my own way as well when Capturing. but this is what I wanted to ask:

So after capture/record from analog back at wavelab, do you mean you render the files, or anyway its already in a folder automatically when capturing ? ( in my case I create a file under recording and capture etc and choose it when recording)

-Do you mean to do this when capturing through analog? I added a screen shot, like this or you meant at the end of the process when you do the in the box master etc

Screenshot 2024-03-25 at 12.03.56 AM
Screenshot 2024-03-25 at 12.03.56 AM1920×1080 166 KB

Thank you again for taking the time to replay
its much appreciated
Itay

19

No problem. When going analog, I insert a dither plugin on the Reference Track that is feeding my analog chain. This way it is after all Clip Effects and other clip changes like level etc. It’s the very last thing in my digital path before it goes to the output.

In theory, this reduces the bit-depth to 24-bit instead of truncating it from floating point to 24-bit

20

Cool I understand , so it’s on “track”.

And then at the end after capture ?
Rendering ?