Do you play your midi rig like one giant instrument.....

[thisDude]

...or do you compose your midi and render the parts dry to audio and mix with that?

 

I have been using my rig like one giant instrument, running automation and effects on the live output, but I notice a lot of interaction going on between different parts... little things like the apparent flanging of hi-hats, when tweaking a low pass filter on the synth. Maybe it's a different issue all together, but for some reason, elements in my songs don't sound as "separated" as commercial stuff.

[Stephen Fortner]

In my experience, what you describe amounts to the best argument for rendering each MIDI track as audio, then approaching the whole thing as a mix engineer would. Fortunately, given today's fast computers and very capable DAW apps, we have that option.

 

I especially used to notice it when doing a project entirely with one keyboard workstation. There are a lot of tweaks one can learn, but at the end of the day, nothing increases the size of this sonic box you're working within.

 

Heck, really golden-ear engineers point out similar phenomena if tracking everything through the same front end, even if that's some really nice boutique pre and compressor. They'll intersperse "lesser" equipment on certain tracks to avoid everything having the same sonic footprint. I'm making no value judgment as to how legit that particular claim is, just passing it along.

[Jeebus]

Interesting question. There's obviously no right or wrong way, and both have their merits and disadvantages. I've never settled on recording in one shot running my rig as one giant midi instrument, but I have experimented with that method. I prefer the latter option, and I think it will help you get the "seperated" sound you're looking for. Can you give some more details on your method? Are you recording a bunch of midi gear, passing it through a mixer and recording the stereo output? Or are you recording multiple simultaneous tracks to a DAW? Another thing that can help with that seperation you want is the way you are mixing the audio.

 

"I especially used to notice it when doing a project entirely with one keyboard workstation. There are a lot of tweaks one can learn, but at the end of the day, nothing increases the size of this sonic box you're working within."

 

Agree 100%.

[T.H.]

I record everyting MIDI all the BELLS & WHISTLES

(cause it's easier to change if need be)

 

Then once it's exactly how I want it, transfer it

to AUDIO. ((wiping hands)) less hassle dealing with the audio, because exactly the way I want it.

 

NOTE: Never erase the MIDI after transferring it

I keep the MIDI & AUDIO all in one WINDOW, just in case you MUST make a change, Play the MIDI back, make changes, then transfer to AUDIO again.

[DJDM]

That's a great question td. I like to think of my software system as one giant recording studio. I am able to tweak and make changes to the various parts here and there to make them stand out.

But I should note that I do tend to favor a more homogenous sound on the whole. Since I do electronica and it exists only virtually, I like the idea of a consistency of space. I think it adds uniformity to my sound which works for me. Kind of like the Motown or Sun studios approach. Those facilities created an unvarying sound that was both appealing and very unique.

But there is also a very valid argument for not doing everything in the same space.

In the end it is up to you and your ears. If you can get good results out of the gear in front of you that people respond to then there is little more you need to do. But if you feel your sounds are too contiguous and lack form then break them out and try to spread your sounds out among other separate utilities. Say for instance the separate tracks into a DAW approach.

There is no surefire way to make a track sound spectacular. Mostly great tracks seem to come from studios that take chances, try new things but have a vision for their sound in mind. It is very easy to get caught in a rut that allows you to create tracks quickly and with a modicum of technical requirements during tracking. What you need to figure out is what it will take to make them sound distinctive. One way to do this is to bring in a second or third party to listen to your work and give you feedback. That can have a profound impact on your results.

Just some thoughts.

[Gus Lozada]

Maybe I am not following you...

 

Are you using a single synth/workstation and are you asking if it might be better to record all the tracks as audio and mix them instead of doing all the mixdown inside of the keyboard?

 

Or... do you have multiple synths and what would you like to know is if instead of mixing them using MIDI automation and just recording the stereo master you should bounce everything to audio and mix in a DAW?

[Alndln]

I leave my software drum plugin(DR-008) live because it's got 10 audio outs and I can hide them from my track screen and acsess them from the board view.At this point all of my rack synths have either died or need the soldered in batteries replaced, so apart from my korg everything else is softsynths.I render those early for 2 reasons,one : CPU obviously,Two : I have so many patches and samples that if I dont decide early I'll spend forever going through alternate choices.I'd rather spend other time for getting familiar,besides I can alway's keep the midi tracks and hide/archive them if I change my mind at any point,or to correct mistakes,but that's rare because I usually know right away.These days I use midi as a means to an end and not dwell on it anymore.

[zele]

>> we use the the Midi rig as one big giant--sequence everything in the Kurzweils (32 tracks)..rarely ever record the audio outputs to a HD >>> ..we use the 2480 Rolands' 24 inputs as the front mixer to the 16 outputs of the dual Kurzweils, leaving 8 mixing channels of Roland drum machines or dig I/O of our effects--(using the automix and effect/Eq/etc.of the VS) and then mis the recorded 24 tracks of audio on the VS HD and the 24 channel "Midi mix" to a Masterlink.

[Mr Darling]

Maybe my situation doesn't apply here . We have few midi outboard, but also a digital desk.

I leave most midi parts as midi, mixing them at the desk along all the audio files coming from logic (that ends on 8 seperate tracks on the desk)

I render midi parts to audio if I have no free faders on the desk (I do not like two sounds on one fader) or if I feel I can tweak the sound better using plugs.

 

I never felt any timing issues leaving the midi live, maybe becuase I have a good midi interfirance (AMT8) , use a seperate midi cable to each instrument and keep those short.

 

I'm still not comfertable with letting the computer handle all the pressure - this is the main reason I'm still ubsing an outbaord sampler.

[coyote]

I guess I play my midi rig like one giant instrument, but I do it in a different context. I'm not recording via midi; I play live. (When I DO record, I play each part as audio into an 8-track tape recorder. Old-fashioned, I know ) But I have a single device control all my patches for my multiple keyboards & modules, and every keyboard at one time or other during a set is controlling some other board or module. And when I get the FCB1010 programmed, even non-midi switching of devices will be controlled via midi.

[thisDude]

I don't think that any one particular solution is going to help my specific situation, but I'm trying to get an idea of "the method" that others are employing. It could very well be that the tools I'm working with just aren't up to par.

 

The question of whether you are using your rig as one giant live instrument or bouncing and mixing with real audio was just my initial hunch as to what was causing the problem. The frequency "interaction" may be going on at the monitoring level. I'm going to try some different settings this weekend along with the "bounce and mix" method and see what happens. It could also be my el cheapo power supply on my cpu. Something tells me slight power fluctuations aren't good for the sound, especially when the DSP is ripping everything live.

 

The challenge will be in deciding which automation to include in my initial rendering and which to do during the mix. The filters on my nord have a lot more balls than any of the res filter plugins I have. I guess that's why mix engineers make the big bucks.

 

Can you give some more details on your method? Are you recording a bunch of midi gear, passing it through a mixer and recording the stereo output? Or are you recording multiple simultaneous tracks to a DAW? Another thing that can help with that seperation you want is the way you are mixing the audio.

I've got a Cubase DAW with Reason rewired into it (mostly for the drum machine and sampler) and various VST intruments. The Nord and occasionally an S90 on loan from a friend are run into an audiophile24/96 sound card and everything is monitored thru that.... and yes, I'm sure it has a lot to do with the mix. Looks like I'm going to have to figure out what all those seemingly useless plugins do.

 

Or... do you have multiple synths and what would you like to know is if instead of mixing them using MIDI automation and just recording the stereo master you should bounce everything to audio and mix in a DAW?

Yes. I know there is no right or wrong way (assuming that it sounds good)... I'm just trying to get a feel for which method others are using.

[Kendrix]

Some and some.

 

If one midi module is serving up more than one sound ( and has only one set of outputs) and I want to EQ or effect them differently I record one to audio.

 

Also, my sampler has digital outs.

I usually render some tracks from it to audio via a digi connection to my recorder. In this way i eliminate on DA-AD conversion cycle and I do notice improved sonics from this.

 

For some other tracks its one giant instrument.

 

However, i sometimes have mis-fired midi notes occur and timing is not exactly identical in every midi pass. I need to carefully monitor for this.