Bitwig and Multi Output Instruments Grouping

[stash98]

record each channel from a multi out VSTi

Just use an additional track with its input set to the VSTi's particular output? Or am I misunderstanding what you need?

No, but that does not work for some reason.

I make a new audio track and then in the input I go to Tracks, choose the device and click device chains. It shows all the different outs. So I pick one (lets say Kick Pre) and arm the audio channel. When I trigger the drum, the fader for the device master channel and the individual channel both show and play the audio, but the Audio channel I just set the input for does not.

[stamp]

No, but that does not work for some reason.

I make a new audio track and then in the input I go to Tracks, choose the device and click device chains. It shows all the different outs. So I pick one (lets say Kick Pre) and arm the audio channel. When I trigger the drum, the fader for the device master channel and the individual channel both show and play the audio, but the Audio channel I just set the input for does not.

I just received an e-mail about this from support.
It works but only for recording the audio and there's no monitoring on the receiving track.

The only way to actually send audio to a track and have it monitored is using audio receivers.

Initially I thought it was a bug but support said it's designed to work like this. Coming from any other daw this is really weird.

[Karbon L. Forms]

Why do you need to monitor the recording track if you can monitor the source channel? (Honest question, as I'd like a send-channel-to-dedicated-fully-functional-track button myself!)

I think setting a tracks input to be from another track sets it to a no-monitor-input mode for some reason to do with architectural decisions. It can record input OR it can playback it's own clips thru the chain, but live throughput is not a thing UNLESS you reach around this invisible block with an AR.

I wish a dev would explain this to us. Something along the lines of "In order to implement awesome feature X we had to restrict inter track routing/monitoring for reason Y, and the AR workaround works due to mechanism Z". Maybe some feedback avoidance or temporal logic (chicken/egg) type reason? Noob-proofing? I can imagine a reason but can't think of one. lol. A PDC reason maybe? That would make sense.

[Yokai]

Why not just set up a group with all the routing in place, then create a short clip on the “Group Master”, and then drag that master clip into your User Library? You’ll be given the option to save all the devices in the clip as part of the stored clip.

Then when you need this setup again, just drag out that group clip fro your User Library. Done.

You only have to set up all the individual tracks and routings one time...the first time when you’re setting up the group clip to save into your library.

[Yokai]

As an easier alternative, set all your typical hardware/audio receiver routings/groups in a project, then collapse and de-activate all the groups. In 2.5, anything that’s deactivated uses zero RAM, zero CPU, and incurs zero latency. You can even tell Bitwig to auto-hide all deactivated tracks/groups so they’re out of your way u less you need them.

[stash98]

No, but that does not work for some reason.

I make a new audio track and then in the input I go to Tracks, choose the device and click device chains. It shows all the different outs. So I pick one (lets say Kick Pre) and arm the audio channel. When I trigger the drum, the fader for the device master channel and the individual channel both show and play the audio, but the Audio channel I just set the input for does not.

I just received an e-mail about this from support.
It works but only for recording the audio and there's no monitoring on the receiving track.

The only way to actually send audio to a track and have it monitored is using audio receivers.

Initially I thought it was a bug but support said it's designed to work like this. Coming from any other daw this is really weird.

Thanks. It absolutely worked, but when I add an audio receiver the monitoring is still not visible in the monitor and just in the channel. Not a huge deal, it basically makes the audio receiver unecessary for me unless I am missing something.

I wonder if there is a way to record to mono. For kicks and snares that would be nice.

[stash98]

Why do you need to monitor the recording track if you can monitor the source channel? (Honest question, as I'd like a send-channel-to-dedicated-fully-functional-track button myself!)

I think setting a tracks input to be from another track sets it to a no-monitor-input mode for some reason to do with architectural decisions. It can record input OR it can playback it's own clips thru the chain, but live throughput is not a thing UNLESS you reach around this invisible block with an AR.

I wish a dev would explain this to us. Something along the lines of "In order to implement awesome feature X we had to restrict inter track routing/monitoring for reason Y, and the AR workaround works due to mechanism Z". Maybe some feedback avoidance or temporal logic (chicken/egg) type reason? Noob-proofing? I can imagine a reason but can't think of one. lol. A PDC reason maybe? That would make sense.

I dont need to monitor, so it is ok. It just threw me off. My guess is maybe Bitwigs decision is latency related. In ableton if you monitor a channel it adds latency to it. So you have to turn that off. The only thing that ableton does that is nice is it shows a grayed out signal going through the channel so you at least know there is signal piping through. Still not the end of the world especially if you can answer my question below:

For monitoring and recording external synths, in Ableton we use the External Instrument plugin. The catch with this is that you will get latency unless you fool the input by setting it up as a MIDI track and monitoring with that and than routing the audio to an audio channel with no monitoring that you record to. It’s a pain.

How do you set up latency free recording of external synths in Bitwig? I have a feeling it is a lot smoother, but I am not sure.

[Karbon L. Forms]

Don't have one to try, but the HW Instrument is the equivalent device. I have used it to run BWS and Maschine (standalone) side-by-side. Never noticed any latency. Think there's plenty hardware/Eurorack guys here to answer your question hopefully.

[stash98]

- update below--

[stash98]

Don't have one to try, but the HW Instrument is the equivalent device. I have used it to run BWS and Maschine (standalone) side-by-side. Never noticed any latency. Think there's plenty hardware/Eurorack guys here to answer your question hopefully.

I did it. It looks like there is a little latency but i may just be the synth itself. I am interested in if there are best practices for this that I may not be doing.

Also is there a fast way to quantize audio? I see little blue markers on the transients, but I am not seeing how to quickly get them on the grid. I am on the demo and the audio event options say [demo] next to them so maybe it does not work? the time strecthign options are vast so I figure I must be missing something here.

[Yokai]

Don't have one to try, but the HW Instrument is the equivalent device. I have used it to run BWS and Maschine (standalone) side-by-side. Never noticed any latency. Think there's plenty hardware/Eurorack guys here to answer your question hopefully.

I did it. It looks like there is a little latency but i may just be the synth itself. I am interested in if there are best practices for this that I may not be doing.

Also is there a fast way to quantize audio? I see little blue markers on the transients, but I am not seeing how to quickly get them on the grid. I am on the demo and the audio event options say [demo] next to them so maybe it does not work? the time strecthign options are vast so I figure I must be missing something here.

There is not yet some auto-magic "Quantize Audio" button, but in practice you only need a few stretch markers in strategic places to quantize an audio clip in a musical and natural way.

1. Go into the Event Editor for an audio clip. CLIP mode is probably easiest.

2. Turn on the Stretch mode you want for the clip. I recommend the "Elastique Pro" for the least artifacts after stretching (transients are relatively untouched and strong in this mode). Unlike the similar-sounding "Complex Pro" mode in Ableton (which sounds AWFUL most of the time and is a CPU hog), the Elastique Pro mode in Bitwig sounds great on nearly all material and is low CPU impact. It's my #1 stretch mode I reach for unless I'm trying to creatively mangle the sound somehow.

3. In the Onsets mode of the editor, zoom in close on the transients closest to the downbeats on the "1" (n.1) of every bar in the clip. Make sure the blue onset marker is truly sitting where the downbeat transient spike is in the waveform. (Just like in Ableton, Bitwig does moderately well at placing the onsets but NOT perfectly at placing them.)

4. Now flip to the Stretch mode of the editor. Hover your mouse at the BOTTOM of the waveform underneath each of those onsets at the downbeat of each bar in the clip, then left click to place a stretch marker for that onset. Drag the stretch marker to the grid line at the start of the bar. Repeat this process for the downbeat of each bar in the clip.

After step 4, you have the downbeats of the clip lined up with the grid. 8 bars? Then you have 8 stretch markers all sitting on the downbeat gridlines for each bar in the clip. If for some reason this still doesn't sound "quantized" enough for your taste, then repeat this process for the downbeats on the "3" (n.3) of every bar. If you have to go much deeper than the 1 and 3 downbeats in each bar, then either your audio clip is truly whack (or you've got the wrong "original tempo" specified for it in the Stretching section of the inspector), OR you're just trying to over-quantize and take all human expression out of an audio recording.

In practice, I'd wager that a lot of the urge to "quantize audio" is based on specifying the WRONG original tempo for a clip when stretching it. If you've got the true, correct original tempo specified, your audio events should be fairly lined up with the grid and any variation falls within normal, "human and musical" timing ranges.

Bonus Thought: My guess as to why the Bitwig devs have never implemented an auto-magic "Audio Quantize" feature in Bitwig after YEARS of people asking is because Bitwig knows that an automatic algorithm for doing that is pretty hit or miss, and you'll ALWAYS get much better results with human eyeballs and a human brain making good decisions about what, exactly, to move onto a grid line.

[stash98]

Don't have one to try, but the HW Instrument is the equivalent device. I have used it to run BWS and Maschine (standalone) side-by-side. Never noticed any latency. Think there's plenty hardware/Eurorack guys here to answer your question hopefully.

I did it. It looks like there is a little latency but i may just be the synth itself. I am interested in if there are best practices for this that I may not be doing.

Also is there a fast way to quantize audio? I see little blue markers on the transients, but I am not seeing how to quickly get them on the grid. I am on the demo and the audio event options say [demo] next to them so maybe it does not work? the time strecthign options are vast so I figure I must be missing something here.

There is not yet some auto-magic "Quantize Audio" button, but in practice you only need a few stretch markers in strategic places to quantize an audio clip in a musical and natural way.

1. Go into the Event Editor for an audio clip. CLIP mode is probably easiest.

2. Turn on the Stretch mode you want for the clip. I recommend the "Elastique Pro" for the least artifacts after stretching (transients are relatively untouched and strong in this mode). Unlike the similar-sounding "Complex Pro" mode in Ableton (which sounds AWFUL most of the time and is a CPU hog), the Elastique Pro mode in Bitwig sounds great on nearly all material and is low CPU impact. It's my #1 stretch mode I reach for unless I'm trying to creatively mangle the sound somehow.

3. In the Onsets mode of the editor, zoom in close on the transients closest to the downbeats on the "1" (n.1) of every bar in the clip. Make sure the blue onset marker is truly sitting where the downbeat transient spike is in the waveform. (Just like in Ableton, Bitwig does moderately well at placing the onsets but NOT perfectly at placing them.)

4. Now flip to the Stretch mode of the editor. Hover your mouse at the BOTTOM of the waveform underneath each of those onsets at the downbeat of each bar in the clip, then left click to place a stretch marker for that onset. Drag the stretch marker to the grid line at the start of the bar. Repeat this process for the downbeat of each bar in the clip.

After step 4, you have the downbeats of the clip lined up with the grid. 8 bars? Then you have 8 stretch markers all sitting on the downbeat gridlines for each bar in the clip. If for some reason this still doesn't sound "quantized" enough for your taste, then repeat this process for the downbeats on the "3" (n.3) of every bar. If you have to go much deeper than the 1 and 3 downbeats in each bar, then either your audio clip is truly whack (or you've got the wrong "original tempo" specified for it in the Stretching section of the inspector), OR you're just trying to over-quantize and take all human expression out of an audio recording.

In practice, I'd wager that a lot of the urge to "quantize audio" is based on specifying the WRONG original tempo for a clip when stretching it. If you've got the true, correct original tempo specified, your audio events should be fairly lined up with the grid and any variation falls within normal, "human and musical" timing ranges.

Bonus Thought: My guess as to why the Bitwig devs have never implemented an auto-magic "Audio Quantize" feature in Bitwig after YEARS of people asking is because Bitwig knows that an automatic algorithm for doing that is pretty hit or miss, and you'll ALWAYS get much better results with human eyeballs and a human brain making good decisions about what, exactly, to move onto a grid line.

Great post, thank you. Honestly audio quantize has worked well for me in the past but I also agree on the sound quality.

Where I disagree is I also have recordings where I need to quantize a decent amount of audio. Live playing that I want to purposefully edited to be ultra tight is something I like to do at times. I love it on super percussive rhythm guitars for example. I don't think Bitwig would take the route of "we know better than you about this". That's not a good move by them. I hope they would add this feature and let us decide for ourselves on what we want. I say this because hand editing a lot of transients (even when well-played) is soul sucking.

One thing I prefer about Bitwig is that it appears you can grab a marker and move it without affecting everything in front or behind it if you don't activate markers there as well. That's a really nice little benefit.

[da10us]

I’m afraid this will take forever. It’s the only flaw in this pretty good software in my opinion. We need ‘normal’ routing on multi-out instruments. My work around is individual Kontakt/Omnisphere/Vienna/etc instances. Basically multi-out in Bitwig is useless to me.

[stufflin]

I still do it the 2016 way. :/
Going to try the groupsends though. I hate it that i loose my levelmeters when using audioreceivers.

[Zacchino]

Can the OP send a formal FR for this issue? Any modern DAW should work flawlessly with Kontakt & co.