• Register
  • Help
Page 1 of 2 12 LastLast
Results 1 to 10 of 12

Topic: dynamics with gpo and jabb

Share/Bookmark
  1. #1

    dynamics with gpo and jabb

    I was wondering how people in the forum apply dynamics to there music. I take a non rendered notation and convert it to midi, and then bring it in to SONAR. I then use the piano roll for note accents and some crescendo and decrescendo effects. After that, I save each track to an audio format. I have tried a few times to record my audio tracks to one track, using the master volume control for dynamics, but always ends up sounding like I am am using the VOLUME CONTROL and it does not sound very authentic to me. So I end up saving all of the tracks to one audio track, where I use SONARS audio edits to get dynamics. For classically oriented music, I sometimes have trouble getting the big picture of the musics dynamics, and it comes out lackluster. Any personal approaches and thoughts would be appreciated.
    Thanks, Jay.

  2. #2

    Re: dynamics with gpo and jabb

    Quote Originally Posted by jaynkate01 View Post
    I was wondering how people in the forum apply dynamics to there music. I take a non rendered notation and convert it to midi, and then bring it in to SONAR. I then use the piano roll for note accents and some crescendo and decrescendo effects. After that, I save each track to an audio format. I have tried a few times to record my audio tracks to one track, using the master volume control for dynamics, but always ends up sounding like I am am using the VOLUME CONTROL and it does not sound very authentic to me. So I end up saving all of the tracks to one audio track, where I use SONARS audio edits to get dynamics. For classically oriented music, I sometimes have trouble getting the big picture of the musics dynamics, and it comes out lackluster. Any personal approaches and thoughts would be appreciated.
    Thanks, Jay.
    Hi, Jay - Since I do my work first in Sonar, using a MIDI keyboard, things are a little different for me. A lot of the dynamics are already there since I played the various lines, and recorded CC data in the tracks.

    But when you start in notation and export - I thought if you had dynamic markings in the score, that the corresponding MIDI data would be there in the exported MIDI file. It would be simpler and less complex than all the "human error" variation of a DAW MIDI project, but it would be at least a start. I guess you mean that you export from Finale without any dynamic markings?

    Lots of work is still needed, however, even when the tracks have been developed with a MIDI keyboard. More passes recording MIDI data, hand editing in the PRV - and finally, the work that can be done with audio.

    But yes, Definitely you would have a problem automating the master fader for your final mixes. The Master really should only rarely be automated, some people go so far as to say "never." When the Master moves very much, you get the effect you described - of a volume knob being turned up and down. It's only useful for things like a fade out of an entire song ala pop song production, or sometimes for suddenly shifts of the entire orchestra if you need absolutely everything to suddenly shift down to ppp. Most of the time, however, you need to keep your Master in one position.

    It's the individual tracks that you can safely automate to some degree, and also the group buses - e.g. you want all of the woodwinds (which are grouped to a woodwinds bus) to come up a bit for a passage. By automating much smaller units of the full mix, everything else is staying at the same volume and so the results can be natural. You still need to avoid fader moves that give the volume knob effect.

    When you say you record all your tracks to one track - you're talking about mixing down to a 2-track master I think. You obviously still need to do that (I think exporting the 2-track is preferable to bouncing inside the project) - but as per the above, with detailed volume work being done on single tracks and/or buses. A lot of volume automation can be pretty subtle. You've bounced an instrument to a track, but during the mixing part of the project, you'll find that one or two notes are getting a bit lost. A tiny temporary boost of those notes, using a volume envelope, will fix that.

    And so forth. Basically I'm saying that the problem you described is due to an over simplification of the mixing process. Leave your master alone, with rare exceptions.

    BUT you still need to do as much MIDI work as possible prior to bouncing to audio. Constant volume work in each track, tweaking of velocities etc. You can get at least a rough mix before you ever go to audio. Working with all the bounced tracks just gives you a lot more detailed control over the sound later on.

    Randy

  3. #3

    Re: dynamics with gpo and jabb

    Yo Randy! Thanks for the info on this. I went back to the piece and reworked the dynamics. It still has some spots where you can hear the seams, but it is not to obvious. I also applied some panning. If you get a chance, give it a new listen.
    Thanks, Jay

  4. #4

    Re: dynamics with gpo and jabb

    Quote Originally Posted by jaynkate01 View Post
    Yo Randy! Thanks for the info on this. I went back to the piece and reworked the dynamics. It still has some spots where you can hear the seams, but it is not to obvious. I also applied some panning. If you get a chance, give it a new listen.
    Thanks, Jay
    You must be talking about the "Ghost" piece in the LR - Listening to it again now. Yes, much better panning, with each instrument having its own space.

    And the dynamic work is also much better. You said you can "hear the seams" - Yes, those are audible because there's a distinct click at the spots where you've suddenly dropped the levels. You won't get clicks if you don't make your vertical automation data perfectly perpendicular. You have to give that sudden change a bit of a slope, otherwise the results are exactly like when reel to reel tape was splice - you never did a straight cut with the razor blade, but always sliced at an angle to avoid a pop.

    You can hear that pop at around 2:06, and the one around 3:01 doesn't pop really, but you can hear the reverb volume suddenly change. Hmmm - you're not recording the reverb with your bounces are you? You need to keep your reverb plug-in still operating in real time during your final mix.

    Randy

  5. #5

    Re: dynamics with gpo and jabb

    Randy! I listened to those points in time that you mentioned. It sounds like the volume decrease cut off the reverb tail. I am not sure how it is supposed to work, but that's not right. I had reverb on each instrument and then exported them to one audio file. I then brought that file into sonar to work with the dynamics. I will just have to keep hacking away until I get all of this right. Thanks, Jay

  6. #6

    Re: dynamics with gpo and jabb

    Quote Originally Posted by jaynkate01 View Post
    Randy! I listened to those points in time that you mentioned. It sounds like the volume decrease cut off the reverb tail. I am not sure how it is supposed to work, but that's not right. I had reverb on each instrument and then exported them to one audio file. I then brought that file into sonar to work with the dynamics. I will just have to keep hacking away until I get all of this right. Thanks, Jay
    Hello again, Jay

    The first place I mentioned has a click, as if you have automation that is suddenly going straight down. But there's more to go over here. Your new post sounds like you're still doing it the way you were before. Here are some points:

    --You had reverb on each of the instruments? But you shouldn't be recording the reverb on those bounced tracks. Keep the reverb plug-in playing live when you do the final mix. Understand? The tracks should always remain dry.

    --You don't need to export tracks, you just bounce the instrument tracks down to new tracks inside the project. No need to re-import them.

    --You exported to one file? But that's what you said you were doing before. You want individual tracks for each instrument so you can work with volume automation on all of the tracks, leaving the Master fader alone.

    --The dynamics work is done on a per-instrument and per-group basis.

    Let me know what you're not clear on.

    Randy

  7. #7

    Re: dynamics with gpo and jabb

    Randy! I get it. I don't use the master fader. (and I did not for this render). I bounce the tracks within the project so I can make volume adjustments per instrument making sure that there is no reverb on these tracks. I could also group instruments and make adjustments on more than one instrument in the same track. When the adjustments are satisfactory, I then bounce all of the tracks to a new audio track, taking care that the bounced tracks have no reverb. When I do my final adjustments on the "final mix track", I should have the reverb on for the "final mix track".

    I am pretty sure this is what you are telling me. (I hope so anyway).
    Thanks for your patience with this.
    Jay

  8. #8

    Re: dynamics with gpo and jabb

    Quote Originally Posted by jaynkate01 View Post
    Randy! I get it. I don't use the master fader. (and I did not for this render). I bounce the tracks within the project so I can make volume adjustments per instrument making sure that there is no reverb on these tracks. I could also group instruments and make adjustments on more than one instrument in the same track. When the adjustments are satisfactory, I then bounce all of the tracks to a new audio track, taking care that the bounced tracks have no reverb. When I do my final adjustments on the "final mix track", I should have the reverb on for the "final mix track".

    I am pretty sure this is what you are telling me. (I hope so anyway).
    Thanks for your patience with this.
    Jay
    Howdy, Jay - There are of course many variations in how to do this. I'm trying to get across the most basic Sonar mixing 101 version. This was good, parroting back what you think you should do as per what I've said.

    And now - for some corrections. Class is in session again!

    --"'I don't use the master fader." - I didn't mean to not use a Master, just to not do volume automation on it. It's helpful to have a Master to avoid peaking out the final stereo pair in the Console View which is your interface's controls. One DEFINITELY never touches those interface controls. But you may have things going well, just a few spots where it peaks out. Then to use the Master as a way to pull everything back a smidge at that hot spot - that's a legitimate use of the Master. Also, most people put a compressor on the Master to tame the output a bit. You can have a Compressor at very gentle settings that helps you out but does Zero damage to your audio.

    --"I bounce the tracks within the project so I can make volume adjustments per instrument making sure that there is no reverb on these tracks." Right. And there's no difficulty in not including the reverb, because all you do is select the MIDI track and its associated Audio track, then bounce. The signal isn't going through a bus that has reverb on it. When you're moving from the MIDI realm, the signal you're bouncing is bypassing any reverb as you bounce.

    You can bounce a lot of tracks at once. Select all your MIDI and Audio tracks, and in the Bounce dialogue, choose Tracks as the source - then the list of all the Audio tracks will show up in that window.

    --"I could also group instruments and make adjustments on more than one instrument in the same track."
    I think you have that right, it got a bit sketchy there at the end. I think you meant "more than one instrument in the bus"--? But yes, you work on the volume relationship between all the woodwinds, for instance. But now you want to be able to control the entire woodwind section as a unit. Direct their output (bottom slot in the Console's track modules) to a Bus that you label "Woods." That bus goes to the Master. Now, when you need the woods to step back a bit, or move forward a bit, you can work with automation on the bus, moving the entire group at a time. But individual special automation was still done on each instrument's track.

    Note that you could have a Reverb Send on the Woods Bus, since they can usually share the same setting. No need for all those Sends on each track - Just the one send from the Bus to the Reverb Bus. Same for the Brass Bus, which has all the brass grouped - Strings, Percussion- whatever else you have in the piece. You may have a few instruments by themselves, like Piano or Harp - then they would need their own individual Reverb Send from their tracks.

    --"When the adjustments are satisfactory, I then bounce all of the tracks to a new audio track, taking care that the bounced tracks have no reverb. When I do my final adjustments on the "final mix track", I should have the reverb on for the "final mix track". No. There's an unnecessary step there. What you just described is making two master mixes for some reason. You just want one. Here:

    Do your mixing session(s). That involves panning everything where you want, automating each instrument's tracks as needed, then each group bus as needed. Getting all the Reverb settings the way you like. Moving the Master only when absolutely necessary. You haven't recorded any Reverb on to the tracks - the Reverb is playing live when you play the project.

    When you're set - Select the entire project, everything in it, plus a bit at the end to avoid chopping off reverb. Then under File, choose Export. The Source Category defaults to Entire Mix - that's what you want. Direct the Export to where you want - It's logical to have that go to the same project folder that contains everything in a particular project. Label the export, usually the name of the piece. Choose Sample Rate and Bit Depth - You can turn Dithering off if you've maintained the same Bit Depth throughout. I routinely record and Export at a Sample rate of 48000 and a Bit Depth of 32.

    That's your 2-track Master. - Ideally you then have a mastering program to do final touch ups to that. I use Sound Forge - Audacity is what you should get if you don't have it. Amazing that it's a free program - it's excellent. Bring the Export into there, check that the levels are good, trim excess time at the end. And maybe some volumes still bother you - bring up an entire section's volume if you want, within reason. Take out DC Offset.-- Make that puppy as good as you can get it.

    THEN make an MP3 copy of that to share. You can do that in Audacity, or bring it into Sonar by itself so you can make the MP3.

    OK - I added more detail this time. Hope it's helpful.

    That last sentence was your big problem, Jay - Do Not bounce everything to a 2-track, and then keep doing more work on it in Sonar. No need to do that. When you mix everything down to your master - that's the final step, outside of touching up that you hopefully also start doing in an additional sound editing program like Audacity.

    Randy

  9. #9

    Re: dynamics with gpo and jabb

    Wow! That is an excellent breakdown of things. This should help me out tremendously. I have saved this thread for reference. Now I need to practice the techniques that you described.

    I am always so impressed with your willingness to help people out on the forum. I, for one, appreciate it to no end.
    Thanks, Jay

  10. #10

    Re: dynamics with gpo and jabb

    Hi, Jay - I'm glad my spontaneous replies on this thread are useful. They're not especially well organized or succinct, but if you can extract the most useful bits from all the text, great!

    Your reply inspired me to do some screenshots this morning that may be good supplemental visual guides. I didn't add arrows and explanatory text, feeling that if you look around, things will make sense and the visuals could help lock in the mixing tips:

    ONE: A view of three vocal tracks with volume automation envelopes. Below are the project's Buses, all with automation except for the Master which is set just a bit below 0 DB. Note that the kind of detail in those volume envelopes is achieved by recording the automation in real time. Arm the volume sliders for recording automation, and all moves are captured as you play the project.



    TWO: A view of 9 of the sample project's 40+ tracks.

    From top to bottom: The Trim pots have been used, adjusting the overall volume of the tracks. After working with the mix for several days, I needed to lower the volume of all tracks, and in different amounts. Instead of having to throw out all my volume automation work, I adjusted each track's global volume with the Trim knob.

    EQ is on for all vocal tracks, rolling off bass, boosting highs. The first vocal track has a compressor in its FX bin.

    All of the visible tracks have varying amounts of Send going to the Reverb Bus. The first vocal track has its Send button armed for automation, because it was a case where different amounts of reverb were needed at different points.

    The red outline around a track fader indicates it's armed to record automation.

    At the bottom of the channel strips, right above the name slots, each track's output is directed to the appropriate Bus.

    On the right hand side are all of the project's buses.

    Things to notice in the bus section: The Master bus has EQ on - EQ balancing for the entire sum of all the tracks. Bass frequencies pile up quickly, and usually some roll off is needed for the final mix.

    A Compressor is in the Master's FX bin. It's set at a moderate level to help even out the dynamics of the project.

    The master's volume is just a fraction under the optimum volume of 0 DB, and no automation is used. The slider stays at that position throughout the project.

    The master's output is going to my interface made by Alesis.

    The Reverb bus has had its fader moved down a bit. This was a decision made later in the mixing process. Instead of working with all the controls in the reverb plug-in, I simply reduced the amount of reverb by lower its fader.

    The rest of the buses have all of their volume sliders armed to record automation. The results were seen in the first screen shot.

    Some buses have Sends for various group instruments going to the Reverb Bus.

    At the far right is the Interface's module - NEVER to be touched. It's good to keep that module visible to make sure the mix's volume never peaks into the red.



    THREE: A screen shot I did awhile back to simultaneously show three ways of monitoring when a signal goes into the red i.e. when there's digital clipping.

    1) In the track view, upper left, there's a slim VU meter which shows red at the top when the signal's gone into clipping.

    2) In the Console View, upper right, all the meters involved show the same red clipping signal: The offending audio track, the Master Bus, and the Interface's meters.

    3) In the Bus track view, always available at the bottom of the main Track View, the Track Inspector at the far left shows the clipping, and the arrow is pointing to the Waveform Preview button which has been turned on. This is a great tool. During playback, the wave form is drawn in real time. Wherever there's clipping, red shows up on that spot. Using that visual as a guide, the user can hand insert a volume envelope on the offending track precisely to momentarily take the volume down until it's no longer in the red at that spot.



    Hope the visual aids are helpful!

    Randy

Go Back to forum

Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •