Referencing Audiofiles like Samples
When using imported audio tracks Reason´s song filesize increases dramatically because every audiofile is included within each save.
Saving sequentially quite often here, saying the used diskspace for songfiles add up easily into GB when working with audiotracks.
The filesize has actually quite often kept me from continuing working with imported audio so far.
Working with synthesis/samples only keeps it a few MB at max per songfile.
Simply including an optional reference like it´s done with samples or vice versa so to speak would easily solve this issue.
I could simply reference the files during the working process while including them once a song is done for example.
Saving sequentially quite often here, saying the used diskspace for songfiles add up easily into GB when working with audiotracks.
The filesize has actually quite often kept me from continuing working with imported audio so far.
Working with synthesis/samples only keeps it a few MB at max per songfile.
Simply including an optional reference like it´s done with samples or vice versa so to speak would easily solve this issue.
I could simply reference the files during the working process while including them once a song is done for example.
- stillifegaijin
- Posts: 251
- Joined: 27 Oct 2020
No. It's a file organization nightmare. No, No, No. Use Pro Tools if you wanna deal with that.
And do you actually have audio tracks references in multiple songs? Still...No.
And do you actually have audio tracks references in multiple songs? Still...No.
TBH, when i import audio files, i just get them in and delete the source material if i don't pretend to use it on other projects. I even think other daws, copy the files, specially if they have to oversample them or change bit rate to the project's bit rate. So most times, there's an import and conversion to include the file in the project folder.
But my grips with the reason file format is not actually that. My problem with this format is more related to version management. Most of what you do in a project is on tens of MB at most (midi data, sequencer data, parameters, modulations, automation, etc), and you still need some sort of versioning of those changes and if your software doesn't do that you need to create snapshots of your project. When you do this with an application like cubase, you can manage that by creating a new version, and the audio is not copied. In reason, need a snapshot? It copies audio. Need a new mix version? Copy audio. Your client asked a source a second down and you need to keep it? Copy audio. Specially on client's stuff, where i like to keep snapshots of every bigger change (even with cubase's version management it's a good option to keep it that way), i have client's projects that should take about 2 gb and are more than 15 because of freaking versions (i usually keep the base project unmixed, the first mix, and any big change, so assume 3 to 5 versions of a 2 gb project plus the first one)!
This is something i really hope gets reviewed, I've heard logic allows both project structures (consolidated and separated) so lets see.
But my grips with the reason file format is not actually that. My problem with this format is more related to version management. Most of what you do in a project is on tens of MB at most (midi data, sequencer data, parameters, modulations, automation, etc), and you still need some sort of versioning of those changes and if your software doesn't do that you need to create snapshots of your project. When you do this with an application like cubase, you can manage that by creating a new version, and the audio is not copied. In reason, need a snapshot? It copies audio. Need a new mix version? Copy audio. Your client asked a source a second down and you need to keep it? Copy audio. Specially on client's stuff, where i like to keep snapshots of every bigger change (even with cubase's version management it's a good option to keep it that way), i have client's projects that should take about 2 gb and are more than 15 because of freaking versions (i usually keep the base project unmixed, the first mix, and any big change, so assume 3 to 5 versions of a 2 gb project plus the first one)!
This is something i really hope gets reviewed, I've heard logic allows both project structures (consolidated and separated) so lets see.
I’m with mcatalao on this, always have been with regards to Reason and audio (ranted about it since just after Record was released!).
So far it’s fallen on deaf ears, and many folks have been suggesting this for well over 10 years now.
It would actually be far more efficient than the Pro Tools model IMO because it would keep things more organized as long as it included a few key functions.
•Versions (saving multiple version within the same song file)
•Import/Export audio data minus song data (already in place for the most part)
•Import/Export song data minus audio data
This last point is the key to making it work well IMO. It would allow you to import mixer channels/settings from one song to another. It would allow making/saving mixer templates when mixing multiple songs with similar track layouts. It would allow you to import channel data from one mix version into another to allow consolidating the best parts of various mixes.
Versions is such a basic concept and mirrors the way most of us work (or want to work), and this includes workflows beyond delivering multiple mix versions.
•Creative exploration (breadcrumb trail): It’s extremely handy to be able to explore options when building tracks, and in many cases the changes go well beyond what would be easy to explore with ‘undo’. Similarly, data import/export would be necessary when exploring these options in order to consolidate the best parts of various versions.
•Overdubbing on complex songs: to reduce latency it would be very simple to create an “overdub” version consisting of a stereo audio file, add your overdubs at the lowest latency settings, then consolidate those new tracks back into the main version.
•Bouncing down stacked tracks: To make mixing simpler it’s often handy to bounce down stacked tracks such as vocals, but you don’t want to loose access to the original individual tracks. Versions makes this workflow simple by creating a version with the stacked vocals bounced to a single stereo track/channel while still giving near instant access to the originals without requiring keeping track of separate song files and doing copy/paste routines to keep things in ‘sync’.
•Collaboration: versions would allow much easier workflow when collaborating by allowing each person to keep their original version and also import changes and additional tracks back into the song file. Everything in one place, never loosing any changes or alternate versions of the song/arrangement and able to mix/match all ideas to explore all your options. When receiving new files from your collaborator all you do is select “import new” and it would automatically create a new version with the new tracks/ideas imported in place. That means you don’t need to exchange the entire song file with all the audio files each time you share ideas, you simply “export new” and send THAT file, which is then imported by your collaborator.
•And finally, Mix Versions: this is probably what most folks think about when speaking of a ‘version’ system, and it’s handy but hardly the only reason for wanting this feature. For those who are not familiar, every client based project I’ve worked on requires multiple mix versions. It’s also useful even when working solo, but at preset many folks don’t bother because it’s actually MORE complicated to try to keep track of multiple separate song files (each containing redundant audio files) than not.
For example, with a good ‘version’ system, you could manage multiple mixes/arrangements (vocal up/down, full version and edited version, etc) making changes on one and having them easily integrated into the others (could be an automatic function or a manual function). This way you could do things like bump up the drums on one version and easily copy it across to the others, and much more.
I could go on, but I’ll stop here - I keep posting this idea when it comes up in hopes that eventually it may affect a change - but not holding my breath!
So far it’s fallen on deaf ears, and many folks have been suggesting this for well over 10 years now.
It would actually be far more efficient than the Pro Tools model IMO because it would keep things more organized as long as it included a few key functions.
•Versions (saving multiple version within the same song file)
•Import/Export audio data minus song data (already in place for the most part)
•Import/Export song data minus audio data
This last point is the key to making it work well IMO. It would allow you to import mixer channels/settings from one song to another. It would allow making/saving mixer templates when mixing multiple songs with similar track layouts. It would allow you to import channel data from one mix version into another to allow consolidating the best parts of various mixes.
Versions is such a basic concept and mirrors the way most of us work (or want to work), and this includes workflows beyond delivering multiple mix versions.
•Creative exploration (breadcrumb trail): It’s extremely handy to be able to explore options when building tracks, and in many cases the changes go well beyond what would be easy to explore with ‘undo’. Similarly, data import/export would be necessary when exploring these options in order to consolidate the best parts of various versions.
•Overdubbing on complex songs: to reduce latency it would be very simple to create an “overdub” version consisting of a stereo audio file, add your overdubs at the lowest latency settings, then consolidate those new tracks back into the main version.
•Bouncing down stacked tracks: To make mixing simpler it’s often handy to bounce down stacked tracks such as vocals, but you don’t want to loose access to the original individual tracks. Versions makes this workflow simple by creating a version with the stacked vocals bounced to a single stereo track/channel while still giving near instant access to the originals without requiring keeping track of separate song files and doing copy/paste routines to keep things in ‘sync’.
•Collaboration: versions would allow much easier workflow when collaborating by allowing each person to keep their original version and also import changes and additional tracks back into the song file. Everything in one place, never loosing any changes or alternate versions of the song/arrangement and able to mix/match all ideas to explore all your options. When receiving new files from your collaborator all you do is select “import new” and it would automatically create a new version with the new tracks/ideas imported in place. That means you don’t need to exchange the entire song file with all the audio files each time you share ideas, you simply “export new” and send THAT file, which is then imported by your collaborator.
•And finally, Mix Versions: this is probably what most folks think about when speaking of a ‘version’ system, and it’s handy but hardly the only reason for wanting this feature. For those who are not familiar, every client based project I’ve worked on requires multiple mix versions. It’s also useful even when working solo, but at preset many folks don’t bother because it’s actually MORE complicated to try to keep track of multiple separate song files (each containing redundant audio files) than not.
For example, with a good ‘version’ system, you could manage multiple mixes/arrangements (vocal up/down, full version and edited version, etc) making changes on one and having them easily integrated into the others (could be an automatic function or a manual function). This way you could do things like bump up the drums on one version and easily copy it across to the others, and much more.
I could go on, but I’ll stop here - I keep posting this idea when it comes up in hopes that eventually it may affect a change - but not holding my breath!
Selig Audio, LLC
I like how that's implemented in Cakewalk. I can have audio stems of full song length just referenced from another location on the disk, then automatically chop them into clips on the tracks based on silence, move the clips around, and all that without creating any copies or touching originals.
Not to mention multiple mix variants in one project that I can switch between in a second.
Not to mention multiple mix variants in one project that I can switch between in a second.
That first feature is great, until you move or delete the referenced file since it is in no way saved with the song file! Not sure I see that as a “feature” I would have any use for since I like all related audio files in one place.orthodox wrote: ↑25 May 2022I like how that's implemented in Cakewalk. I can have audio stems of full song length just referenced from another location on the disk, then automatically chop them into clips on the tracks based on silence, move the clips around, and all that without creating any copies or touching originals.
Not to mention multiple mix variants in one project that I can switch between in a second.
Wish Reason Studio had listened to my “versions” request 12 years ago because they would have been first to market with the feature, leading the pack instead of (not)? following…
Selig Audio, LLC
Of course I have other options as well. I can bounce the clips in place individually to the project directory or bundle them within a big project file like Reason does.selig wrote: ↑25 May 2022That first feature is great, until you move or delete the referenced file since it is in no way saved with the song file! Not sure I see that as a “feature” I would have any use for since I like all related audio files in one place.orthodox wrote: ↑25 May 2022I like how that's implemented in Cakewalk. I can have audio stems of full song length just referenced from another location on the disk, then automatically chop them into clips on the tracks based on silence, move the clips around, and all that without creating any copies or touching originals.
Not to mention multiple mix variants in one project that I can switch between in a second.
Curious then what use case for linking audio files like samples, and what advantages to this approach that make it superior over copying (other than saving a few seconds)?
Selig Audio, LLC
If I generate the stem files with some other program (Reason), it's convenient to reference them where they are so I can replace them later without modifying the project that uses them.
Also, multiple projects can reference the same file(s).
Like said, it´s less about technicality in this case, it´s mainly about FILESIZE
Imagine working with GB of video footage and saving it all along with every save file. Harddrive would be full in an instance. Naturally it´s done referenced.
It´s not quite as bad with audio, yet the same circumstance applies.
Regarding versions, sounds interesting and would aswell merely contain the original audiofile once. Then it´s a lot more effort to create and a simple audio file referencing would at least solve the filesize quite easily.
Imagine working with GB of video footage and saving it all along with every save file. Harddrive would be full in an instance. Naturally it´s done referenced.
It´s not quite as bad with audio, yet the same circumstance applies.
Regarding versions, sounds interesting and would aswell merely contain the original audiofile once. Then it´s a lot more effort to create and a simple audio file referencing would at least solve the filesize quite easily.
Where is the difference to working with samples after all? Simply put a footage folder into your songfolder and done. That´s exactly how I do it with videoediting.
Aswell it´s an OPTIONAL saving feature, the inverted principle of sample saving where reference is default but samples can aswell be included in the song file.
Saying, like stated before, I could work with referenced audio during the proccess and include the audiotracks with the songfile when having finished.
In addition in After Effects for example there is aswell a save with a "gather all footage" feature where it copies every file imported into the project from wheresoever and puts it into a single footage folder that is referenced from then on. Easy project saving/moving.
Aswell it´s an OPTIONAL saving feature, the inverted principle of sample saving where reference is default but samples can aswell be included in the song file.
Saying, like stated before, I could work with referenced audio during the proccess and include the audiotracks with the songfile when having finished.
In addition in After Effects for example there is aswell a save with a "gather all footage" feature where it copies every file imported into the project from wheresoever and puts it into a single footage folder that is referenced from then on. Easy project saving/moving.
stillifegaijin wrote: ↑25 May 2022No. It's a file organization nightmare. No, No, No. Use Pro Tools if you wanna deal with that.
And do you actually have audio tracks references in multiple songs? Still...No.
In terms of collaboration it´s much easier to exchange files once both have the same original files on their harddrives because it´s merely the songfile itself having to be sent.
Working with proxies aswell allows simple outside editing by whosoever and the file within the song can simply be updated to a new version by replacing the original file like orthodox mentioned.
From what it sounds like Cakewalks way of handling is basically what I´m talking about. You can simply choose between referenced or integrated saving.
Working with proxies aswell allows simple outside editing by whosoever and the file within the song can simply be updated to a new version by replacing the original file like orthodox mentioned.
From what it sounds like Cakewalks way of handling is basically what I´m talking about. You can simply choose between referenced or integrated saving.
I get the first part, which is why I’ve made the suggestion to support that feature with ‘versions’. It’s the import audio/data part that handles this function, though it is not commonly needed in my experience. Examples would be doing noise reduction on a recorded track where you need to process it externally and then re-integrate it with the original file That said, every DAW would have to FIRST consolidate the file before this could work if the audio track consisted of more than a single clip, which is very common in modern production. For example, when you punch in using Pro Tools, you create new audio files.
So even in Reason as it is today it’s not a complicated process to do this. You export your audio as a single file, process it, then re-import it. Again, the only way this works differently in other DAWs is when the audio is already (or later) consolidated into a single audio file, right? Or am I missing something here?
I’ve worked many years and never needed multiple projects to reference the same files. What is the use case for this?
Selig Audio, LLC
Possibly it´s because you work with a different type of audio files like vocals or instrument recordings?
I created a samplelibrary from the modular aswell with longer recordings in it.
I like importing these and chop them for merely using pieces, not for one song only. I could use some parts of the audiofile in one song while I could use other parts in a new one. Also using identical recordings but differently used or altered across songs is possible. Just like I do with samples, hence referencing comes in handy. And reimporting isn´t easily done at all once you started chopping. The mentioned versions-feature would lack this option aswell as it would merely reference internally.
I created a samplelibrary from the modular aswell with longer recordings in it.
I like importing these and chop them for merely using pieces, not for one song only. I could use some parts of the audiofile in one song while I could use other parts in a new one. Also using identical recordings but differently used or altered across songs is possible. Just like I do with samples, hence referencing comes in handy. And reimporting isn´t easily done at all once you started chopping. The mentioned versions-feature would lack this option aswell as it would merely reference internally.
I’m sorry, but I must be super dense this morning as I still don’t see any advantage other than saving disk space, which is no longer an advantage if you process the audio externally like is also being mentioned.moofi wrote: ↑29 May 2022Possibly it´s because you work with a different type of audio files like vocals or instrument recordings?
I created a samplelibrary from the modular aswell with longer recordings in it.
I like importing these and chop them for merely using pieces, not for one song only. I could use some parts of the audiofile in one song while I could use other parts in a new one. Also using identical recordings but differently used or altered across songs is possible. Just like I do with samples, hence referencing comes in handy. And reimporting isn´t easily done at all once you started chopping. The mentioned versions-feature would lack this option aswell as it would merely reference internally.
Speaking of which…Re-importing after chopping, though rarely needed, is easy to accomplish with the versions suggestions I’ve made over the years (and one of it’s advantages). By time stamping the clips on the timeline it’s simple to export/process/import and get things right back to where they were. But again, this is rarely needed but handy if you do.
The ONE place where sharing a file IS advantageous is if you need to make changes to the original file and have it apply to ALL uses, though I can’t come up with a reason for needing to do that (is that why you’re wanting that approach?).
Selig Audio, LLC
No, I don´t want to go back with versions in that case, I would like to keep the chopping but swap the file it´s done to.
Say I´m working with a 30 second recording or sample for that matter. I take that sample and chop it in the sequencer into parts I would like to use and simply rearrange/multiply those parts. Then I got an idea how the original recording can sound even better like processing it through a pedal, a speaker, a module or whatsoever.
Aswell, the first recording could possibly be a dummy, like a not yet "mastered" version, then it´s mastered and I can simply swap the original file to the mastered version.
Now how to simply replace the chopped file? Unless the versions version remembers the chopping and is able to apply those chopping changes to a newly imported file, I don´t rellay see how it could solve this issue. And like said, implementing references is pretty easily achieved, while anything like applicable protocols or recorded actions like in Photoshop, not only complicate the working process but are over the top for the task where a simple swap is all that´s needed.
Say I´m working with a 30 second recording or sample for that matter. I take that sample and chop it in the sequencer into parts I would like to use and simply rearrange/multiply those parts. Then I got an idea how the original recording can sound even better like processing it through a pedal, a speaker, a module or whatsoever.
Aswell, the first recording could possibly be a dummy, like a not yet "mastered" version, then it´s mastered and I can simply swap the original file to the mastered version.
Now how to simply replace the chopped file? Unless the versions version remembers the chopping and is able to apply those chopping changes to a newly imported file, I don´t rellay see how it could solve this issue. And like said, implementing references is pretty easily achieved, while anything like applicable protocols or recorded actions like in Photoshop, not only complicate the working process but are over the top for the task where a simple swap is all that´s needed.
selig wrote: ↑29 May 2022I’m sorry, but I must be super dense this morning as I still don’t see any advantage other than saving disk space, which is no longer an advantage if you process the audio externally like is also being mentioned.moofi wrote: ↑29 May 2022Possibly it´s because you work with a different type of audio files like vocals or instrument recordings?
I created a samplelibrary from the modular aswell with longer recordings in it.
I like importing these and chop them for merely using pieces, not for one song only. I could use some parts of the audiofile in one song while I could use other parts in a new one. Also using identical recordings but differently used or altered across songs is possible. Just like I do with samples, hence referencing comes in handy. And reimporting isn´t easily done at all once you started chopping. The mentioned versions-feature would lack this option aswell as it would merely reference internally.
Speaking of which…Re-importing after chopping, though rarely needed, is easy to accomplish with the versions suggestions I’ve made over the years (and one of it’s advantages). By time stamping the clips on the timeline it’s simple to export/process/import and get things right back to where they were. But again, this is rarely needed but handy if you do.
The ONE place where sharing a file IS advantageous is if you need to make changes to the original file and have it apply to ALL uses, though I can’t come up with a reason for needing to do that (is that why you’re wanting that approach?).
What you describe above sounds like something for a sampler like Grain or a beat slicer to handle, in which case you can already do all of the above that you've mentioned. In fact, that approach seems far superior to the simple timeline editing you can currently do in Reason, no?moofi wrote: ↑29 May 2022No, I don´t want to go back with versions in that case, I would like to keep the chopping but swap the file it´s done to.
Say I´m working with a 30 second recording or sample for that matter. I take that sample and chop it in the sequencer into parts I would like to use and simply rearrange/multiply those parts. Then I got an idea how the original recording can sound even better like processing it through a pedal, a speaker, a module or whatsoever.
Aswell, the first recording could possibly be a dummy, like a not yet "mastered" version, then it´s mastered and I can simply swap the original file to the mastered version.
Now how to simply replace the chopped file? Unless the versions version remembers the chopping and is able to apply those chopping changes to a newly imported file, I don´t rellay see how it could solve this issue. And like said, implementing references is pretty easily achieved, while anything like applicable protocols or recorded actions like in Photoshop, not only complicate the working process but are over the top for the task where a simple swap is all that´s needed.
FWIW, Versions can handle that scenario plus it could also easily deal with the clips scenario (even sourced from multiple audio files) using import/export features and timeline stamping.
And personally, I would always want to keep the original file around if I'm doing additional processing in a song, whether I use a sample or an audio file, so you end up with a copy of the file either way.
[I just went back and re read your original post, and am more confused than before - I've used tons of audio and never hit a limit where I couldn't continue to work, so I'm not sure I'm following your workflow example in the original post (which may explain why I'm still confused now, sorry!)]
Selig Audio, LLC
I bounce mix channels from Reason and import stems and mix them in Cakewalk. Hence the work in Cakewalk does not involve audio editing, it is much like what I do in a video editor. I can move clips around, add volume and fx automation, but then I suddenly want to change the instruments or alter the notes. I would have to re-import some 30 stems so that they get on the same tracks containing the settings and automation I made. I can also have multiple versions of the mixing project. So having references to external files is a nice thing to have.selig wrote: ↑27 May 2022So even in Reason as it is today it’s not a complicated process to do this. You export your audio as a single file, process it, then re-import it. Again, the only way this works differently in other DAWs is when the audio is already (or later) consolidated into a single audio file, right? Or am I missing something here?
I’ve worked many years and never needed multiple projects to reference the same files. What is the use case for this?
When I'm done, I can press 'Consolidate' and get the audio copied to the project.
Perfect.
orthodox wrote: ↑29 May 2022I bounce mix channels from Reason and import stems and mix them in Cakewalk. Hence the work in Cakewalk does not involve audio editing, it is much like what I do in a video editor. I can move clips around, add volume and fx automation, but then I suddenly want to change the instruments or alter the notes. I would have to re-import some 30 stems so that they get on the same tracks containing the settings and automation I made. I can also have multiple versions of the mixing project. So having references to external files is a nice thing to have.selig wrote: ↑27 May 2022So even in Reason as it is today it’s not a complicated process to do this. You export your audio as a single file, process it, then re-import it. Again, the only way this works differently in other DAWs is when the audio is already (or later) consolidated into a single audio file, right? Or am I missing something here?
I’ve worked many years and never needed multiple projects to reference the same files. What is the use case for this?
When I'm done, I can press 'Consolidate' and get the audio copied to the project.
No, it´s not I cannot create an arbitrary number of slices and multiply/rearrange them like in the sequencer. Regarding the necessary tasks there´s all I need for now available in the sequencer with multiple tracks above each other for easy manual adjustment. It aswell has an entirely different workflow and feel compared to a sampler even if it could forfill the given tasks. Anything patternbank- or even MIDI-based is simply a different way of working compared to adjusting/arranging stuff in the sequencer and has thatfor usually a different outcome. It´s more like a technique by itself, not a substitution btw. If you look at sequencer-centric DAW, it´s how people work there quite a lot, even for single repeating bassdrums/hihats, where I would naturally use a drummachine.
And I also like to keep every version for potential revisions or whatsoever at some point.
I know, Versions would handle that then how could I easily replace one file by the other within Versions? What import/export features in combination with timestamps would it offer?
Anyway I would simply like to have it the way orthodox describes Cakewalk in that regard. Like said, it´s how the videoediting world works and having experienced the advantages and workflow it comes with I´m wondering why it isn´t done that way in Reason, especially because samples are already handled both ways.
It´s about swapping, not replacing one by the other, so yes, I would have two files anyway, then the size issue isn´t really related to swapping files but to saving the complete audio along with every file. It simply unnecessarily bloats the file, at least here because I save quite often. Let´s say I got a few hundred MB of audiofiles in a project and save it 30 or even more times, it easily eats GB.selig wrote: ↑29 May 2022
What you describe above sounds like something for a sampler like Grain or a beat slicer to handle, in which case you can already do all of the above that you've mentioned. In fact, that approach seems far superior to the simple timeline editing you can currently do in Reason, no?
And I also like to keep every version for potential revisions or whatsoever at some point.
I know, Versions would handle that then how could I easily replace one file by the other within Versions? What import/export features in combination with timestamps would it offer?
Anyway I would simply like to have it the way orthodox describes Cakewalk in that regard. Like said, it´s how the videoediting world works and having experienced the advantages and workflow it comes with I´m wondering why it isn´t done that way in Reason, especially because samples are already handled both ways.
selig wrote: ↑29 May 2022
FWIW, Versions can handle that scenario plus it could also easily deal with the clips scenario (even sourced from multiple audio files) using import/export features and timeline stamping.
And personally, I would always want to keep the original file around if I'm doing additional processing in a song, whether I use a sample or an audio file, so you end up with a copy of the file either way.
I just went back and re read your original post, and am more confused than before - I've used tons of audio and never hit a limit where I couldn't continue to work, so I'm not sure I'm following your workflow example in the original post (which may explain why I'm still confused now, sorry!)]
-
- Information
-
Who is online
Users browsing this forum: No registered users and 6 guests