When I start recordings with a band, I want to feed their headphones /TM hardware outs  with mixes fron the total mix TM hardware inputs  ( no latency from daw (Nuendo).
If  -as an example- the bass wants to correct something, I`d like to be able to copy all settings I made in the hardware inputs (can`t be heard in playback modus) to the daw outs (except the bass), so the headphone mixes stay the same. And afterwards go back to TMhardware ins for more tracking .. (using master mute to have only one feed)
For me, -volume, pan, mono/stereo, routing- would do... (madi pci card).

I think you want to look into using Asio direct monitoring (ADM).
Here the DAW controls totalmix when Hardware monitoring is activated in the daw.
You have zero latency monitoring during recording and you can monitor the daw during playback. When dropping in, totalmix switches from playback to input.
Georg

Use submixes and snapshots in TM FX.

As you want to have near-zero latency monitoring lets exclude the DAW for a moment.
The DAW gets the signal unaltered from HW inputs, so your recording is "safe".

Now lets talk about TM FX submix mode.
Every HW output (bottom row) is a submix of its own.
So in TM FX you have as many submixes as you have HW outputs.

To solve your 1st problem.
If I understand you correctly, then every Musician wants to have a submix of its own.
i.e. a little bit more from the drums, a little bit less from the Bass or whatever.

In TM FX "submix mode":
1. you click to the HW output (bottom row) where i.e. "Phones1" are connected to for "Musician1".
2. next you change the level of
    - HW inputs (audio from devices which are connected to the inputs of your recording interface, top row in TM FX)
    - SW playback channels (audio from PC/applications, like DAW, maybe a backing track)
3. If your recording interface has a FX chip then you can also EQ the inputs or outputs to create a smoother submix.
    Note: this does not change the recorded audio in the DAW, audio from HW inputs will be passed through
    unmodified to the application. So your recording is safe !
4. This you perform for every HW output to which i.e. phones are connected to for every musician.
    By this every musician gets the submix that he want to hear.

These settings and routing you store to one of the 8 Snapshots in TM FX "-> see below "Snapshots" on the right side.
All settings can also be safed to a file on disk File -> Save Snapshot As and later be restored.
You can also save the workspace, the manual tells you the difference whats being saved in Snapshots and Workspaces.
Additionally you can save "Snapshot" AND "Workspace" to up to 32 "Workspace Quick Select" slots together.
This enables you to quickly restore "Snapshot" and "Workspace" together, also with an easy key command ALT-1 for the 1st workspace, ALT-2 for the 2nd workspace.
This is nice to restore TM FX settings quickly to your last proven state.
Useful if you work i.e. in different projects or environments.

Now to your next requirement, you want to mute inputs if you playback from DAW.
Well, then mute the inputs, create proper routing and save it to snapshot 2.
Then for recording you use snapshot1 and for playback snapshot 2.

If you want control room features for your outputs either controlled
- by mouse in the TM FX control room or
- by ARC USB
then assign them by using the "Assign" button.
Only one HW output "Main Out" is being controlled by FM FX control room at a time.
With "Monitor B" you can switch between "Main Out" and "Main Out B".

You can of course also save to another snapshot, that the phones HW output is "Main Out", then the Control Room features work on the phones. Its free assignable to you and storeable in Snapshots, whats Main Out, Main Out B, Phones 1..4, etc.

Further information found in the manual.
Each manual describes the TM FX capabilities of the device that you own.

Additionally you can watch RME tutorial videos which are very compact and to the point to one topic,
so very easy and time efficient to watch, you can find a collection of new and old videos here in my blog:
https://www.tonstudio-forum.de/blog/ind … al-Videos/

I also put together a step by step guide for TM FX 1st setup.
Maybe the one or other point might also be interesting to you: https://www.tonstudio-forum.de/blog/ind … rnal-equi/
Its available in two languages, german and english.

If you should really need ASIO direct monitoring as a feature (ADM), then I remember a thread month ago where RME told, that ADM runs best, when the inputs are mono and the outputs are stereo.
In TM FX you can easily reset the Mix to have exactly this ADM layout:
TM FX -> Options -> Reset Mix -> Set inputs mono / outputs stereo (ADM)

But I am not sure whether ADM is really what you need in your situation.
You want to control the gain for the different submixes and have zero latency monitoring.
You do not want to control the gain in the DAW for the inputs to record at lower input level.
I think this would happen if I am not wrong.

For recording I personally would use the optimum level for each input and would adjust
the final level later in the mixing phase.

For the phones mixes simply use TotalMix FX for zero latency submixes.

One more thing to note. If your backing track consists of different tracks, then you can send them in the DAW to different Outputs of your Recording Interface.
The Audio will be send to TM FX first to the SW Playback Channels (not directly to a HW output).
This is very useful because then you can
a) reroute as you need to one or more real HW outputs and
b) also adjust the volume for every submix (HW output) of your recording device.

And this is the strength of TM FX. Its perfect to create phones submixes.
You can ideally mix audio directly coming from HW inputs latency free (near zero-latency)
You can alter the signal with FX only for the submix (your recorded signal remains unaltered)

You can add audio coming from the PC (Windows, DAW, some other application) and here is TM FX
an additional layer between Audio Application and the real HW outputs. Therefore you can reroute so nicely
and create perfect submixes for phones etc as you need.

I tried to "capture" this in the following drawing from my step by step primer for TM FX:

So you are only looking for a way how to copy a submix from one HW output to the other ?

Its easy, see manual.

For an UFX+ in chapter 27: http://www.archiv.rme-audio.de/download … plus_e.pdf

Georg

when I use ADM, I only get the DAW ciue sends (Nuendo still provides only 4 stereo cues-additional sends do not have the Talkback/click option)...and the hardware ins are always on.....
Pierre

You can copy / paste from a HW input to SW playback channel, but only if both are mono or stereo.
But it has limitations, you can only copy level and pan.
If you copy / paste then take care
- copy <channel>
- paste input mix <channel>

But it can not copy FX settings, if you have a FX chip in TM FX, because this is only possible on real
physical I/O ports (HW inputs, HW outputs), not for audio which comes through the connection to the PC.

Otherwise see comments from RME, Mr Carstens (MC).

What I would do .. perform those settings in the DAW ... and voila, you do not have to split every channel into a dedicated SW playback channel and have all these issues.

That is technically impossible, sorry.

Again no. That is basically the same as before, it breaks the internal structure. You can copy an input to output routing to a different input (so only the input changes), and an output with all inputs and software playbacks to another output (so only the destination changes). So far nobody needed to access software outputs or their routing in such a way as you do, which also seems to mean that everyone else works differently.

Implementing something like this will have strong limitations (for example the independent EQ of input and hardware output cannot simply be merged to the hardware output alone).

That is not how it is implemented. It would also require to throw away the current playback fader position, which might have been in effect and used before. So it would only work correctly if there is zero playback as long as the inputs are used. But this is all theoretical, we will not implement such things, sorry.