questions about setting up signal chain for recording at home

[gene12586]

Hi all,

I have some very beginner questions about setting up my home studio signal chain with some equipment I have access to. Sorry in advance for the lengthy message. Here's the equipment:
Audio interface: Motu896mk3
Computer: 2020 27 inch Imac w/ 3.3 ghz, 6 core processor (w/ garageband)
Mics: Sure sm7b, Sure sm58, Sure sm57, two Rhode condenser mics
Monitors: Adam A7x monitors
Headphones
Mixer: Yamaha stagepas 600i

For a project I'm working on, I want to first be able to record myself playing my acoustic-electric guitar while singing at the same time, and I want some reverb on the vocals. Then I want to overdub with a doumbek (middle eastern hand percussion instrument). Then after I record that, I want to overdub with an electric guitar. I should also mention that I currently have no sound treatment on my walls, but I plan to be getting some in a few months. So several questions:

1) I've heard that direct monitoring is preferred to input monitoring to avoid lag... but as I said I also want reverb when I record the vocals while I'm playing acoustic. So it seems to me then that a logical thing to do then is to have a signal chain as such: Guitar and vocals microphone plugged into mixer, then mixer plugged into audio interface, then one line going from interface to monitors (or headphones) and another line going to computer. Is this doable? The issue is that whenever I watch videos of people recording acoustic guitar and vocals, I usually see that the acoustic is miced and both the acoustic mic and the vocals mic feed into an interface, then interface goes to computer where effects are applied (in this case, reverb), then computer back to interface, and then interface to monitors/headphones. But isn't this the input monitoring approach? So there would be lag, and wouldn't that interfere with the recording? Overall, what are the pros and cons of using the direct monitoring setup I proposed vs. the input monitoring setup? Which would work better in terms of quality and so on? Also, I tend to not be a huge fan of reverb applied using computer software; I like the sound of the reverb on the mixer better... so this would seem to be another reason that the direct monitoring option I presented above might be better.
2) For the two options I listed above, when I'm just recording the acoustic guitar and vocals, would it be better to monitor through the monitors or through headphones? I realize sound quality will obviously be better through the monitors.... My logic tells me that for the direct monitoring approach, I can get away with monitoring using the monitors (instead of headphones)because there will be no lag. Is that right? But what about for the input monitoring approach? Wouldn't using monitors instead of headphones screw up the recording because of the lag? So would I use headphones for the input monitoring approach?
3) When I overdub the doumbek, I'll mic the doumbek and feed it into the interface, and then the interface will either send out to the monitors/headphones (in the case of direct monitoring approach) or will go to computer and then back to interface and then out to monitors/headphones (in the input monitoring approach). But at the same time since I'm overdubbing over the pre-recorded acoustic/vocals track, I'll also have the pre-recorded track feeding from the computer to the interface and then to the monitors/headphones. Regardless or whether I go the direct or input monitoring approach, will having sound coming from two different sources be a problem? In other words, on the one hand sound will come from the computer for the pre-recorded stuff, and on the other hand the sound from the doumbek will come from the doumbek... so will the two sound sources be combined smoothly into one in the interface and then sent out together to monitors/headphones? Or am I missing something here?
4) As mentioned, I'm planning to then overdub with electric guitar (I plan to plug the electric guitar into my guitar amplifier, mic the amplifier, and then run the mic into the interface. What I'm confused about is how it's possible for one to overdub with an electric guitar in this fashion. I'll have the prerecorded track of the acoustic guitar and vocals coming from the computer to the interface and then out of the monitors (or headphones), but then I'll also have the electric guitar sound live coming out of the amplifier and this will be routed to the interface, which would then potentially go to the monitors/headphones as well as to the computer (but then I would have the electric guitar sound coming out of two different sources, both the guitar amp and the monitors/headphones, which I imagine won't work). So is the idea that I need to somehow make it so that the amplifier to interface sound doesn't then go to the monitors/headphones and instead only goes to the computer? Is that the way to do it? Is that possible? Or is there some other way that's typical to do this? Also, for doing this, is it preferable to monitor through the monitors or headphones? If headphones are the better way to go, how would this even be possible if the sound from the electric guitar isn't being fed into the headphones? (in other words, I wouldn't be able to hear the electric guitar sound because I have headphones on, and the only sound coming from the headphones is the pre-recorded computer track)
5) In comparison to the direct monitoring approach I listed in question 1 above, I thought of another way to do the signal chain with direct monitoring, and I'm wondering if this would be better: acoustic guitar and microphone for vocals both fed into the mixer, then the mixer feeds out to the speakers that came with the mixer, then I mic the mixer speakers, then feed the mic to the interface, then the interface goes to the monitors/headphones and also to the computer. Would this be a better approach than the other direct monitoring approach I mentioned in question 1 above? Or no? I don't have any compelling reason for thinking it would be better or worse; I just thought I'd ask out of curiosity.

Any help anyone can give would be so very much appreciated.

Thanks,

Gene

 

[rob aylestone]

The post has lots of questions. But surely you’ve tried it? Direct vs computer monitoring is something only you can decide. You plug things in and see if the latency bothers you? I never use the direct feature on my interface ever. My latency is low enough I can play comfortably. My concert pianist collaborator is hyper sensitive to it and any delay seems to throw him. Surely as you know what you want, have the gear, you just give it a shot and see what the results are. You seem to be micromanaging the process instead of experimenting. Your iMac might be speedy or a snail, we don’t know. You have multi stage questions that have a possibility per section, so while the process could work, it could stall almost anywhere. You have some pleasant mics, so experiment. There are no prescriptions, you do what works for you. Is it preferable? Might be is the only answer viable.

one thing that is certain is that recording while monitoring via speakers is rarely workable. closed back or iem is the sensible route for separation.

be aware that many videos are composites. Some people are very good at miming. If it sounds really clean and wonderful, very often it’s been cheated. It’s perfectly possible to record multiple sources live, but it takes care and attention and experience. Don’t always believe what you see.

 

[gene12586]

The post has lots of questions. But surely you’ve tried it? Direct vs computer monitoring is something only you can decide. You plug things in and see if the latency bothers you? I never use the direct feature on my interface ever. My latency is low enough I can play comfortably. My concert pianist collaborator is hyper sensitive to it and any delay seems to throw him. Surely as you know what you want, have the gear, you just give it a shot and see what the results are. You seem to be micromanaging the process instead of experimenting. Your iMac might be speedy or a snail, we don’t know. You have multi stage questions that have a possibility per section, so while the process could work, it could stall almost anywhere. You have some pleasant mics, so experiment. There are no prescriptions, you do what works for you. Is it preferable? Might be is the only answer viable.

one thing that is certain is that recording while monitoring via speakers is rarely workable. closed back or iem is the sensible route for separation.

be aware that many videos are composites. Some people are very good at miming. If it sounds really clean and wonderful, very often it’s been cheated. It’s perfectly possible to record multiple sources live, but it takes care and attention and experience. Don’t always believe what you see.

Thanks for your reply. I edited the original post to include the Imac specs.
What you're saying about micromanaging makes sense. I've been setting up the room in the best way I can, and definitely plan to do some experimenting in the near future when I have things set up... but I thought I would get any feedback I can off the bat to see if anyone has any strong opinions for the different options I listed.

Why is recording while monitoring via speakers rarely workable? Is it because most rooms don't have proper sound treatment? Or something else?

 

[ecc83]

One reason why loudspeaker monitoring whilst recording is a bad idea is that you will get acoustic feedback with a live mic or at the very least a coloured sound. You might get away with a tiny bit of 'foldback' to keep you in time while you add other parts but you do not want that recorded, as Rob says, you need some closed back headphones.

The Yamaha mixer is of little use, except if you want some 'comfort' reverb on voice then you should be able to send a feed from the MOTU to the mixer and monitor it on cans.

Rob is right again about having a 'king go! Just hook stuff up and play and listen. Are you afraid you will damage something? Well you won't. You don't appear to have any capacitor mics (they are good tho!) so just leave phantom power off, a potential source of bother but in reality, not a lot.

Dave.

 

[gene12586]

One reason why loudspeaker monitoring whilst recording is a bad idea is that you will get acoustic feedback with a live mic or at the very least a coloured sound. You might get away with a tiny bit of 'foldback' to keep you in time while you add other parts but you do not want that recorded, as Rob says, you need some closed back headphones.

The Yamaha mixer is of little use, except if you want some 'comfort' reverb on voice then you should be able to send a feed from the MOTU to the mixer and monitor it on cans.

Rob is right again about having a 'king go! Just hook stuff up and play and listen. Are you afraid you will damage something? Well you won't. You don't appear to have any capacitor mics (they are good tho!) so just leave phantom power off, a potential source of bother but in reality, not a lot.

Dave.

Thanks for your feedback Dave.
When you say what you say about feeding from the MOTU to the mixer for reverb, I'm a little confused.... Are you saying the chain is: vocals mic and acoustic guitar > interface > then signal goes to both computer and to mixer... and then I monitor by plugging headphones to the mixer?? Thanks.

 

[mjbphotos]

I understand that some people really *need* reverb when recording vocals, however, if you do it long enough, you will realize its not necessary. The problem with recording your tracks with reverb from the mixer is that you can't "undo" that reverb - you are stuck with how it sounds when tracked.
The other thing to consider is that the Stagepass mixer does not have a stereo input section for mic/guitar (no pan control on the channels), so if you have both vocals and guitar plugged into it, your output to your interface will be mono - guitar + vocals on same mono track in your DAW.

 

[ecc83]

Thanks for your feedback Dave.
When you say what you say about feeding from the MOTU to the mixer for reverb, I'm a little confused.... Are you saying the chain is: vocals mic and acoustic guitar > interface > then signal goes to both computer and to mixer... and then I monitor by plugging headphones to the mixer?? Thanks.

Er? Yes! The MOTU should be able to send a "Zero Latency" output to a line input on the mixer*. You can then add reverb to THAT input and hear it from the mixers headphone jack.
OR!

You could run EVERYTHING through the mixer and feed that to the interface for recording. That would need the mixer to be able to give you a 'clean feed' of the reverb so you did not record it.

BUT!

I don't have the gear. YOU do! FCS plug it all up and have a bloody DO!

Dave.

 

[ecc83]

I understand that some people really *need* reverb when recording vocals, however, if you do it long enough, you will realize its not necessary. The problem with recording your tracks with reverb from the mixer is that you can't "undo" that reverb - you are stuck with how it sounds when tracked.
The other thing to consider is that the Stagepass mixer does not have a stereo input section for mic/guitar (no pan control on the channels), so if you have both vocals and guitar plugged into it, your output to your interface will be mono - guitar + vocals on same mono track in your DAW.

Yes, problematic Mike (see below) the Yammy is not even as useful as a £50 Behringer in this situation. I wonder if that MOTU has any inserts?

Dave.

 

[ecc83]

Just found this: https://www.soundonsound.com/news/motu-896-mk3

Oh! FFS! That interface can do EVERTHING! Tell you what Gene, you ship it to me and I will buy you a Behringer UMC404HD and a wee Behrry mixer with FX and you can play with that. Give me 6 months and I will have the MOTU sussed and we can swap back. Might be a bit over 6mnths because I would like my son in France to have it for a time!
Seriously, you have some serious reading and concentrating to do.

 

[rob aylestone]

I can't comment on garageband, because it's just not on many people's list of software to use, but of course, it's free ......

I'm a cubase user - so if I want to play a guitar and sing, this is what I'd do. I'd plug a vocal mic into channel 1, and a guitar mic into channel 2. I might, if the guitar had a pickup stick that into channel 3. In Cubase - I'd select the routing to send 1 input 1 to track 1, 2 to 2 etc. I'd set the levels on the interface (Rather a nice one you have, I note) Let's say I want some reverb on my voice, and a different reverb for the guitar. In Cubase, and most DAWs are similar - I'd add a reverb to the track 1, and another to track 2. ABSOLUTELY NOT to the input channel, because that would embed the reverb into the recorded track and I want to be able to tweak it later. For speed and easiness - I'd set the headphones to simply listen to the main stereo out. Then I can adjust the faders to give me a mix I can sing and play to, with as much or little of the reverb. Arm 3 tracks for record, bash rec and off you go. The few mS latency I have, and with your interface, almost certainly what you have, can be ignored. It need be no more complex than that. I don't know about garageband. That may or may not be able to do this, but Reaper, Cubase and the other popular DAWs can all do this. You have some nice mics and nice interface - Garageband is probably the limiting factor.

PS - send me the interface, and I'll match daves mixer and add a Shure SM57.

EDIT
On the reverb front, personally the reverb is a performer choice. Some cannot sing without reverb, some can.

 

[LazerBeakShiek]

I understand that some people really *need* reverb when recording vocals, however, if you do it long enough, you will realize its not necessary. The problem with recording your tracks with reverb from the mixer is that you can't "undo" that reverb - you are stuck with how it sounds when tracked.

Yeah, but thats kind of par . When performing the piece, you would push the reverb with your vocal force to create excitement. You will sing differently because of it. Adding reverb after does not allow the performer to make the most of the effect.

 

[gene12586]

I understand that some people really *need* reverb when recording vocals, however, if you do it long enough, you will realize its not necessary. The problem with recording your tracks with reverb from the mixer is that you can't "undo" that reverb - you are stuck with how it sounds when tracked.
The other thing to consider is that the Stagepass mixer does not have a stereo input section for mic/guitar (no pan control on the channels), so if you have both vocals and guitar plugged into it, your output to your interface will be mono - guitar + vocals on same mono track in your DAW.

Thanks Mike. I get what you’re saying about the reverb. But if I like the reverb sound of the mixer better than what I get from the software, does it really matter so much that I’m stuck with what I get out of the mixer? Also, if I don’t have reverb, then I hear no reverb live, right? I feel like that will take away from the vibe, which seems to me possibly like the most important factor in getting a good take (at least for me). Or is it possible to get live reverb somehow by using the DAW?

 

[gene12586]

Er? Yes! The MOTU should be able to send a "Zero Latency" output to a line input on the mixer*. You can then add reverb to THAT input and hear it from the mixers headphone jack.
OR!

You could run EVERYTHING through the mixer and feed that to the interface for recording. That would need the mixer to be able to give you a 'clean feed' of the reverb so you did not record it.

BUT!

I don't have the gear. YOU do! FCS plug it all up and have a bloody DO!

Dave.

Thanks Dave. Running everything through the mixer was the option I initially presented…. What do you mean by needing the mixer to give a “clean feed of the reverb so you did not record it”?? Totally newb here with the terminology, sry.

 

[rob aylestone]

I do lots of classical music. Singers who sing for hours each day, concentrating on tiny details. They have spent their lives in concert halls and churches with significant reverberation. It’s as vital for them as monitors are for stage singers. Put them in a studio with dead acoustics and a close mic and their voice has gone. They sing louder, their voice gets worse. Give them reverb and their comfort zone returns. A busker who has spent years singing outside doesn’t need or want reverb. To them, it’s unnatural. To a cathedral chorister no reverb is bizarre, and of course it’s why everyone sings better in the bath!

 

[gene12586]

Just found this: https://www.soundonsound.com/news/motu-896-mk3

Oh! FFS! That interface can do EVERTHING! Tell you what Gene, you ship it to me and I will buy you a Behringer UMC404HD and a wee Behrry mixer with FX and you can play with that. Give me 6 months and I will have the MOTU sussed and we can swap back. Might be a bit over 6mnths because I would like my son in France to have it for a time!
Seriously, you have some serious reading and concentrating to do.

Amazing, Dave, thanks for pointing this out! I didn’t realize this was possible. So basically as I understand it, this interface has an on board mixer and I can use it to apply reverb to the vocals input and send that to headphones to monitor? Then I take it I would just turn fader all the way down for the track that’s getting recorded in the DAW. And it seems I can apply the reverb in the interface not only to inputs that are sent to headphones, but also to the outputs sent to DAW if I so chose, which is cool. Only thing is that this is technically still digital reverb, right? (whereas I tend to like the analog sound). What’s the difference between adding the stagepas mixer in between the vocals mic and interface, adding reverb in the stagepas mixer, sending that to the interface, and then sending to headphones (i.e., direct monitoring)? I’m not sure I understand the difference other than whether I’m adding the reverb on the stagepas or using the onboard mixer on the interface…

 

[gene12586]

I can't comment on garageband, because it's just not on many people's list of software to use, but of course, it's free ......

I'm a cubase user - so if I want to play a guitar and sing, this is what I'd do. I'd plug a vocal mic into channel 1, and a guitar mic into channel 2. I might, if the guitar had a pickup stick that into channel 3. In Cubase - I'd select the routing to send 1 input 1 to track 1, 2 to 2 etc. I'd set the levels on the interface (Rather a nice one you have, I note) Let's say I want some reverb on my voice, and a different reverb for the guitar. In Cubase, and most DAWs are similar - I'd add a reverb to the track 1, and another to track 2. ABSOLUTELY NOT to the input channel, because that would embed the reverb into the recorded track and I want to be able to tweak it later. For speed and easiness - I'd set the headphones to simply listen to the main stereo out. Then I can adjust the faders to give me a mix I can sing and play to, with as much or little of the reverb. Arm 3 tracks for record, bash rec and off you go. The few mS latency I have, and with your interface, almost certainly what you have, can be ignored. It need be no more complex than that. I don't know about garageband. That may or may not be able to do this, but Reaper, Cubase and the other popular DAWs can all do this. You have some nice mics and nice interface - Garageband is probably the limiting factor.

PS - send me the interface, and I'll match daves mixer and add a Shure SM57.

EDIT
On the reverb front, personally the reverb is a performer choice. Some cannot sing without reverb, some can.

Hi Rob, thanks for all of this. So is what you’re saying that I can add reverb live while singing/playing using the DAW, and that any latency produced by this will be negligible?

Also, I’m curious why you might both mic the acoustic-electric guitar and also send it via cable to two separate interface channels?

And about your edit… I definitely need reverb when I sing, particularly for the style of stuff I do – lots of ambience and deep spacey stuff (think vocals for Go Slowly by Radiohead) – if I were to do everything dry and add reverb later, I feel like it would detract too much of the feel in real time. But again, I’m going to experiment, as you’ve all suggested, and see what works best.

 

[gene12586]

Yeah, but thats kind of par . When performing the piece, you would push the reverb with your vocal force to create excitement. You will sing differently because of it. Adding reverb after does not allow the performer to make the most of the effect.

Thanks LazerBeak. This is exactly what my logic is, as I just wrote above to Rob.

 

[rob aylestone]

If you have a jack socket, there’s an output available. They never sound ‘real’ but they do sound totally isolated, with none of the voice. As some of the guitar will contaminate the vocal recording it can’t hurt, costs nothing and gives you two guitar faders, to blend. It might help your choices. You don’t have to use it but you could.

Let’s sort the latency thing. Every time you press record and listen to another track there will be latency. The in to out process always takes time. My latency by today’s standards isnt tiny, it’s 9ms. As I use VSTis all the time. Every note Cubase plays is 9ms late. I can start to feel latency at 11ms. Not as a ‘delay’ but as a sort of feeling. Sometimes if I’m a bit silly I can set up effects that double the latency. I can’t play to these without cheating. Cubase let’s me fix it by sending data from a recorded track early, which with the new delay, then arrives on time. I also am not certain if the 9ms displayed on the screen is real and accurate. Is my 9 somebody else’s computer system‘s 11?

what matters is you. Everyone has their point where the delay messes you up. On a modern computer with modern device drivers, latency hadn’t been an issue for me now since probably 2010, when I couldn’t quite afford a fast enough PC.

get your system running and try it. Don’t plan and stress. I’d bet you don’t have an issue, but if you do, there are things you can then, but only then, try. You’ll have far more things to spend time on. In fact, I don’t think on my current Cubase, version 11, I have even bothered to even see what my latency is? No need, it works fine and that is the only thing. The control on my interface to mix real direct sound with computer sound is set to 100% computer. Everything plugged in I hear via the machine.

 

[gene12586]

If you have a jack socket, there’s an output available. They never sound ‘real’ but they do sound totally isolated, with none of the voice. As some of the guitar will contaminate the vocal recording it can’t hurt, costs nothing and gives you two guitar faders, to blend. It might help your choices. You don’t have to use it but you could.

Let’s sort the latency thing. Every time you press record and listen to another track there will be latency. The in to out process always takes time. My latency by today’s standards isnt tiny, it’s 9ms. As I use VSTis all the time. Every note Cubase plays is 9ms late. I can start to feel latency at 11ms. Not as a ‘delay’ but as a sort of feeling. Sometimes if I’m a bit silly I can set up effects that double the latency. I can’t play to these without cheating. Cubase let’s me fix it by sending data from a recorded track early, which with the new delay, then arrives on time. I also am not certain if the 9ms displayed on the screen is real and accurate. Is my 9 somebody else’s computer system‘s 11?

what matters is you. Everyone has their point where the delay messes you up. On a modern computer with modern device drivers, latency hadn’t been an issue for me now since probably 2010, when I couldn’t quite afford a fast enough PC.

get your system running and try it. Don’t plan and stress. I’d bet you don’t have an issue, but if you do, there are things you can then, but only then, try. You’ll have far more things to spend time on. In fact, I don’t think on my current Cubase, version 11, I have even bothered to even see what my latency is? No need, it works fine and that is the only thing. The control on my interface to mix real direct sound with computer sound is set to 100% computer. Everything plugged in I hear via the machine.

Thanks much for the thorough explanation Rob. That all makes sense.
Cool feature you mentioned on Cubase. Will have to see if I can figure out how to do something like that with my software.

 

[TimOD]

If this recording thing is something that's important to you, and you want to do it forever, this is what I think you should do, because I do it, and it works great and is simple to set up (hopefully I won't confuse myself and you): Treat the room. Get an out-board reverb unit and set it up in a rack. (I have a Phonic in the rack; a good unit, but impossible to find). You'll need to do this eventually anyway. Get a dedicated monitoring device, like the PreSonus Central Station. Put it in the rack. Get a patchbay(s) and put that in the rack. It appears your mixer does not have AUX sends and returns. It also is designed for stage work. Use it for that only. Get a mixer with AUX sends and returns, like the Soundcraft GB2R (pictured). You have the computer and the interface. It looks like that MOTU has plenty of outs and ins. Run the reverb unit's outs and ins through the bay. From the bay they go to the mixer's AUX inputs and Stereo AUX returns. This is straightforward; hooking up the bay to your various bits of gear adds massive flexibility to your rig. I see that the MOTU has a virtual mixer; I believe you can do further routing with that, but get the MOTU's main outs on the bay and/or Central Station. Determine the amount of cables you need, and the length, and hook it all up. Start out simple, and add cables and connections that you think you need as you go. The idea here is to monitor from either the Central Station or the mixer the combined signals--what's going in, and what's coming out. You can do it through the MOTU too, of course, but I'm not sure how flexible that is. NO latency, because you're using the outboard reverb to add treatment via the mixer that your monitoring the signal from, the stereo track and your new vocals. The dry signal ONLY of your voice (or whatever--I use it always on steel guitar, for instance) gets recorded, or you can record the reverb, together or separately. Lastly, get a Behringer ADA 8000. For the outs more than the ins. Put it in the rack . . . .
So, your voice is routed to the mixer, where you use the AUX sends to give the reverb unit your voice, and it returns it to the mixer via the AUX returns. You adjust the amount to taste. The computer is recording your dry, un-effected voice, simultaneously. As long as the MOTU has the functionality of routing inputs in real time to its outputs, you're good to go. I use an RME RayDat, but I think the MOTU will work.

So: I know it seems kind of dickish for me to declare "hook it up," but that is the way to learn how to do it. (Private message me and I'll give you a much more detailed look-in on how my system is set up. We can private message here, right?) It's actually pretty easy. You need to learn it. Why do it this way? There is never a latency issue. It's all in real time. You can record the reverb from the out board, or not. Either "attached" to your voice, or not, or as a separate track. It's in stereo, or mono. You can record through the mixer: the GB2R in question is a very good mixer, with great pre-amps, but I only use my Audient ASP 880, and sometimes a modded ADA 8000.
All of the suggestions above by the other posters will work for you; this is how I do it, and it is simple and reliable. You'll get used to it quickly, I can promise you that.