Recording/mixing workflow.

Hi ardour community,
I’d like to share thoughts about workflow and who does what and why, I also have questions about some specific aspects of Ardour. This post is meant to be a rough workflow example for beginners and is meant altogether to improve our workflow. I’d be delighted to hear how each of us tackle this
let’s start with a bit of background:
I run in my spare time a project studio running Ardour on Ubuntu. I use a rme HDSPe AES with SSL alpha-link converters. I use Audient preamps and several other 500 series preamp, most of them DIY. I have a few hardware EQs and compressors, more on the way to be built. Analog interfacing workflow is becoming more and more something that I need to tackle for smooth running sessions. More on that later.
I record mostly acoustic bands with the few occasional private people wanting to sing in a real studio mic.

So far my workflow is:
track the band live except the lead instrument/singer which is firstly singing along as a reference for the other musicians. I create 1 session per song and do 1 long take until I have 2-4 useful and good takes.
I then decide which take is the best and track the lead on top of it, using playlists for the different takes. I do the same for solos or any other kind of re-recording.

Tracking session finished, everybody is happy and drink.

I then start editing the material. I decide which take is the best, mark it, cut and paste to edit the mistakes out. I used to use the start/end markers to mark the final track but they move along when I record new material beyond the markers, so I ended up using a range instead and snap the start/end markers to the range when I’m definitely really sure I won’t change anything on the edit. Any better ideas?
When I edit the lead instrument I start to use playlists. I read about how great playlists are in Ardour, use them happily but I’m still a bit unsure about their benefits, especially when done tracking. how do you people switch fast between takes in order to compare them? Should I record every take on top of each other in one track and then use the option to visualise them all at once? Then why bother with playlists? to keep the track “clean”? experience here is welcome.
I know I should use more the different views in order to avoid some scrolling and speed up the editing process, not yet in my workflow… do you have useful tips on how to speed up editing?
I then let the band validate the edit and start mixing.

I don’t have much to say about the mix process itself, there is enough documentation online and tons of great plugins. But I’m still a bit unsure about the analog interfacing: I have 2 mono EQs, a mono compressor and a stereo one. So far I mostly used the mono comp this way: create a new track, use a send, compensate latency and record through the compressor on the new track, disable the original track and continue mixing. It’s time consuming but it works. I could also directly route the output of the track to the hardware, but I would need to disable the plugins already loaded, and the latency compensation gets lost, I’d need to synchronize the tracks by hand, which ends up not being faster, as far as i see it. How do you guys do interface your hardware? Use an insert on the original track and record the output on the new track?
About patching the hardware: At the moment, the different units are directly hooked to the converters/soundcard and I patch digitally using Ardour. I could also use a hardware patchbay (I “just” need some cables) and assign each track to a patchbay number and then use real cables to connect tracks and hardware. I don’t see a method being better than the other one, except if i have more hardware effects in/out than my soundcard where i would need a hardware patchbay. Feedback and thoughts on this point welcome.

Note that I often use the same signal chain across tracks and sessions. Inside a session, is there a way to copy the entire signal chain of a track in another one? If I have 5-6 FX, drag’n’drop is really time consuming. Is there a way to share this across sessions? Or are presets+screenshots+pen & paper the way to go?

Comes pre-mastering: I import each song on a new track, create a new stereo “master” track where I route the output of each song. Note that I use hardware so I need this extra “master” track in order to record the hardware output. Why import each song on a different track? If I need to change the gain of one track, I don’t need to use automation for the next track.
I then set up volumes and maybe EQ and compression on each track separately and record the result on the “master” track. I create the different ranges for the export, solo the “master” track and export.
Here i don’t get the difference between CD-ranges and normal ranges, especially if I tick the “CD-track” box in the range editor window. I only got problems when I did this for both. I ended up using only the range markers with “CD-track” checked. I create a .cue file and using cue2ddp create a proper DDP image for the CD plant.

So that’s my workflow so far, I hope it can help some even though the post raises a lot of questions.

The tracking and editing workflow has to be something you figure out. What works for you will largely depend on how you organize and approach your project as well as what constraints you have to deal with. There are many different tools and methods for this and you have to fine what best fit what you want to accomplish. If playlists for example don’t provide any advantage for you, don’t use them.

I could also use a hardware patchbay (I "just" need some cables) and assign each track to a patchbay number and then use real cables to connect tracks and hardware.

Your audio card has a built in routing matrix. Patch it it the matrix.

Inside a session, is there a way to copy the entire signal chain of a track in another one?

Hold shift while selecting to select multiple, then drag and drop all at once.

Here i don't get the difference between CD-ranges and normal ranges

Your workflow doesn’t really have much use for them. They are more for when a single session is going to be sliced into separate CD tracks. (For instance a live concert, or if you are recording book on t̶a̶p̶e̶ CD)

IMO you are relying too much on external hardware. Really for a small operation it’s not needed. I can understand it it’s some sort of tube based hardware that adds a unique coloration to the sound, or for instance bouncing it off of tape to get the subtle tape saturation. But for most utility functions like EQ and compressor, it’s just a lot of work with little to no benefit. The software compressors and EQs are just as good if not better. It can make sense to use them to tame a erratic signal source at the recording stage, but beyond that it’s fantastically tedious. People like to gush over outboard processing hardware in magazines, but most of that is boutique relics of a dying workflow. Spend money on mics, preamps, AD/DA converters, monitoring equipment and room treatments. Most everything else is superfluous.

Hi Reuben and thanks a lot for your feedback.
I’m starting to figure out that hardware isn’t THE solution, especially while mixing on a small scale as I do. But hey, it’s there, I like to track through it and I like to build it. But I definitely keep in mind to expand my mic locker first before starting more outboard.
I actually mixed the latest album nearly without hardware outboard. Patching and recording it is a tedious work as you said. Even though it was easier to setup the one hardware compressor I used on lead vocals.
Thanks about the soundcard matrix tip, I forgot about it.
Finding the right workflow is a fascinating thing, I’m still open to share and collect ideas about it.