Topic: Total mix with live mix and streaming feed.

I'm a new Totamix/fireface  user and despite viewing scores of videos, reading the manual, and surfing; I still have a difficult time with what should be a simple task. Basically, how I think Totalmix loopback works and how it really works isn't the same.

I need to have multiple live hardware and software inputs on a live mix and the ability to monitor it, at a minimum,  on  the headphone output. I also need to have the audio stream sent through a streaming app for a video feed. OBS has a 3 frame buffer, which means video and audio are off 150 milliseconds or so. It would make it easier to dial in the delay if I could monitor that mix also.
Here's what I've tried to do. Mix microphones / hardware inputs and software sources with the send fx bar set to 0db and output two submixes to two unused outputs, one with Fx receive off, the other with Fx on, essentially a wet and dry bus. Fx echo is set to 100ms. I then tried to loopback both outputs to their respective inputs and send those inputs  to different  headphone outs  to monitor the audio. It doesn't work and I don't know why. I verified that the Fx is working as expected. Can someone provide a more detailed explanation of how loopback works or why this configuration fails? How can I have a live mix for musicians on site and a synced(delayed) mix for off site viewers?

2 (edited by waedi 2024-11-30 04:51:50)

Re: Total mix with live mix and streaming feed.

first, please tell wich interface model you use.

If you send the input and software signal to the output and to the FX, the output will have dry and wet signal mixed, but you want to have FX wet signal only on one output for the OBS stream, right ?

On those input and software signals make the FX send fader fully up.
Don't rout the signal to output OBS.
On the hardware output channel with OBS Loopback lift up the FX fader fully.  Only the Fx signal with your delay is there.
As expected.
Don't mess up the routing.
The FX-return fader does not replace the signal with the FX, it just ads Fx to that channel.
But in your case you want the only-wet signal.

M1-Sequoia, Madiface Pro, Digiface USB, Babyface silver and blue

Re: Total mix with live mix and streaming feed.

waedi wrote:

first, please tell wich interface model you use.

If you send the input and software signal to the output and to the FX, the output will have dry and wet signal mixed, but you want to have FX wet signal only on one output for the OBS stream, right ?

On those input and software signals make the FX send fader fully up.
Don't rout the signal to output OBS.
On the hardware output channel with OBS Loopback lift up the FX fader fully.  Only the Fx signal with your delay is there.
As expected.
Don't mess up the routing.
The FX-return fader does not replace the signal with the FX, it just ads Fx to that channel.
But in your case you want the only-wet signal.

The interface is a UFX III.
Yes, I want a wet only to OBS or ZOOM to compensate for the video delay, and a dry mix for live in room monitors.
I had it  routed like above but was trying to get the wet signal to headphones to try and dial in the delay. I thought if I looped back it would allow me to add another route from the corresponding HW input fader like it was hardwired.

4 (edited by waedi 2024-11-30 05:41:23)

Re: Total mix with live mix and streaming feed.

No. Loppback does not loop back the signal into Totalmix that is too dangerous for feedback loop.
The Loopback does forward the signal into the input channel for the DAW (or OBS).

The wet signal is available on every output channel (also headphone) by lifting the FX-return fader.

M1-Sequoia, Madiface Pro, Digiface USB, Babyface silver and blue

Re: Total mix with live mix and streaming feed.

waedi wrote:

No. Loppback does not loop back the signal into Totalmix that is too dangerous for feedback loop.
The Loopback does forward the signal into the input channel for the DAW (or OBS).

The wet signal is available on every output channel (also headphone) by lifting the FX-return fader.

That's what I suspected but numerous places, including RME's videos, say that loopback brings the output to the corresponding. input. What it appears to do is virtualize the hardward output and make it available to the OS.

FWIW, they could route it to the corresponding input if they excluded the ability of sending it back to the same channel. It would allow for really complex routing schemas and minimize the risk of a software positive feedback loop.

Regardless, they really should explain in detail what loopback does. The manual and the videos clearly give the wrong impression.

Thanks for your help; it cleared things up.

6 (edited by ramses 2024-11-30 17:06:28)

Re: Total mix with live mix and streaming feed.

> The manual and the videos clearly give the wrong impression.

To which part of the manual and to which video do you refer to?

Loopback can be implemented pre- and post-fader, you can see this is the block diagram or it is also documented in my Excel.
https://forum.rme-audio.de/viewtopic.php?id=35156

By loopback you can record the submix of the particular HW output on the corresponding HW input.

The audio signal which might be connected on the corresponding HW input is not lost, you can route it to another output and if you want also enable loopback there to record it on the corresponding HW input.

You kind of complained that you can not route the looped back signal to another submix, maybe it is sufficient to you to copy or mirror this submix to another HW output?

BR Ramses - UFX III, 12Mic, XTC, ADI-2 Pro FS R BE, RayDAT, X10SRi-F, E5-1680v4, Win10Pro22H2, Cub14

Re: Total mix with live mix and streaming feed.

ramses wrote:

> The manual and the videos clearly give the wrong impression.

To which part of the manual and to which video do you refer to?

Loopback can be implemented pre- and post-fader, you can see this is the block diagram or it is also documented in my Excel.
https://forum.rme-audio.de/viewtopic.php?id=35156

By loopback you can record the submix of the particular HW output on the corresponding HW input.

The audio signal which might be connected on the corresponding HW input is not lost, you can route it to another output and if you want also enable loopback there to record it on the corresponding HW input.

You kind of complained that you can not route the looped back signal to another submix, maybe it is sufficient to you to copy or mirror this submix to another HW output?

I'm mostly trying to understand how the device works. I'm sorry if it comes across as complaining.

Here's one of several videos : https://www.youtube.com/watch?v=Xpt7LczeAvU


The issue I am trying to work around is that a looped back output can't be routed to the  headphones or monitor. I didn't consider a mirror or copy, but it seems like I could make that work.

Re: Total mix with live mix and streaming feed.

simply copy the submix to phones submix.

BR Ramses - UFX III, 12Mic, XTC, ADI-2 Pro FS R BE, RayDAT, X10SRi-F, E5-1680v4, Win10Pro22H2, Cub14

Re: Total mix with live mix and streaming feed.

Concerning loopback and description in manual and videos.

I couldn't get it even after reading manual few times, watching videos few times. I had to ask here on forum, how it works. And I am 40+ years in computers and have PhD. in electrical engineering....

So, it seems to me, it is not because of some reason very clear, how that works from description in manual and videos.

10 (edited by ramses 2024-11-30 22:03:27)

Re: Total mix with live mix and streaming feed.

Kubrak wrote:

Concerning loopback and description in manual and videos.

I couldn't get it even after reading manual few times, watching videos few times. I had to ask here on forum, how it works. And I am 40+ years in computers and have PhD. in electrical engineering....

So, it seems to me, it is not because of some reason very clear, how that works from description in manual and videos.

Seriously? It's all in the manual, even with a nice block diagram included.

Frankly, I don't understand what is incomprehensible about this. You press loopback and can record the mix of your hardware output at the corresponding input to save you having to plug in a loopback cable (from the output to the input). It says so right there. If necessary, just ask support or in the forum and you'll get an answer.

https://www.dropbox.com/scl/fi/i7bttbr4w9fbg3xvo0wev/2024-11-30-Loopback.jpg?rlkey=crulknq7ww3lqtswxx7852g0n&st=vpie4am1&dl=1

Or look at that video .. very nicely explained in less than 2 minutes:
https://www.youtube.com/watch?v=m-Zeruz-9Zk

BR Ramses - UFX III, 12Mic, XTC, ADI-2 Pro FS R BE, RayDAT, X10SRi-F, E5-1680v4, Win10Pro22H2, Cub14

Re: Total mix with live mix and streaming feed.

ramses wrote:
Kubrak wrote:

Concerning loopback and description in manual and videos.

I couldn't get it even after reading manual few times, watching videos few times. I had to ask here on forum, how it works. And I am 40+ years in computers and have PhD. in electrical engineering....

So, it seems to me, it is not because of some reason very clear, how that works from description in manual and videos.

Seriously? It's all in the manual, even with a nice block diagram included.

Frankly, I don't understand what is incomprehensible about this. You press loopback and can record the mix of your hardware output at the corresponding input to save you having to plug in a loopback cable (from the output to the input). It says so right there. If necessary, just ask support or in the forum and you'll get an answer.

When I got my UCX II three years ago it took some time before my brain could wrap itself around TotalMix FX and digest it. TotalMix FX is a pro tool and pro tools usually have a learning curve in whatever field.

Re: Total mix with live mix and streaming feed.

ramses wrote:
Kubrak wrote:

Concerning loopback and description in manual and videos.

I couldn't get it even after reading manual few times, watching videos few times. I had to ask here on forum, how it works. And I am 40+ years in computers and have PhD. in electrical engineering....

So, it seems to me, it is not because of some reason very clear, how that works from description in manual and videos.

Seriously? It's all in the manual, even with a nice block diagram included.

Frankly, I don't understand what is incomprehensible about this. You press loopback and can record the mix of your hardware output at the corresponding input to save you having to plug in a loopback cable (from the output to the input). It says so right there. If necessary, just ask support or in the forum and you'll get an answer.

https://www.dropbox.com/scl/fi/i7bttbr4w9fbg3xvo0wev/2024-11-30-Loopback.jpg?rlkey=crulknq7ww3lqtswxx7852g0n&st=vpie4am1&dl=1

Or look at that video .. very nicely explained in less than 2 minutes:
https://www.youtube.com/watch?v=m-Zeruz-9Zk

In my case, I wasn't trying to record anything and thus didn't concentrate on that part of the manual. My bad.
I also watched the that video as well as several others. In the video he says something along the line of "loop back functions like putting a cable from the output to the input but lies entirely within Total Mix". Hence my confusion because the implementation of loop back doesn't really do that. You are correct,  the diagram shows that loop back pretty much leaves the hardware input out of the loop. Had I seen that, I would have not been confused. 
Now I know. I've done as suggested and copied the mix to the second phones jack. Thank all of you.

13 (edited by ramses 2024-11-30 23:12:46)

Re: Total mix with live mix and streaming feed.

Hi Muffin. Sorry, I don’t mean to sound presumptuous or overconfident, but I see things a bit differently. Before purchasing my UFX in 2014, I watched all the Syntax videos—they were quick and easy to go through. With that foundation, reading the manual became much easier. Along the way, I took a few notes and focused more closely on the sections about driver installation and configuring TotalMix FX and the Control Room. Other sections I just skimmed.

After that, I had a good understanding of where to find the relevant information in the manual. The manuals are well-structured, and with the search function in any PDF viewer, it’s easy to quickly look up specific keywords.

I don’t recall ever having any issues or finding the system complicated. On the contrary, I found it logically designed and well-thought-out.

If you know that audio can be recorded unaltered directly from the hardware inputs, understand the role of SW playbacks, and see how easily routing can be configured in submix mode and saved in Snapshots and Workspaces, or how and why to assign output channels into control room, then you’ve already grasped the essential basics—and it’s really not that difficult.

Every DAW is much more complicated due to a lot more functions and features. The Cubase 14 Pro operation manual has 1560 pages and the plugin reference has 245 pages and this is not all...

BR Ramses - UFX III, 12Mic, XTC, ADI-2 Pro FS R BE, RayDAT, X10SRi-F, E5-1680v4, Win10Pro22H2, Cub14

Re: Total mix with live mix and streaming feed.

ramses wrote:

Hi Muffin. Sorry, I don’t mean to sound presumptuous or overconfident, but I see things a bit differently. Before purchasing my UFX in 2014, I watched all the Syntax videos—they were quick and easy to go through. With that foundation, reading the manual became much easier. Along the way, I took a few notes and focused more closely on the sections about driver installation and configuring TotalMix FX and the Control Room. Other sections I just skimmed.

After that, I had a good understanding of where to find the relevant information in the manual. The manuals are well-structured, and with the search function in any PDF viewer, it’s easy to quickly look up specific keywords.

I don’t recall ever having any issues or finding the system complicated. On the contrary, I found it logically designed and well-thought-out.

If you know that audio can be recorded unaltered directly from the hardware inputs, understand the role of SW playbacks, and see how easily routing can be configured in submix mode and saved in Snapshots and Workspaces, or how and why to assign output channels into control room, then you’ve already grasped the essential basics—and it’s really not that difficult.

Every DAW is much more complicated due to a lot more functions and features. The Cubase 14 Pro operation manual has 1560 pages and the plugin reference has 245 pages and this is not all...

My post was just to chime in on Kubrak's experience which mirrors mine, though coming from another field of research and non-pro-audio experience. As I wrote: It has a learning curve, which is to be expected for a pro tool. That does not mean there is no room for improvement, though, and RME do that continuously with the focus on pro audio users (which I'm definitely not).

I'm not sure why you had to bring up reading the manuals and other documentation, something I do as a pro in another field. I'm sure Kubrak does the the same. To be honest, I found it condescending.

15 (edited by ramses 2024-11-30 23:57:20)

Re: Total mix with live mix and streaming feed.

> To be honest, I found it condescending.

I’m sorry if my response came across as condescending—that wasn’t my intention at all. It seems there might have been a misunderstanding.

Just to clarify: Music is a hobby for me, but I’m also a professional in other fields. I’ve been through the learning process too, starting as a beginner 10 years ago. Like everyone, I had to figure things out, but I didn’t face major difficulties.

When I mentioned videos and manuals, it was simply to highlight that a certain methodology can be helpful. The information is available—you just need to find a way to make the most of it.

So, if something seems difficult, remember that others might have a different perspective or are just sharing what worked for them. That’s what I was trying to do.

Instead of seeing it as criticism or assuming I was being condescending, I’d appreciate it if you could focus on the topic and explain what exactly is unclear about the examples—that would help me understand your point better.

BR Ramses - UFX III, 12Mic, XTC, ADI-2 Pro FS R BE, RayDAT, X10SRi-F, E5-1680v4, Win10Pro22H2, Cub14

Re: Total mix with live mix and streaming feed.

ramses wrote:

> To be honest, I found it condescending.

Then you misunderstood me completely.

BTW .. I am also pro in other fields, music is only my hobby. How can it be condescending?
I also started as beginner 10y ago.

I referred to videos and manuals only for one reason .. that a certain methodology can help.
The information is there .. you only need to find a way to make the best use out of it.

As I wrote, and with this I'll take my leave from this particular part of this thread.

Re: Total mix with live mix and streaming feed.

Regarding the quote, I rephrased (but was not fast enough).

BR Ramses - UFX III, 12Mic, XTC, ADI-2 Pro FS R BE, RayDAT, X10SRi-F, E5-1680v4, Win10Pro22H2, Cub14

Re: Total mix with live mix and streaming feed.

My setup is simpler, but here's what I do:

1. Use an empty output (like Madi 1/2) in loopback. It can be set to mirror the main out, and I also mirror the main out to the headphones so those are hearing the same thing. Make sure that fader is up, say 0db.

2. When setting the audio source properties in OBS, make sure you don't have that downmix checkbox checked. This then lets you select the actual hardware input (I choose Madi 1/2). Note that you can use the compressor on that output which can be helpful.

Lastly, to check the audio delay needed in OBS, I record a little clip of a clap and then bring it into Final Cut to see about how many frames off the audio is from the visual clap. Don't worry about the math too much, just set it to say 180ms. Do the same thing again and see if it lines up. If visual clap to early, add say 30ms delay, check again. It probably takes 5 minutes to do this and then you are done.

Also, I don't think you really want to monitor with that audio delay because you'd be hearing that 180ms delay on your voice.

Good luck!

19 (edited by Muffin 2024-12-02 18:53:35)

Re: Total mix with live mix and streaming feed.

agotheridge wrote:

My setup is simpler, but here's what I do:

1. Use an empty output (like Madi 1/2) in loopback. It can be set to mirror the main out, and I also mirror the main out to the headphones so those are hearing the same thing. Make sure that fader is up, say 0db.

2. When setting the audio source properties in OBS, make sure you don't have that downmix checkbox checked. This then lets you select the actual hardware input (I choose Madi 1/2). Note that you can use the compressor on that output which can be helpful.

Lastly, to check the audio delay needed in OBS, I record a little clip of a clap and then bring it into Final Cut to see about how many frames off the audio is from the visual clap. Don't worry about the math too much, just set it to say 180ms. Do the same thing again and see if it lines up. If visual clap to early, add say 30ms delay, check again. It probably takes 5 minutes to do this and then you are done.

Also, I don't think you really want to monitor with that audio delay because you'd be hearing that 180ms delay on your voice.

Good luck!

I did similar for my video calls except that the delay is done in the UCX II using Echo FX, and like you I certainly don't monitor using the delayed audio as that would be awful to listen to.

The Echo has steps of 10 ms but that works fine for my usage.

Re: Total mix with live mix and streaming feed.

agotheridge wrote:

My setup is simpler, but here's what I do:

1. Use an empty output (like Madi 1/2) in loopback. It can be set to mirror the main out, and I also mirror the main out to the headphones so those are hearing the same thing. Make sure that fader is up, say 0db.

2. When setting the audio source properties in OBS, make sure you don't have that downmix checkbox checked. This then lets you select the actual hardware input (I choose Madi 1/2). Note that you can use the compressor on that output which can be helpful.

Lastly, to check the audio delay needed in OBS, I record a little clip of a clap and then bring it into Final Cut to see about how many frames off the audio is from the visual clap. Don't worry about the math too much, just set it to say 180ms. Do the same thing again and see if it lines up. If visual clap to early, add say 30ms delay, check again. It probably takes 5 minutes to do this and then you are done.

Also, I don't think you really want to monitor with that audio delay because you'd be hearing that 180ms delay on your voice.

Good luck!

The only reason I wanted to monitor the delayed output is to make life a little easier when setting up the video, but unless you change the frame rate it shouldn't be needed  very often. Additionally, I was trying to learn more about the software. Although I contend that loop back could be more powerful if it worked the way I thought it worked, it doesn't matter, as copying the mix works.

I actually bypass OBS's audio, seems superfluous for my use. I also think having a dedicated delay, as opposed to echo would be a nice feature. 10ms increments is OK but 5ms increments would be better for faster frame rates.

21 (edited by Muffin 2024-12-02 20:39:24)

Re: Total mix with live mix and streaming feed.

SwampFox wrote:
agotheridge wrote:

My setup is simpler, but here's what I do:

1. Use an empty output (like Madi 1/2) in loopback. It can be set to mirror the main out, and I also mirror the main out to the headphones so those are hearing the same thing. Make sure that fader is up, say 0db.

2. When setting the audio source properties in OBS, make sure you don't have that downmix checkbox checked. This then lets you select the actual hardware input (I choose Madi 1/2). Note that you can use the compressor on that output which can be helpful.

Lastly, to check the audio delay needed in OBS, I record a little clip of a clap and then bring it into Final Cut to see about how many frames off the audio is from the visual clap. Don't worry about the math too much, just set it to say 180ms. Do the same thing again and see if it lines up. If visual clap to early, add say 30ms delay, check again. It probably takes 5 minutes to do this and then you are done.

Also, I don't think you really want to monitor with that audio delay because you'd be hearing that 180ms delay on your voice.

Good luck!

The only reason I wanted to monitor the delayed output is to make life a little easier when setting up the video, but unless you change the frame rate it shouldn't be needed  very often. Additionally, I was trying to learn more about the software. Although I contend that loop back could be more powerful if it worked the way I thought it worked, it doesn't matter, as copying the mix works.

I actually bypass OBS's audio, seems superfluous for my use. I also think having a dedicated delay, as opposed to echo would be a nice feature. 10ms increments is OK but 5ms increments would be better for faster frame rates.

RME added room EQ with delay to a number of device, including my UCX II, but the max delay of 42.5 ms was too short for my usage for video calls where I talk. So the Echo workaround, that I read on this forum, is what I still have to use. Listening to my own sound about 200 ms delayed is not pleasant.

Re: Total mix with live mix and streaming feed.

Muffin wrote:
ramses wrote:

> To be honest, I found it condescending.

Then you misunderstood me completely.

BTW .. I am also pro in other fields, music is only my hobby. How can it be condescending?
I also started as beginner 10y ago.

I referred to videos and manuals only for one reason .. that a certain methodology can help.
The information is there .. you only need to find a way to make the best use out of it.

As I wrote, and with this I'll take my leave from this particular part of this thread.

@ramses I got your e-mail and I apologize for my tone in my post.

Re: Total mix with live mix and streaming feed.

No problem, simply misunderstanding, don't worry. But thanks for your kind posting :-)

BR Ramses - UFX III, 12Mic, XTC, ADI-2 Pro FS R BE, RayDAT, X10SRi-F, E5-1680v4, Win10Pro22H2, Cub14

Re: Total mix with live mix and streaming feed.

SwampFox wrote:
agotheridge wrote:

My setup is simpler, but here's what I do:

1. Use an empty output (like Madi 1/2) in loopback. It can be set to mirror the main out, and I also mirror the main out to the headphones so those are hearing the same thing. Make sure that fader is up, say 0db.

2. When setting the audio source properties in OBS, make sure you don't have that downmix checkbox checked. This then lets you select the actual hardware input (I choose Madi 1/2). Note that you can use the compressor on that output which can be helpful.

Lastly, to check the audio delay needed in OBS, I record a little clip of a clap and then bring it into Final Cut to see about how many frames off the audio is from the visual clap. Don't worry about the math too much, just set it to say 180ms. Do the same thing again and see if it lines up. If visual clap to early, add say 30ms delay, check again. It probably takes 5 minutes to do this and then you are done.

Also, I don't think you really want to monitor with that audio delay because you'd be hearing that 180ms delay on your voice.

Good luck!

The only reason I wanted to monitor the delayed output is to make life a little easier when setting up the video, but unless you change the frame rate it shouldn't be needed  very often. Additionally, I was trying to learn more about the software. Although I contend that loop back could be more powerful if it worked the way I thought it worked, it doesn't matter, as copying the mix works.

I actually bypass OBS's audio, seems superfluous for my use. I also think having a dedicated delay, as opposed to echo would be a nice feature. 10ms increments is OK but 5ms increments would be better for faster frame rates.

30 fps is roughly 33ms per frame. So as long as the audio hits in say a 33ms window of that frame, that's the best you can do anyway. I think!

Re: Total mix with live mix and streaming feed.

agotheridge wrote:

30 fps is roughly 33ms per frame. So as long as the audio hits in say a 33ms window of that frame, that's the best you can do anyway. I think!

10 ms is fine for 30FPS.
I usually stream at  60 FPS, which is 16.67 msec per frame. It makes being one frame off barely noticeable even when looking very closely.  5 millisec increments would make sync'ing to the frame more reliable at 60fps. Having said that, it's not the end of the world to be a frame off. Either way it would be nice to have a delay setting that wasn't part of the echo effect as it would be more straight forward to 'read'.

Re: Total mix with live mix and streaming feed.

SwampFox wrote:

10 ms is fine for 30FPS.
I usually stream at  60 FPS, which is 16.67 msec per frame. It makes being one frame off barely noticeable even when looking very closely.  5 millisec increments would make sync'ing to the frame more reliable at 60fps. Having said that, it's not the end of the world to be a frame off. Either way it would be nice to have a delay setting that wasn't part of the echo effect as it would be more straight forward to 'read'.

Or you could do 1ms increments in the advanced audio properties of the OBS audio device. Just seems simpler to me.

Re: Total mix with live mix and streaming feed.

agotheridge wrote:
SwampFox wrote:

10 ms is fine for 30FPS.
I usually stream at  60 FPS, which is 16.67 msec per frame. It makes being one frame off barely noticeable even when looking very closely.  5 millisec increments would make sync'ing to the frame more reliable at 60fps. Having said that, it's not the end of the world to be a frame off. Either way it would be nice to have a delay setting that wasn't part of the echo effect as it would be more straight forward to 'read'.

Or you could do 1ms increments in the advanced audio properties of the OBS audio device. Just seems simpler to me.

smile Kinda made me laugh thinking that adding additional audio routes through OBS would be simpler.