Most efficient way to use plugins?

Hi,

Let’s say I have 10 different guitar tracks. More or less the same thing with a few variations.
This tracks were recorded dry in direct input.
I want more or less the same sound for every tracks: chorus effect, same amp sim.
Maybe a different impulse response here and there.

I feel like putting the same plugins in direct insert is a was of CPU.
Should I create a bus with auxiliary send (since the tracks are mono, should the bus be mono as well?) and return.
Or should I export each track with plugins enabled one by one and so that I don’t the plugins at all anymore?

More generally, how do you do when you have many tracks and many plugins?
Some effects can be applied only once (Noise Gate, HPF…) keeping them enabled during all the mixing seems counterproductive. On the other hand, if you have 20 tracks or more, to export each track and re-import it seem never ending…

I’ll recommend an ampsim for each track. When it’s a really clean sound, you may get away with an ampsim in a group, but adding up different DI tracks has an effect of the saturation and compression behavior. If you like the sound, do it!

Effectwise use sends. I find it more easily to define the amount of effect with just the send fader as to adjust every dry/wet knob individually. And as you said, CPU-handling. If you use chorus or a delay, it might be reasonable to use a stereobus (just think of ping-pong-delays f.e.).

cheers eighty

Thank you. I’ll do that.

How do I route the signal?

If I have “guitar track 1” with amp sim plugin.
I route the signal before amp sim to “chorus bus” through aux send.
How do I feed the signal back from chorus bus to the amp sim (on guitar track 1) without larsen?

Go to the effect bus after the amp. That’s like putting your effects in the FX loop of a real amp.
Then you can route the amp tracks and the effect bus to the Master or a guitar bus.

This routing simulates a FX loop in parallel mode.

There are of course other options but i tend to mostly use this method.

Then you can route the amp tracks and the effect bus to the Master or a guitar bus.

Mmm, it sounds like parallel processing, not exactly what I want.

My idea of sending all the tracks to a single chorus bus is only to save some CPU. But I’d like to sound as if the chorus was directly on the track.

What I’m looking for is a way to have the signal routed like this for each track :

dry -> chorus -> amp -> cab -> eq -> pan and fader -> (to guitar bus -> master bus)

With a chorus bus, it has to be like :

guitar track --> [[send to [chorus bus] return wet sound to]] -> guitar track -> -> amp sim -> impulse --> eq and stuff -> pan and fader -> Guitar Bus (where I’ll add some reverb and more eq and compression)

Is it possible?

I’m not sure, i get it right.

Sure you can send all tracks to one chorus bus, but you can not get seperated wet signals back to your guitar track, since the ‘chorus out’ is the sum of all tracks fed into. For what i understand is an ampsim more CPU hungry than a chorus, anyway.

You can route all tracks into one group with chorus, amp, cab, eq. But that’s like plugging multiple guitars into one amp setup and this will affect the resulted sound. If that sounds as you want it, go for it.

No, you cannot insert a bus in a track’s signal flow.


Let’s take a step back.
Is your system’s DSP load high so that you have to do this optimization?

…or is this only for convenience, in order to do the settings only once, and share them for all tracks. If so, I’d group the tracks and then create a “subgroup bus”. All common FX can go there.

image

I’d still keep reverbs separate (aux-send from the subgroup bus to a shared reverb-bus).

Oh. So that’s not possible.

I guess I’ll stick to insert then.

What I will do is treat each track with chorus + amp + cab, then export each track one by one, then reimport the new track to process the panning, balance and eq.

You don’t need to export. Just select all regions, right click and klick “freeze”.

Could be that you have to restart the session to get this working properly.

Update: could be that just rightclicking one region in the track might be enough to “freeze”/consolidate the whole track so you don’t need to select all regions.

what “feeze” does? I’ve never used it.

Does it re-record the track with the plugins on?

edit: I’ve tried with one duplicate tracks, it’s not exactly what I want.

@ Robin,
I thought about doing that at first, but I don’t control the panning as accurately with this method. At least to my ears.

To make things more clear: I have 4 tracks with one single region on each (yep) going all the way from the beginning to end of the song.
The other tracks won’t be all the way, it will be some variation here and there. But some will be 70% left, some will be 25% right, and so on. That’s why I need a lot of tracks.

shortly put, yes. It does an “offline bounce” of the track with the plugins on, then creates a new region with the bounce, puts the bounce into the track and deactivates the plugins to free the CPU.

If you feel the need to change the settings later, you can unfreeze the track and have access to all plugin parameters again.

1 Like

If the balance of the group is 100% width and centered, there should be no difference with the panning in Robin’s suggestion. If there is either something else is causing it (Look at your plugins) or there is an issue that needs to be addressed (Unlikely, but not 100% out of the question).

  Seablade

As an example, here’s how I do it. This has both cases covered.

There’s two tracks of Electric Guitar and each need different treatments so they each get their own ampsim, comp, eq, etc
There’s also two Accoustic Guitar tracks and these share a common treatment so I send them to an Accoustic Gt Bus that does it all.
Finally, the two Elec Gt tracks and the Acc Gt bus have an Aux send to a common Guitar Reverb bus to bring some cohesion between them.

What is not visible on the picture is that all of these finally end up on a global Guitar Bus that has no effect, it’s just a convenience to manage a global volume for all guitars.

shortly put, yes. It does an “offline bounce” of the track with the plugins on, then creates a new region with the bounce, puts the bounce into the track and deactivates the plugins to free the CPU.

I thought he would do that: create a copy of the region with the plugins on, then we can unfreeze, remove the original region and the plugins, play the copy.
But I can’t see any copy in the track (the track layering is on “stacked” so I should see it if they were 2 region at the same place).
Maybe I can freeze the track and duplicate the region or copy-paste?

@Vincent Tassy

You have a lot of plugins. Don’t you have any delay issue and if so how do you fix it?

Thanks everybody. :slight_smile:

Small suggestion, look into VCAs for this purpose, could make your life much easier as you start to learn to really utilize them, but at the least this is exactly what they are designed for as there is no processing for them.

Seablade

If the plugin properly reports it’s latency, Ardour will automatically compensate for it so you shouldn’t have a delay issue.

 Seablade

Seablade
Small suggestion, look into VCAs for this purpose, could make your life much easier as you start to learn to really utilize them, but at the least this is exactly what they are designed for as there is no processing for them.

Thanks ! I sure will ! I read about them but never took the time to try them out. I should do that on my new project.

newbie44
You have a lot of plugins. Don’t you have any delay issue and if so how do you fix it?

No latency issues. I’ve used more plugins than this before without problem :wink:

Just to be clear, there is no relationship between number of plugins and latency. You could have thousands of plugins and zero latency.

There are plugins that introduce latency (and Ardour will compensate everything correctly if they report it correctly). But this is quite different from the idea that using plugins adds latency. It does not.

1 Like

I ran into this problem on my album, and my solution was:

  1. Create a guitar bus for each kind of tone
  2. Run compression individually on each track
  3. Run amp sim, tone shaping, IR loader on the main bus
  4. Set up a chorus, reverb, and ambience bus (the ambience bus is verb > delay > verb > lopass/hipass EQ)
  5. Add the effects as sends onto the main track
  6. Add the effects onto individual tracks as needed.

“Run compression individually on each track”

May I ask why?

(as opposed to a compressor on the master bus)