That's a good observation about the default volumes. And here's what I've found: When I have a full orchestra set up, in the passages where all the instruments are playing, the volume almost goes into dreaded digital over-load. I experimented--Any higher volumes on those settings, and distortion is sure to set in with large numbers of instruments playing in a project.
Thanks, Randy. A simple, practical, and useful explanation!
One thing I have experimented with is moving the Kontakt volume settings up so that when the instrument receives a CC1 message of 126 (maximum), its output will push up close to, say, -3 db on the meters in my sequencer (Sonar ... not the Kontakt meters). The benefit is that I get a "truer" representation of the volume when I draw CC1 messages. CC1(64) would be mf, CC1(32) would be pp, CC1(126) is ff, for example.
However, the effect you describe results. Bundle a bunch of tracks together playing f, and you pop the meter.
Could you use this method described above, but drop the overall volume with an envelope or through a bus when the orchestra is tutti?
Hello again, Kent--Glad my explanation made sense. I'm quite sure that what I described is what the programmers hand in mind as they figured out the default volume settings.
I use Sonar also, so we have the same point of reference.--And here's my
impression from what you described--that you're using a broader range of cc1
volume data than you need to. A value of 32 is extremely quiet, to the point of becoming virtually inaudible if the loudest passages have been playing back at a comfortable level for the listener.
I rarely go lower than a cc1 value of 64 in my projects. Having a good wide dynamic range is a common compliment on my work, so I don't think I'm compromising my dynamic possibilities with this approach.
I'm not quite sure what you meant when you said:
"...I get a 'truer' representation of the volume when I draw CC1 messages..."
?--Truer than what?---There's no doubt in my mind that the most satisfactory cc1 usage is done by recording it, making a volume pass over a previously recorded track. I often am using the wheel as I record, but for more difficult sections, I do the volume separately. Getting into the Piano View to do some fine-tuning of the cc1 data by hand is always necessary, but it's extremely rare for me to do all the cc1 that way--My own ears easily show me that the most natural, organic effect is done by recording.
I hope you don't mean you're inserting one cc1 value to control the volume of a whole section of music?--Like using cc7 with traditional synths? That's not the way cc1 was intended to be used. It's a constantly changing parameter which emulates the ebb and flow of acoustic instruments.
If you're concerned about various sections varying too wildly in volume--I understand. When a project is worked on over the course of many days, sections worked on during different days can end up having very different points of reference--I mean, what seemed "loud" on one day will end up playing back too softly when set along side another day's work.
All that is very correctable with the Sonar Velocity application--one of the MIDI "effects." It's a very detailed window where you can select the data for a section and change it by a percentage. Looking at a zoomed out view in the Piano Roll window, and you can get various sections of a piece to match up better. And the advantage of shifting volumes up and down by percentages is that the good, detailed cc1 data isn't compromised--all the detail is still there, just lower or higher in overall value.
"...Could you...drop the overall volume with an envelope or through a bus when the orchestra is tutti?..."
Yes, but the effect could be unsatisfactory, with the piece's volume decreasing right at the moment when it should be Increasing.
Using busses is really pretty much essential for balancing out the orchestral sound, from section to section. But it's a constant juggling act--you can't just suddenly drop the volume when there's an especially loud and full section.
One thing to keep in mind is the necessity of starting things out low when you're mixing. Console sliders at about half way up is a good rule of thumb. Plenty of head room is available that way. Then when you Do come to your loud sections, they're going to provide plenty of contrast to what came before, but without peaking out your meters.
If you find that the entire mix ends up too soft--Fine, simply adjust the trim pots so you don't have to re-do any envelopes or automation you may have recorded.
When I've done all the fine tweaking I can muster, first in the MIDI realm, then in the audio realm, I still often find that I need at least some subtle compression and/or limiting to tame the recording. So don't forget those tools.
I'm working on a new piece today, and I'm following your suggestions of maintaining the default GPO volume settings but using higher cc1 values than I have in the past. So far, so good. (Yes, I do keep cc1 messages moving.)
However, it does seem a bit odd to me that a cc1 value of, say, 16 is virtually inaudible. It raises the question, "Then what is 16 for?"
Which brings me to another point. Based on your approach, when you do the final mixdown for burning, say, an MP3, what is your reference point for the final volume settings?
For example, how do you set final levels for a very quiet funeral song vs. levels for a song with a robust, all-instrument finale? I've been referencing to the meters (using my ears only when I'm not aurally fatigued), but I have a sneaking suspicion that you have a few tricks up your sleeve!