Finishing music is not fun. You either strive to create a body of music and accept the graft that goes into it, or you are satisfied with your own personal enjoyment of jamming. It’s really just down to what kind of artist you are, and I don’t think we even really have a choice - it’s just about how we feel we need to express ourselves.
That said, I love jamming but I don’t get a real sense of accomplishment unless I put in the work and finish stuff. And that always 100% requires a DAW.
I embrace dub style production, with constant interaction with the console and the effects. If you can combine this with a DAW, recording your MIDI and audio so you can make small refinements and edits, it’s literally the best of all worlds. I also use Bitwig, and leverage its immense modulation system and hardware integration.
My most efficient workflow is to create a music sheet in a dedicated software like Muse score, at least for the chord progressions and song structure. Then I export the midi tracks in Ableton where the fun part can start by sequencing external hardware or VSTs. I rarely do that as it is not the most fun workflow at least at the beginning but when I do it I make my most favorite tracks
I use grooveboxes because I never “finish”tracks. I freakin hate arrangement views and respect all the artists who work this way. I’ve tried, and made it work on occasion, but I just reach so much more of a flow state creating a loop and then variations of that loop and then tweaking/recording in real-time. For artists seeking to release tracks, or score games or film, or to compose music in a traditional way it makes sense to work in arrangement mode at some stage in the process but I love working with living loops…
Fair enough. I went for a little irony in my phrasing. 
but we just got song mode… 
Most boxes I’ve used, I’ve generally struggled with this. It’s slightly off topic since I don’t go daw at all, but having said that, only the 1010 blackbox and the OP-1 Field have both enabled me to actually finish a decent volume of work.
Those two, however, I’ve developed a lightning fast workflow with now. I’m the only blocker, really. If I got a song in my head, it’s done within the hour if I use one of those two boxes.
I tend to not have a just-one-hour-to-finish-this song in my head, tho …
When I was into hip hop, I worked fully ITB. But that was mostly just a loop with few variations, aligned with the lyrics. So relatively easy to arrange it.
With electronic music, doing all the automation that is needed so it sounds interesting is just incredibly tedious work and not fun at all. That’s why I use hardware for it.
My setup is designed in a way, so I can multitrack every track/instrument during live performance via USB. The plan is to just mix+master it in a DAW and maybe correct some mistakes.
TBH, so far I just used the main stereo mix and did a little of EQ and compression/limiting at the end 
Its funny I use software for the same reasons you find to be tedious in it…lol. It’s super fast for me compared to hardware but this is why this stuff is so personal.
Haven’t watched the OP video yet but what used to work for me was quickly recording a few loops (i.e 3 or 4) to create a very basic DAW arrangement so it’d no longer be just a loop. After that it was experiment with adding new sections or variations (either sequenced or just playing the notes), chopping things up and mangling with fx.
Got kinda bored doing things that way after a while but maybe it’s worth revisiting as a starting point.
This - I work with a producer who is super good at finishing stuff. I enjoy writing and sound designing and playing live, they enjoy tweaking and mixing and arranging and adding reverb and stuff, which I’m not so strong on. Does me wonders! I’m lucky to have this as it’s a skill trade of sorts… so find someone who needs your skills and collaborate!
Is that good or bad?
i’ve been using the DAW mostly to arrange the tracks i’ve been working on lately, no more than that…but most of the playing, if not all, is done on the hardware/plugin.
i’ve also decided not to get back to what i’ve played too much, as it feels like doing so it takes away the initial feeling/inspiration thus not catching the moment
as soon as i’m happy with what i hear, i’ll record it, maybe i’ll give it a second go, but that’s just about it really; if i wait and start to fiddle around with parameters or even maybe change a note, it won’t feel the same anymore; so now i just record whatever i’m happy with straight away, even if there’re mistakes, and i must admit, i like it better this way 
very good. fast as hell once all your clips are in existence (or not, just drop a white clip in as needed for one-off bits)
Oh… just like Maschine. Cool.
That is the most sane approach and what I’ve been doing when at my most productive. Getting an AR last year kind of threw a wrench in my workflow, because I was so used to using Drum Racks for percussion (I would use two MIDI tracks and send the same CC with opposite polarity to the track on/off switches for fills). It still feels weird to have percussive patterns as audio clips somehow - completely irrational, I know.
I commit to audio immediately. If I have to use a sequencer then I’ll commit to audio as soon as the sequence is done. This has also improved the timing of what I play and record live - after 100 takes there has to be a learning effect. I used to be a big fan of synth workstation sequencers. That was a big mistake. Today I prefer the audio workflow by far.
My current flow looks roughly as follows.
1010music blackbox: single recording&writing point and pre-arranging;
—> Loopy Pro: drum programming and further/portable arranging. BB has a 16 pads limit, that’s how I know that it is getting time to switch;
—> Logic Pro: final arranging and documentation;
—> professional specialist for mixing and mastering.
Honestly I feel like those AI track splitters are close enough for me to multitracking, your aren’t going to be able to really change the arrangement super easily but you can rebalance stuff or touch something up enough to get some extra polish. Otherwise I am using a tracker to arrange stuff so there isn’t too much risk of messing it up.
Growing up playing in bands we’d go to record in a studio and realize that our best work was based on the philosophy that you put good music in, you get good music out.
It was much easy to already have the song sounding great before recording.
I treat it the exact same now.
It needs to sound good on my machines.
If it doesn’t I’ll just get lost trying to fix it by over editing in a daw.
it would be nice if some box with overbridge had per track FX like the OT.