Track Outputs vs. Sends survey: How do you guys prefer to route in Sonar?

So for quite a while I have been setting my track outputs to None and then using sends to route out to various instrument and FX busses. Now I'm reconsidering this strategy and might use the track outputs to go to main instrument busses and only using sends to output to FX busses and the like. So I was wondering how everyone approaches this and what their reasoning is. Just looking for the pros/cons of both methods. Cheers.

Beepster


So for quite a while I have been setting my track outputs to None and then using sends to route out to various instrument and FX busses. Now I'm reconsidering this strategy and might use the track outputs to go to main instrument busses and only using sends to output to FX busses and the like. So I was wondering how everyone approaches this and what their reasoning is. Just looking for the pros/cons of both methods. Cheers.

If I want them to be completely effected by a buss (The whole signal, not just part of it) I assign output to busses (Drum Buss, Vocal Buss, Guitar Buss) and if I only want some of the signal to be effected, I use sends (Reverb, delay, etc)

Usually use track outputs to master. Drums track outputs generally go to Drum bus. Sometimes like instruments will be routed to a bus. Sends to FX busses. Everything ultimately resolves to master bus

+1

Bob Bone

Hi, Bob. Is that a vote for Output to Inst. busses? It will get rid of that missing output message I get (I know I can disable that message but I like having it unless something really does go wrong). Cheers.

Well the busses would definitely be going to the Master (related tracks > instrument bus > Master) but I'm not seeing the benefit of having the tracks outputting directly to the Master in addition to that (tracks > instrument bus AND master > instrument bus to master). Maybe I'm misunderstanding. Cheers.

Beepster


So for quite a while I have been setting my track outputs to None and then using sends to route out to various instrument and FX busses. Now I'm reconsidering this strategy and might use the track outputs to go to main instrument busses and only using sends to output to FX busses and the like. So I was wondering how everyone approaches this and what their reasoning is. Just looking for the pros/cons of both methods. Cheers.

That makes more sense Beepster to use the faders rather then the send knobs to mix. Generally sends are used to output to shared effects such as reverb and only returning the wet signal of the reverb. 


That being said I don't think there is a right or wrong way to do it. What ever works. Sonars mixer design is loosely based on an analog console. Loosely being the key term.


Side rant:
One thing I never understood about sonars routing is why we can't assign a track to multiple outputs like you can do on an analog console? I know, use sends! But since sonars console design is based on an analog console, including its unneccesary limitations, doesn't make sense you can't do this.  And since I'm off topic a little why no mention mute. We are on how many versions later since automated mute has been added and the code hasn't been patched to make one mute that is has automation like any other console parameter and records the automation from a control surface?

It is common practice to have busses for groups of instruments, such as "Guitars" or "Vocals" so that you can alter the level of them all together relative to the other "groups.  You can also add gentle bus compression etc. to help them "gel".  For these, you would assign the tracks outputs to the bus.  Using sends for this would be pointless, and not work.
 
For effects busses e.g. reverb, you would use sends.
 
HTH

I know a lot of members want more routing options but for me I find the way Sonar routs is as complicated as I can deal with and meets my needs completely. I know they may be very meager as compared to others. 

One more thought, I have found due to the PC I don't use as many aux buses as I once did. ProChannel has caused me to rethink how I go about setting a project up. Having a channel strip on every track and buss has me rethinking the whole idea behind a project. 

Although I usually do have instrument submix and aux send busses like most everyone else, John's comment here sums it up best for me:   

I create buses depending on the song itself.

 
I've had occasional songs with no submix busses at all, where all tracks go directly to the master bus.  In those cases I simply had no need for the additional busses, so why bother.    It's not the norm for me, but it has happened...   But most of the time I prefer the ability grab one fader or envelope and adjust a series of related tracks together, same for FX that help glue things together.  
 
 

Edit: Duplicate post courtesy of IE9.
 

My main issue is because I'm dealing with multiple clones of guitar tracks (so they can be panned left/right) those get grouped. Then there is multiple vocals which need to be grouped and have their own FX busses. Because the drums were recorded with a four mic set up (kick, snare, two overheads) they need some particular attention. I'm going to be using samples to beef up the kick and snare so each of those will be at least two track each so they are both getting their own bus. The overheads get a bus and because I'll be creating a clone of both OHs to create a tom focused set of tracks those will get a bus as well. All drums get routed out to the main drum bus and maybe a dedicated delay/reverb and compression bus. If I decide to go with "New York" compression on the kick/snare that might be an extra two busses as well. Then I have a pre master bus to do overall tweaks on EQ/compression/etc so I can keep the Master as uncluttered as possible. There is more stuff I'm considering as well but I'll see how things progress as I move forward. So yeah... kind of complicated bussing but if I had to mix with just the tracks it would be a nightmare. That's how I was doing things before and trying to keep the clones/doubles/etc in sync with each other was an ultra pain and I always ended up missing something somewhere. After seeing a few of those Groove3 vids and reading up a bit more I realized that I shouldn't be stingy with my busses if there is a legit reason for them. Just glad my brain is able to visualize this type of complex routing now because before I was like a monkey trying to fly an F-15. lol

Tracks go to master, unless there  is a reason for me to route them to a bus.  Drums, bus.  Guitars - yea, I like the ssl bus comp on a collection of guitars.  Vocals - usually for the same ssl reason along w/ reverb (delay usually goes on the track).  It is a per track agenda, or combined as said above.  It is what the instruments call for.

There is no right or wrong, just a style you develop.  If you want the same reverb wash on stuff, use a send.  If you want most of the guitars to make a wash, bus comp.  It is technical, not strategic.

@

Learning about how to set up reverb and delay busses to react to the project tempo and overall mix instead of applying it directly to tracks was a very interesting and eye opening part of my studies. It really gave me a huge insight not only about reverb and delay but bus routing in general. I'm very much looking forward to playing around with these concepts. As always guys you've cast aside some doubt for me here. I am moving forward with my current routing set up (track output to appropriate bus and sends on tracks as needed) and creating a rough mix so I can edit and then overdub with everything audible without blowing out my ears. Thanks again but by all means keep sharing. Always cool to see how others do stuff. Cheers.

I always route logically grouped tracks to a bus, and then route that bus to a master. That way I can apply some very light compression, eq etc across that bus (if required) prior to sending to a master bus. Makes the life of the master bus compressor a little easier as the rouge peaks should have been addressed. Parallel compression, fx and the like I route to send buses, and mix to taste.