How do you make a MIDI note louder during it's duration?

Example if you have a long legato violin note and you want the "velocity" of it to increase overtime. There are a couple of ways I've found to do this, but I'm not satisfied with them. I could make an envelope for the volume of the track itself, or I could create a second track with the same instrument except with a slower attack. Maybe an automation envelope on the attack would work, but not in dimension pro, where the envelope generators are plotted. What methods do you guys use?

A CC7 (or on some synths, CC11) automation envelope.

Actually, it's not volume at all. It's after-touch as suggested by Josh. after-touch can affect many things, depending on how the synth/patch is set up. Most of the time, it is setup to affect volume, but could just as well be cutoff freq or res. After-touch is what is sent when you adjust how hard or soft you are pushing a held key, after the initial press. If your controller does not have this ability (and some don't), or are entering notes by hand on the p-roll, you'll need to select the note and time-range using "Select -> By Filter" and then apply a control change message set. The reason you would want to do this over using volume is that volume will affect all notes being played in that patch. Applying MIDI CC messages to a single note means you can actually sustain a root and third with a strict volume, while having the upper note do some interesting things.

I hope this helps.

Randy B. (rbrowser) nailed it.

Something else to consider is whether or not your string sample responds with a volume swell when you move the mod wheel.   The thing to remember is that the timbre of real strings change with volume changes, so simply increasing the volume won't necessarily yield a satisfying string sound.   The better string libraries accommodate this with multisamples tied to some sort of continuous controller changes.   Maybe check the documentation for the string samples to see what's inside.

JoshWolfer


So if you use CC11, it effects notes differently then CC7? 

How does the plugin know which notes should be affected? There is still something that I'm missing here.

Hi, Josh - This seems to be one of my missions in life, to expound the virtues of using CC11 instead of CC7.

But Paul mentioned using the mod wheel, which sends out CC1, not 11.

Taking his point first - my main sound libraries are from Garritan.  All of the GPO winds, strings, brass, and those in JABB, CoMB, World - they use CC1 for volume control.  As Paul explained, using strings as an example, there's a change in the timbre of the instruments as CC1 swells and falls, and it's critical you use CC1 for volume control in those libraries.

AND CC11 is the alternate control for those GPO instruments.  CC1 is always mentioned first since most people have a mod wheel which can send out the appropriate data.  But CC11 accomplishes the same thing.

Moving away from Garritan, for an extremely long time now, almost since the beginning of MIDI, CC11 has actually been the INSTrument volume control.  CC7 controls volume in a different way.  But what's happened is that many people (the majority?) have come to think of CC7 as The volume control, when it actually isn't.

Many instruments in samplers, and some synth patches, are programmed so that when they're played musically, with a dynamic use of CC11 as the volume control, - their timbres will change, more dull and muted at lower levels, and brighter at higher levels.  CC7 has no such control available to it.

CC7 is intended for setting the Potential volume of an instrument.   Like if you have a multi-timbral synth/sampler, you set up a relationship of balances between all its instruments, and CC7 is the controller doing that.  You leave that setting the same throughout a project.  Then it's CC11 you use to Play the instruments - and well done MIDI projects will have a ton, almost continuous CC11 data in them.  No note can be just played and sustained ala old school synths - Once the note is hit, then it swoops up and down, as per what real-life instruments do when played by good musicians.

Get it?

Randy B.

JoshWolfer


Randy, 

Thanks for taking the time to explain that. I'm going to have to play with my hardware a little more =).

So that's how they get those creamy dynamic shifts using things like synthobia. The horns especially need to have dynamics programmed in or they sound lame. I figured that'd be built in to the velocity of the notes, which it is, but that doesn't give you the ability to shift after the note has started. 

Thanks! I'm really looking forward to playing with this some more. 

Well cool, Josh - Thanks for letting me know my post gave you some ideas.

I'll tell you, there isn't a good sampled instrument in creation that won't sound lame to some degree without some work with dynamics/volume.  Some samplers will give you instruments that have built in swoops and dynamics, but that isn't so good, because you're locked in to that same thing happening every time you hit a note.  

I know that the number one reason people will say "This such-and-such library sucks" because they don't understand how to make it come to life.  They'll hit a note, hear the sound - and if it doesn't sound sexy after the initial moment, they think it sucks - when actually it's Them that sucks.  8-)

The best samples have great attacks, and after that initial attack, they're static--and purposely so.  The musician is then supposed to play with the volume in a way appropriate for the passage, just as the real-life counterpart musicians do.  Take a horn, as in your example - let it hit, and then swoop the volume suddenly down and then back up to its full volume - then you have a dynamic horn line.

And think of it as playing the instrument, not so much "programming" it.  If you can record that performance data in real time from your keyboard, you'll get the feel for it, and your MIDI tracks will really start coming to life.

Thanks for listening.

Randy B.

They'll hit a note, hear the sound - and if it doesn't sound sexy after the initial moment, they think it sucks - when actually it's Them that sucks.  8-)

Guilty!

rbowser




...This seems to be one of my missions in life, to expound the virtues of using CC11 instead of CC7.

Thanks for taking the time to explain it again ... I don't think I really appreciated this before reading your post.  I feel some experimentation coming on.

Aftertouch uses a lot of resources in terms of the midi data stream and should be used only for live performance.  For recording and studio work it's best to use control 11.  You can also use control 7, but that is often reserved for the overall track volume, while individual notes are treated with control 11.  Obviously, you need a series of control 11s to create a crescendo or decrescendo.  Finally, another way to do it is to have the decrescendo and/or crescendo built into the sample itself.

Jerry Gerber
www.jerrygerber.com

rbowser


LHousehold, what you're talking about is Playing an instrument rather than just triggering notes.  CC11 (Expression) can be used in a virtually continuous way so that the volume fluctuates the way a violinist plays his instrument.  Listen to any good string recording, and if you listen carefully, it's breath taking how much the volume will fluctuate during a performance - from a whisper to triple F forte.  If you're using a sampler that doesn't respond to CC11 for volume, but that's not likely, then CC7 is second best - that controller actually being meant to set the over-all potential volume of an instrument.

If you don't have an Expression pedal, hopefully you have a keyboard that can have its wheel or other controller assigned to CC11.  If you don't have that, then start drawing expressive volume curves in the PRV.

You're not talking about AfterTouch.  That's usually assigned to control an instrument's vibrato, but it can be used to control other parameters.

Play your instruments.  They are instruments, not just "sounds." Don't just trigger them, breathe life into them with whatever tools you have to generate volume data.

Randy B.

+1 rbowser.
 
I'm about to get a nice MIDI breath controller to control MIDI expression. Should be fun!

jsg


Aftertouch uses a lot of resources in terms of the midi data stream and should be used only for live performance.  For recording and studio work it's best to use control 11.  You can also use control 7, but that is often reserved for the overall track volume, while individual notes are treated with control 11.  Obviously, you need a series of control 11s to create a crescendo or decrescendo.  Finally, another way to do it is to have the decrescendo and/or crescendo built into the sample itself.

Jerry Gerber
www.jerrygerber.com

Hi, Jerry - We've covered the OP's original question well now, but there are interesting sub-topics.  Aftertouch is one of them.

"...Should be used only for live performance..."---!  I've never heard anyone say that before.  And I know I certainly can't agree in the least.  Years ago when we were using hardware sequencers with small memories, and still early in the computer era, Aftertouch needed to be used sparingly because it used up a lot of storage space and could jam up MIDI traffic.  But we've had systems for some now that can easily handle Aftertouch.  It would be a rare situation where a musician is still using a system that would have trouble using AT.

I use the Garritan instruments, and many of them are programmed to have their vibrato controlled by AT.  Rather than setting up a translation of AT to a MIDI controller number, I use it freely as needed.  I record it from my keyboard, I draw it in sometimes in the PRV.  No problems - and there wouldn't be for anyone, unless they're still using a pretty ancient computer.

Randy B.

My understanding of CC7 versus CC11 had always been that the former is locked to volume by definition while the latter is undefined and left to the synth manufacturer to decide how to implement it. I used to have a Roland synth that allowed you to use the expression pedal in place of the mod wheel, for instance.

However, this discussion prompted me to do some research (or at least the lazy man's research: google). That led me to a better understanding of what CC11 is about, and how it differs from CC7. Mainly, that CC11 is defined as a percentage of volume, e.g. 127 = full volume. This source says it well:

By having both a master Volume and sub-Volume (ie, Expression), it
makes it possible to do crescendos and decrescendos without having to do algebraic calculations to maintain the relative balance between instruments. When Expression is at 100%, then the volume represents the true setting of the Volume Controller. Lower values of Expression begin to subtract from the volume. When Expression is 0%, then volume is off. When Expression is 50%, then the volume is cut in half.

CC11 defaults to either 100 or 127 on most synths. Zero All Controllers results in returning to this default value, but Zero Controllers returns the main volume (CC7) to zero.

Just out of interest what you are talking about here will be very possible with a future version of VST and I am referring to VST3 of course. At this stage only Cubase 6 and Studio One has it. Halion is the only synth that can respond to it fully as it does now.

It will allow for full control over every note for velocty, volume, pitch, fliter, fx and in a full chord, total control over every note will be possible. The receiving device has to be fully VST3 compatible though. The current midi spec pails in comparison and the results have to be heard to be believed. A new total level of realsim and control.

A lot of people are sitting on the fence to see where VST3 might go. If it takes off then the obvious thing is for the virtual synth creators to bring out new versions of their products that are fully VST3 capable. And for the DAW writers it means writing it into their software. I am sure Cakewalkwill embrace it if it becomes the norm. And their own synths will benefit from being rewritten to respond to VST3 messages.