Feature request - Master Export

WaveLab
21

Hi PG!

Yes, sorry not an English native speaker here either so please excuse my lack of proper terminilogy and grammar. By Line-up problem I only mean that the individual samples are indeed changed at the very start and end of each file that should be an unbroken stream of audio and this could possibly generate a click or pop and this is by only using Wavelabs internal SRC and dither, not using any 3rd party processing. It’s a different result at those same sample points compared to applying SRC to an entire file if that makes sense?

Agreed that rendering the montage to a hi-res file like you describe and export individual titles from that this is the way to do it but if your main montage is a 96k one you would need to first render a hi-res version of that into a continous file including markers, then export individual titles from that and then render an SRC’d version of that down to a 48k and then to a 44k and export titles from that… and then you need to save everything in order to have recall. For me being used to the simple way Pyramix handles this I just find that the way Wavelab does this and as @Justin_Perkins from a lot of experience has come up with is the best and most fail safe way to do it… increases project complexity and file management so much compared to what I’ve grown used to and really takes away focus from doing work.

22

Hi Justin, I’m pasting the same response here as on Facebook just to have it all saved in one spot.

Yes, that’s exactly what I did and as Steve noted above as well, the resulting audio samples don’t line up when using the Batch Processor in this way.

I see what you’re saying about DDP generation and having processing locked in for doing sides for cassettes or vinyl and I still think it takes too much focus compared to doing the same operation in Pyramix. I hate brining that up but imo that system for exporting masters is just faster and more intuitive to me and is cleaner and potentially less cluttered as well depending on the complexity of the delivery at hand.

I hear you regarding additional processing when needed after the SRC and making a montage to save those settings in would be the only viable option and is lot more fail safe than loading Master Section presets as well. For me, I’d still prefer to do this with the Batch Processor if it worked in order to reduce project management.

My main issue is that I think the exporting and workflow in Wavelab is just PERFECT as long as you don’t need to apply any SRC to a Montage during final master generation and as you say it works great once you develop presets, shortcuts and use macros etc… but that to me kind of says that there’s room for improvement here so that you wouldn’t need to construct your own software within the software REAPER-style because it intuitively isn’t designed to do what you’re looking for if that makes sense? I do agree that your process is the most fail safe and customizable no matter what your needs are in any given situation.

The different montage variations you describe is what I used to do and still have trouble with, that even though they don’t take up any space I tend to end up with too many variations of virtually the same thing and unnecessary duplicates in audio as well. My main idea about a Montage is that it holds my editing, track orders for different releases etc and maybe some processing… and I want to be able to tell at a glance what is what rather than be confused by montage variations that are only there to hold different export settings and details of the same basic audio material and also that if you have 10 tracks in a 96k montage then you’d need to render that to one continuous file which means twice the storage, then you render an SRC’d 44k version of that which again increases project size again and then you export separate files from that while you also need to keep all previous renders and montage variations for recalls which again increase project size, where my view is that you should be good with basically the very first montage and the rest are just deliverables made from that very first montage so that if a change would creep in, that you don’t have to go back and re-render your entire chain of montages.

In Pyramix I have one single session-file referencing one body of audio material that never needs to be duplicated and I handle exports of separate tracks, complete sides for vinyl and DDP from the same 96k timeline at floating point while the system handles all SRC and dithering plus file naming and folder generation for me without the need for additional session files, extra rendering stages etc…

As Monte said, WaveLab is better at about everything else.

1 Like
23

I have turned off comments on the FB thread so it can be just discussed here where it’s more likely something can be done about it.

I think my analogy of Pyramix being like an automatic transmission and WaveLab like a manual transmission is good.

I prefer to be able to manage every detail and nuance with all rendered formats. There are just too many details to leave it behind closed doors., even if it’s easier.

There are many reasons but one that hasn’t been mentioned…sometimes songs or projects have peaks over 0dB after SRC and some don’t. Sometimes I decide to manage it in the 44.1k derivative of the montage before rendering 44.1k versions (or DDP), sometimes I don’t. I don’t see how an automatic rendering process could decide this on the fly and do the right hthing.

Also, I rarely render ALL the master formats at one time. I typically create upload 24-bit/44.1k WAVs to my gapless/lossless web-player for clients to approve, and also send a DDP w/DDP Player for those who are inclined to use that.

Once the project is approved, then I’ll go back and render the 24-bit/96k, reference mp3, and then make a derivative for any vinyl, cassette, instrumental masters, radio edits, etc.

Maybe you can create a video of how Pyramix does it for PG to get some ideas of some things that could be improved in WaveLab but I personally am not feeling like this is a bottleneck of WaveLab in any way.

The way it works is a big reason why I chose to use WaveLab over any other Mac mastering DAW option such as HOFA, Sonoris, Studio One (project mode), SoundBlade (RIP), Triumph, REAPER, etc. I also like it so much I have ZERO temptation to try to learn and tolerate Windows.

24

Yes, the workflow could possibly be implemented in a few different ways within WaveLab or the Batch. Just to describe the master export procedure of Pyramix if someone isn’t familiar with it;

You have an album project which is basically the same thing as a Montage and for the sake of discussion you have the need to export this 96k project with individual titles to separate 48k and 44k files with a few plugins on individual tracks and the master output… I’ll skip some details here, but what happens is that Pyramix first renders the project to their own MTFF-format at the highest resolution the project is in… this being 32bF96k in this case and this file then gets put through their very simplified but purposeful version of the Batch Processor which only job is to correctly output a master file into different formats. In this Batch Processor you choose your desired naming schemes and sub-folder generation quite similar to Wavelab, and then you have an output window where you choose exactly what you want this 32bF96k project output as, and for each output you can select the format (wav, ddp or different encoders), sample rate, bit depth, dither settings, any additional gain changes etc… and if you want individual tracks or continuous or both exported for each output with different naming schemes applied for both options.

The system goes to work and then you’re done.
No need for additional montages, duplicate files etc… exactly what can be done with the WaveLab Batch Processor only difference being that it can’t use a complete hi-res file with markers instead to generate everything from?

The Wavelab way in the first step is exactly the same… your hi-res montage gets rendered into a continous hi-res file but then the similarities stop. Then you need to apply SRC to 48k and render this into a third montage and then apply dither inside that montage and you need to go back and render the 96k timeline into a fourth 44k montage and then you need to apply dither to that and then apply different export presets to every montage and render these out.

Now, if an artist comes back and decides after some thought that tracks 7 and 8 should actually switch places…

  • In Pyramix you open the original project switch these tracks around, hit Generate Master, you click the output formats you want again in the Batch and your new set of files for all delivery formats are ready to go within a minute or two.

  • In Wavelab you would need to open the original Montage, switch the tracks around, re-render the hi-res file replacing the old one in the second montage having to find the previous file paths as well, open the second 96k montage, re-render SRC to 48k replacing files in the 3rd montage while finding this file path, go back to the second montage and re-render SRC to 44k and again find file paths and replace files then open montages and re-apply export presets for each montage and export these to the right locations…

There are so many more steps involved that command focus and you need to remember, re-apply and adjust things in way that is much more prone to mistakes IMO.

25

Yes, PG this works great. Only problem is that as @JSMastering mentioned that you would need multiple master section presets for every sample rate and bit depth and when trying this here it’s still quite a bit back and forth process since you would need to…

  • Load Master Section Preset
  • Go into render
  • Select render preset
  • Update file path
  • Render
  • Load Master Section Preset…

If Wavelab could instead render a hi-res in the background and instead of opening a new montage, just open a window letting me select all the ouptut formats that I wanted and one naming scheme in a single go it could be done way simpler as in two steps rather than a dozen maybe?

26

Yes, good analogy there!

If you need to control +0db peaks after SRC, Pyramix will not be your friend, but coming down from a 96k or TP-limited source generally I can’t say I’ve had any issues with overs.

Since the process is so easy in Pyramix, if you’re going to render one thing you could just as well render everything you need all at once. If all is approved from the client, then you already have everything you need basically and if you need to change anything it’s just quick and easy either way to do it all again.

I’ll see what I can cook up… screen recording on Wintendo is so intuitive and easy isn’t it. :stuck_out_tongue:

27

Exactly.

28

Yea, again, I think the main issue here is that when using the batch converter to convert individual tracks from a continuous piece of audio, there are sample mismatches at each transition. When only sample rate conversion and internal dither is applied.

I think Justin’s detailed approach is a superior, detailed, and more failsafe way to work in Wavelab (especially with gapless audio) but that glosses over the fact that there is a separate issue with the batch converter.

Personally I use both methods depending on what I’m doing.

I can’t comment on how Pyramix does things but I do know that the RX batch converter doesn’t suffer from this issue. You drag the files in, sample rate change and dither is applied, resulting tracks line up perfectly.

29

I do all my processing and sample rate conversion BEFORE I create a master. Solves a lot of the problems you all seem to be stuck with. It sounds like you are trying to save time but in reality it is final project that sufferers AND you have to do a lot of QC which seems to defeat the whole way you are trying to work.

I want my mastering to be precise and I don’t want to worry about some “glitches” along the way. FWIW

1 Like
30

Same here. This is why I use the method I use. Precision and no glitches.

The problem is, we pretty much always need to deliver EP/album masters (and even singles) at various sample rates which isn’t hard to do thanks to the “Create Custom Montage Duplicate” feature. It allows you to recreate a montage at a different sample rate after all your processing is locked in (or not if you prefer which I don’t) and everything else about the montage such as marker names, marker placement, song spacing, etc. is 100% identical and all your rendered master formats are totally cohesive.

It’s one of my favorite things about WaveLab and a major part of why I use it instead of other options on Mac.

The topic of this thread is about adding an automated shortcut for this process similar or exactly like Pyramix can do but I personally think it creates just as many potential issues as it solves.

Automatic transmissions are nice sometimes but if you want control and precision of your car in every situation, manual transmission is the way to go…and WaveLab has all the features needed to make it a fast and accurate process.

31

Thanks as always…

32

I hear what you and @Thomas_W_Bethel are saying, but I’ve not had any issues with this method of working now for the past 3-4 years even though it doesn’t offer you total control in the same way WaveLab does in this regard and I can’t see I witnessed any complaints downstream either from working this way compared to WaveLab. WaveLab is pretty much the only software that does give you this amount of control and flexibility and despite this doesn’t seem to be a major player in the big studios that instead choose to run Sequoia and/or Pyramix? From what I’ve seen they just don’t seem to care that much about this amount of control and rather drive an automatic, but to each their own I guess.

Wavelab is exactly the way I want a piece of mastering software to work and I’ve worked on both Sequoia and Pyramix for a few years. The only thing that I’m missing is to have an intutitive and easy export section that doesn’t require as much user attention to produce deliverables, at least so you have a choice about the amount of control you’d like to exhibit.

I’ll go ahead and create a workflow video of a Pyramix session soon as I can so @PG1 can see if it’s of any relevance to implement any similar in the future.

33

I have seen TP levels change after SRC, which is enough to irk me. In general, I still don’t believe TP overs are a problem in practice. But, it has happened.

But…that’s actually why I suggested it the way I did.

I haven’t worked in Pyramix, so I could be wrong. But, adding some processing presets to WL’s already excellent multi-format export presets could wind up even better than how I understand Pyramix’s render workflow.

What I’m suggesting is essentially wrapping up Justin’s entire export workflow into a couple clicks that you set up ahead of time. Maybe I’m misunderstanding how much he’s tweaking each one of those montages for each album…but the way it works out for me, I don’t even listen to anything between the “initial montage render” and the files that eventually wind up in the project folder. I’m just mechanically doing the same thing unless I eventually find a problem…which means that the whole process could be automated. Then, if you do find a problem, you go back to the “initial montage render” and solve the problem either manually if it’s specific to that album or by tweaking your expanded, processing-included, multi-format render preset if it’s consistent.

The thing about including the result of the “initial montage render” in the original montage on a special “result” track…that’s more of a pipe dream and not necessary. It would just simplify file management a tiny bit more. Whether that slight simplification is worth it taking longer to open the montage you eventually render all the master formats from…that’s a coin toss as far as I’m concerned. But, I don’t think there’s another downside from a user’s perspective.

1 Like
34

@PG1 - Please find a short video clip under the link showing a very basic export procedure using Pyramix outputting 16 and 24 bit WAV-files and MP3 and AAC, + DDP export no SRC involved but as you see the procedure is the same.

What you see is the editing window with a sequenced CD master on one track and a vinyl master on the second track…it’s not very different to how it works in Wavelab in this view. Everything is quite self explanatory I think. The only thing that might stick out is when generating the master file which is done to Merging MTFF format which renders the mixer and puts it in a hi-res container with meta-data like when you render a whole montage and choose to create a new montage from the result carrying over markers etc…it’s this file that Pyramix uses for all the following conversions as you see in the Album Publisher that the Input reads 44k 32bf which is what the MTFF in this case was written at.

As you can see, you create different output formats/presets and if the input sample rate is different to the output the pre-chosen SRC algorithm gets engaged taking care of the conversion, any gain change and then dither. The result is then fed to the file naming section and output where you create naming presets.

If you have any questions please go ahead and ask.

1 Like
35

Indeed, in the video of Plec81 I don’t see any limiter option together with the SRC option. Strange.

36

I have had something like this in mind for a while. This discussion reinforces my wish to do something about it. But my todo list is pretty filled up these days :slightly_smiling_face:
Thanks for the video.

2 Likes
37

Exactly, all processing has to be carried out in the mixer/editor prior to export.

The only processing choices you get with the Album Publisher are;

  • Format
  • Sample Rate
  • Bit Depth
  • Dither Noise Shaping
  • SRC Algorithm
  • Gain Change in hundreds of a db after SRC, before Dither.

If you would want do catch peaks after SRC with Pyramix like @Justin_Perkins proposes you could change the sample rate of the project to whatever you want before exporting and add a limiter on the track or master bus. But that would only work like expected with a similar project to that in the video that has no other processing added. The Merging HEPTA SRC works in realtime to make sure all audio matches the sample rate of the project at all times, because the project can be changed on the fly to whatever SR you want. So even if you’re running a 44k mastering session with plugins and you wish to run it at 96k instead, you just change the project sample rate to 96k and the HEPTA algorithm upsamples all audio in realtime before hitting the mixer.

The other solution would be to render a hi-res file, create a new 44k project from that which would downsample the audio in realtime and add a limiter to the end before exporting it again, but that would but you in the same multiple montage world as Wavelab.

38

What I got out of your description was pretty much what you showed, Plek, so kudos on speaking English much better than I speak…what was that? German? Danish? I’m not even sure.

Anyway…the method Justin developed is the best one we have right now, and it does accomplish all of those things. The price is having more montages. If you’re wiling to deal with the (IMO Convoluted) WL Project format, you can at least open them together in different tabs. So, it’s just more clicks.

I would welcome simplifying it to fewer clicks in the future, but there are other things I want more…some of which I know are on PG’s todo list. So, I’m happy to wait and see.

39

Haha, thanks JS! That’s Swedish btw. :smiley:

I’ve been getting into W11 now for the last couple weeks and boy have I missed using Wavelab properly these last 3-4 years. Been experimenting with different ways to accomplish more or less the same things and Justin’s method is very well thought out and I agree it’s the best way with the current set of tools and is more or less the same method I came up with and used some years ago but grew tired of it thinking that there has to be a better way to meet my more limited needs.

To clarify, my workflow implies that I mostly use a separate DAW for all my creative work due to the routing and processing capabilities. It started with a combination of Nuendo and Sequoia many years ago and as Wavelab got more and more advanced eventually it replaced Sequoia for me. Today I use ProTools for all my processing/creative work and only use a mastering DAW like Pyramix or Wavelab for editing, meta-data and exporting deliverables from, so the big difference to Justin’s workflow is that all audio I import into a fresh Wavelab session already has all processing locked in which makes it especially frustrating for me being forced to re-render things over and over. When I’m done editing and sequencing I just want to export easily and maybe do some SRC on the fly as well and then move on to the next thing. Even now when I’m using it again for exporting I find I’m quickly getting frustrated by having to change and make presets for every single thing and micro manage stuff on a level that I see as time consuming and unnecessary for my needs. Now, Wavelab exports things 10x faster than Pyramix just that it takes up more focus and mental energy to get the same thing done. I just did a project that required I re-render a few montages into sub-montages and the project naturally grew to triple it’s size, and for my usual workflow the trade-off there just doesn’t make sense to me.

I can’t wait to see what PG has in mind for future updates!