Reaper's take system sucks, but rather than vaguely bitch about it...

[Justin]

Originally Posted by chip mcdonald

there will be some aspect someone doesn't like [...] There is no single solution. Just other take emulation modes.

I disagree -- the goal isn't to please everybody. It's to make something that works and works well. We haven't gotten that right (yet).

[foxAsteria]

I do have to say that for me the take system is perfect. Could not be better.

Well after I created a few custom actions, that is...kinda shocking to me how many issues are listed here.

[_Stevie_]

Thanks Justin for taking care of this one.

To me, this is one of the most elegant comping workflows I have seen to date:
https://youtu.be/1S-2x0VdH1Y?t=424

[Robert Johnson III]

Originally Posted by Justin

I disagree -- the goal isn't to please everybody. It's to make something that works and works well. We haven't gotten that right (yet).

If you make something that works well you’ll please (just about) everybody… but I’m afraid that without changing anything big the story will continue… (not because we are used to different solutions in some other DAW)

[Robert Johnson III]

Originally Posted by _Stevie_

Thanks Justin for taking care of this one.

To me, this is one of the most elegant comping workflows I have seen to date:
https://youtu.be/1S-2x0VdH1Y?t=424

Take a look at LogicX track alternatives

https://youtu.be/Af6NCJFzZaw

[_Stevie_]

That's not bad, but reminds me more of Cubase' track versions, than a fully featured take system.

[Robert Johnson III]

Originally Posted by _Stevie_

That's not bad, but reminds me more of Cubase' track versions, than a fully featured take system.

A Logic Track Version includes a completely independent take system, with automation, FX, swipe-comping etc... Never used the Cubase track versions. Used Studio One and Logic extensively. Studio One is elegant, but also a bit more simplified (in my opinion)

[Robert Randolph]

Originally Posted by Justin

The first bit of low hanging fruit (awful phrase that is overused, sorry, but applicable here) is an option to reconcile take lanes when switching takes (using the previous-valid take lane if the current take is empty). Here it is in action:



(first switching takes via clicking, then via T/Shift+T, then via the take selection menu).

Remember as well that takes can be created by overdubs and/or without the performances 'lining up'.

This action does not deal with that scenario particularly well, yet it can be desirable for the take switching to 'stay in order' of the take lanes. Hopefully that makes sense.

The dreaded Other Daws™ all handle this scenario without a hitch.

And thank you very much for looking at this.

[Lokasenna]

Imagine a piano keyboard, eighty-eight keys, only eighty-eight and yet, and yet, new tunes, melodies, harmonies are being composed upon hundreds of keyboards every day in Dorset alone. Our language, Tiger, our language, hundreds of thousands of available words, frillions of possible legitimate new ideas, so that I can say this sentence and be confident it has never been uttered before in the history of user-developer communication:

Thanks, I'm digesting this and finding ways to address some of these behaviors without changing anything big, stay tuned..

(Borrowed from A Bit Of Fry & Laurie)

[Ivo Sedlacek]

For me the take system is simply perfect: you record several take, you see them all in lines, you easily select the best parts and bring the result back to a single track ... (and you can return to multiple takes any time). What more simple and straightforward one could imagine ?

[emarsk]

Originally Posted by Ivo Sedlacek

For me the take system is simply perfect: you record several take, you see them all in lines, you easily select the best parts and bring the result back to a single track ... (and you can return to multiple takes any time).

This is just a description of the generic concept, not of the implementation. The devil's in the details, here: the concept is fine, the issues all come from the specific peculiarities of REAPER's current implementation and how they interact with some workflows; that's what makes "easily" debatable.

Originally Posted by Ivo Sedlacek

What more simple and straightforward one could imagine ?

This is the wrong question because:

• the previous point;
• REAPER's implementation's issues are not about simplicity or straighforwardness, but about usability and robustness;
• there's no need to "imagine": there are other implementations to look at that are just as simple and straightforward but more usable and robust.

P.S.: Thanks Justin for looking into it!

[nofish]

Originally Posted by maxdis

Last time I tried that action, it never worked, don't know why; maybe something has been fixed in the meantime, I'll try again.

Some overhaul for the SWS auto group coming in next SWS version.

Regarding Reaper's take system, it's totally fine for me as long as I record click based stuff.
For freetime performance (where performances don't necessarily line up, as mentioned by R. Randolph) I find it not really useable because of the 'all takes' have the same length' paradigm.

edit:
Some discussion about this here.
Maybe a 'low hanging fruit' to tackle it would be if Comp groups could store split points positions as mentioned in that thread.

[Jae.Thomas]

Originally Posted by nofish

Some overhaul for the SWS auto group coming in next SWS version.

Regarding Reaper's take system, it's totally fine for me as long as I record click based stuff.
For freetime performance (where performances don't necessarily line up, as mentioned by R. Randolph) I find it not really useable because of the 'all takes' have the same length' paradigm.

great point

[_Stevie_]

Yep, absolutely agreed. That's why I posted the Studio One video.

[karbomusic]

Originally Posted by nofish

For freetime performance (where performances don't necessarily line up

Never had a problem with it and I mostly always use free time and misaligned takes. There are definitely tips and tricks that I use but it's so second nature I don't even think about it - and it is very efficient for me - I still say Justin was ahead of the game on a number of take features. Those splits everyone hates are invaluable to me.

[nofish]

karbo, can you link me to posts from you where you've written about that?
Otherwise, feel free to post in the thread I linked above (I think it would fit there), I suppose others may find it helpful too.

[karbomusic]

Originally Posted by nofish

karbo, can you link me to posts from you where you've written about that?
Otherwise, feel free to post in the thread I linked above (I think it would fit there), I assume others my find it helpful too.

I will or as I've promised for a couple of years now, video my working with takes. I'd do it this week which is a good time due to having the week off of work, but my current project (some posts in collab forum) is all about single take live performances with zero punches, edits or overdubs.

Something I always bring up as of late is being specific about which part of the takes system we are discussing. I don't use the comps feature so any improvements there, are beyond me since I don't use it. My comments are usually related to comping a single final version after dropping a bunch of sometimes randomly spaced takes.

FWIW, I always collapse takes (CTRL+L), I've found it so worthwhile that it is very rare that I ever expand them - I feel pretty strongly about how much better working with takes this way is but it visually freaks many out. It has held it's ground though with me for years now and I do some fairly complex projects - not orchestral complex but pretty complex track/take wise.

[chip mcdonald]

Originally Posted by Justin

I disagree -- the goal isn't to please everybody. It's to make something that works and works well. We haven't gotten that right (yet).

I disagree, as FoxAsteria posts below I think it works well as is, I have no issues with it. But I'm not trying to have it somehow relate to another (contradictory)workflow.

I think this is going to go on forever with "well, in Cubase it does this", "in PT it does this". I think the base issue is that there is an inherent intractable problem in the dichotomy of having a floating cursor independent of the loop selection. That creates a situation that is making people want 2 different and opposing outcomes that are mutually exclusive, and it's going to be the dog's breakfast of a "solution" to appease both IMO.

I don't know how much effort you're devoting to "this", but I would imagine trying to concoct something new/different would be more difficult than just making the existing code conform to "PT, Cubase, Logic practices" that people seem already content with.


"Well, in this (other) daw it does this..."

[chip mcdonald]

Originally Posted by Ivo Sedlacek

For me the take system is simply perfect: you record several take, you see them all in lines, you easily select the best parts and bring the result back to a single track ... (and you can return to multiple takes any time). What more simple and straightforward one could imagine ?

I agree. I can't, literally, think of a more straightforward, less action-based methodology.

The problem is that for those of us that have no problem with it, it's just another aspect of Reaper and not a battleground. My fear is that this will lead to some new workflow that wrecks the present streamlined one I like.

[EvilDragon]

I think it'd make more sense to somehow adapt FIPM to those other-DAW comping workflows rather than trying to shoehorn current take system into those.

[serr]

Originally Posted by Ivo Sedlacek

For me the take system is simply perfect: you record several take, you see them all in lines, you easily select the best parts and bring the result back to a single track ... (and you can return to multiple takes any time). What more simple and straightforward one could imagine ?

That's what I thought and have been saying too. Why would anyone want to make more work and complexity out of this after it finally evolved to so simple and straightforward of a system that made the needed complexity of the past a moot point?

Originally Posted by nofish

Regarding Reaper's take system, it's totally fine for me as long as I record click based stuff.
For freetime performance (where performances don't necessarily line up, as mentioned by R. Randolph) I find it not really useable because of the 'all takes' have the same length' paradigm. ...

Well, that's the answer isn't it!
All the folks constructing "beats" to a grid.
Easy to forget about that if you record audio and live bands or studio projects that can and prefer to just play freely.


I don't think we have to worry about anything getting broken in any retooling here. (Those of us not using the takes feature.) Because... we're already not using it!

And just maybe some useful tool will come out of this that we WILL find useful.

[EvilDragon]

Originally Posted by serr

Why would anyone want to make more work and complexity out of this after it finally evolved to so simple and straightforward of a system that made the needed complexity of the past a moot point?

Well, it's not simple and straightforward in many cases which Robert explained very well in the opening post... So it's not an end-all-be-all system.

[chip mcdonald]

Originally Posted by nofish

For freetime performance (where performances don't necessarily line up, as mentioned by R. Randolph) I find it not really useable because of the 'all takes' have the same length' paradigm.

The problem is that the "fix" will likely break the previous implementation. For instance I preferred having vol/pan envelopes over the same track lane - that "fix" annoys me where before it wasn't even given a thought. I now have to click a bunch of stuff and open an extra lane,right click and select, jump through hoops. I'm afraid this is what will happen with Takes. There won't be a "happy medium" or solution, there are inherent contradictory aspects IMO.

I think ultimately there will have to be a "mode" switch to appease what appears to me to be 3 basic workflows that are hybrids of the Major Daws.

[karbomusic]

Originally Posted by chip mcdonald

"Well, in this (other) daw it does this..."

That isn't what Justin is addressing, he's looking at shoring up certain issues with the current system which do in fact need a little nip and tuck that wouldn't break or affect anyone other than for the better, and one of those is exactly he showed example wise. :/

What I do agree with is the overall methodology is fine, there are just far too many with preconceptions and assumptions that they've previously learned then try to force reaper to fit those, that will never result in happiness, ever - but if they look at it from "what can this system do for me" and think of the result "a successful track" instead of hyper focusing on the steps they expect to work the way they are used to, they'll do much better and likely faster and more efficiently.

You see this all the time in this very thread "this is how I'd expect it to work" - but it doesn't and there are other ways that are as efficient and I'll say often better.

[karbomusic]

Originally Posted by EvilDragon

Well, it's not simple and straightforward in many cases which Robert explained very well in the opening post... So it's not an end-all-be-all system.

Correct, many in his list are things that just need to have the rough edges smoothed out - but there are a couple that are out of not knowing how to use the system to it's fullest as well.

[Robert Randolph]

Originally Posted by EvilDragon

I think it'd make more sense to somehow adapt FIPM to those other-DAW comping workflows rather than trying to shoehorn current take system into those.

FIPM and item lanes are very close to already being the best takes system on the market.

There's just not support for the comping process.

[Robert Randolph]

Originally Posted by nofish

For freetime performance (where performances don't necessarily line up, as mentioned by R. Randolph) I find it not really useable because of the 'all takes' have the same length' paradigm.

I'd like to point out that it's not necessarily free-time performance.

Takes can be created when doing overdubs, or when needing to loop over a section already containing takes.

This is where about 50% of the issues come from, but at the same time this is a common scenario in many workflows. It's also a scenario that is demonstrably possible to do well.

[chip mcdonald]

Originally Posted by karbomusic

That isn't what Justin is addressing, he's looking at shoring up certain issues with the current system which do in fact need a little nip and tuck that wouldn't break or affect anyone other than for the better, and one of those is exactly he showed example wise. :/

I know that's not what he's addressing, but in the context of this particular thread "no one is going to be happy".

What I do agree with is the overall methodology is fine, there are just far too many with preconceptions and assumptions that they've previously learned then try to force reaper to fit those, that will never result in happiness, ever - but if they look at it from "what can this system do for me" and

We agree.

think of the result "a successful track" instead of hyper focusing on the steps they expect to work the way they are used to, they'll do much better and likely faster and more efficiently.

I agree, but in the context of this thread it's always going to suck to some people. A nip and tuck of the present system could be great, but it's not going to affect the plaintiffs here.

[nofish]

Originally Posted by Robert Randolph

I'd like to point out that it's not necessarily free-time performance.

Point taken, I've referenced you not 100% correctly here.

[Justin]

... making Comps include item mute state, this is nice, too.

[karbomusic]

Lock active take would be really nice if it also locked or optioned to lock with T/SHIFT+T - I constantly use takes collapsed (excellent idea btw) so if I lock to active take, I'd love it to be locked in all cases. I hardly ever make suggestions due to their already being so much noise but is one that would be helpful in preventing rogue keystroke accidents that put me in WTF just happened mode.

[EvilDragon]

Originally Posted by Justin

... making Comps include item mute state, this is nice, too.

While you're at it...

https://forum.cockos.com/showthread.php?t=209873

[nofish]

Wanted to throw in this as an idea:
https://forum.cockos.com/showthread.php?t=32313

(Note: instead of 'take' it should say 'layer' there)

Together with the to come 'include item mute state in Comps' this could provide for a quite nice comping way in layers recording mode imo.

[Robert Randolph]

Originally Posted by nofish

Wanted to throw in this as an idea:
https://forum.cockos.com/showthread.php?t=32313

(Note: instead of 'take' it should say 'layer' there)

Together with the to come 'include item mute state in Comps' this could provide for a quite nice comping way in layers recording mode imo.

I really like this idea.

Along with some extra mouse modifiers/actions for slicing/moving items in layers, this is a 'quick fix' that fixes all of my complaints with the current system used for takes.

It also leaves the current system intact, which allows us to use some of the really cool functions that it provides.

[karbomusic]

Originally Posted by Robert Randolph

I really like this idea.

Just an FYI... It's the same thing in posts #113/109 and possibly a few other posts - nofish and I have been suggesting this since 2009/2010 or so. One reason I keep mentioning it is that it's the shortest path to solve 99% of the take/split complaints without stepping on the toes of other workflows and is essentially the same way video tracks work in reaper now, and most all video apps, which means it even fits "intuitive" requirement.

[Robert Randolph]

Originally Posted by karbomusic

Just an FYI... It's the same thing in posts #113/109 and possibly a few other posts - nofish and I have been suggesting this since 2009/2010 or so. One reason I keep mentioning it is that it's the shortest path to solve 99% of the take/split complaints without stepping on the toes of other workflows and is essentially the same way video tracks work in reaper now, and most all video apps, which means it even fits "intuitive" requirement.

It's also the solution I proposed in the original post, because I had read a number of old posts on the forum about it.

[karbomusic]

Originally Posted by Robert Randolph

It's also the solution I proposed in the original post, because I had read a number of old posts on the forum about it.

Sounds popular.

[Robert Johnson III]

Originally Posted by nofish

Wanted to throw in this as an idea:
https://forum.cockos.com/showthread.php?t=32313

(Note: instead of 'take' it should say 'layer' there)

Together with the to come 'include item mute state in Comps' this could provide for a quite nice comping way in layers recording mode imo.

Maybe one suggestion… What will be heard on playback should be portrayed in the top lane (seems visually more ergonomic, then collapse the "layers" when done)

[karbomusic]

Originally Posted by Robert Johnson III

Maybe one suggestion… What will be heard on playback should be portrayed in the top lane (seems visually more ergonomic, then collapse the "layers" when done)

Cubase uses the bottom-most, video apps use top-most, but in Reaper's video prefs there is the option to choose between bottom-most or top-most having priority so it should likely be the same option available for audio.

[Vagelis]

Originally Posted by nofish

Wanted to throw in this as an idea:
https://forum.cockos.com/showthread.php?t=32313

(Note: instead of 'take' it should say 'layer' there)

Together with the to come 'include item mute state in Comps' this could provide for a quite nice comping way in layers recording mode imo.

Yes please,coming from Cubase and after 10 years of using it i can say this was a very useful feature for selecting individual parts of audio faster and i miss a lot!Combined with Reaper's take system will be the best!
+1!!!