B
banjo71
New member
I just got done mixing/mastering a project and I’ve learned a lot, thanks to the help of the band involved. Somebody asked me to share some insight that’s become apparent to me, as a small town, small time engineer who hasn’t had a lot of experience in doing this for a band, but enough to make this the tenth (count ‘em!) CD I’ve done around here locally.
Usually I refer the mix to someone else to master, but in this particular case I was paid to do both.
Recording
The band recorded their own material with some type of portable studio recorder. So they took care of mic placement. The most important thing they did for me was to record in 44.1kHz at a 24-bit depth. I would have preferred 48kHz (that’s the maximum frequency my A/D converter works), but that’s ok.
They recorded all the instruments and vocals in stereo (two separate mics) which means each song had a folder with 20+ tracks once I imported them all. Multiply this by 12 songs and this is a lot of Gigabytes folks. My Pro Tools can only handle 32 tracks anyway, and I got really close to not being able to do this project.
Mixing
The first thing I did was take each mono track and normalize it**, so when I combined two tracks for an instrument they were equal signal strength pretty much. Next, I recorded those two tracks for every instrument and vocal into a stereo track. Finally, I made the old mono tracks inactive and hid them.
I start by fading all the volumes on each stereo track down to -10dB. Next I gently compressed each instrument and vocal, if it sounds like they need it. Next, I added eq only where an instrument or vocal is needing it. (This is using RTAS plug-ins in Pro Tools).
Then I add two stereo auxillary tracks – reverb and small delay. I add sends from each stereo track for both reverb and delay, and then work a little of each into the mix.
Next I panned the instruments. Then the fun and challenging part… mixing with the volume faders using the auto-volume controls. It’s not really easy to explain this part, it just gets done eventually.
So then I added a master fader and raised/lowered the volume to make the signal fall between -10 dB and -4 dB, approximately.
Then I bounced the tracks to disk. This is the mix for the first song. I repeated all this for each song until you’re out of songs. I bounced them to a folder called Mixes.
Mastering
Each program is different. I use Steinberg’s Wavelab Essential 6. I create a new session called Master. I imported each song, in order according to the band, onto the region. I looked to make sure each song looked about the same volume in general. I put track markers at the beginning and end, and between each song at the splits. Now the region sounds like the mixes made in Pro Tools, with the splits at each new song.
Then I added some VST plug-ins: eq, compression, stereo expander, and peak limiter. On the eq, I added a bit of treble at 7kHz, take away some mids at about 1.6kHz, add some warmth at about 350Hz, and then do a low frequency cut at 50Hz, and a high frequency cut at about 17kHz. Minimal soft opto compression is what I usually use, no more than 3 dB of reduction in the mix. Stereo imaging, like eq and compression is really up to the engineer, I make it suit my taste for separation. Finally, I use a peak limiter set to a soft clip at about -0.9dB. Important: I found I need to save this preset plug-in chain! You can create other saved chains and compare sounds.
Then once I’ve chosen a chain I like best, I push the Master up into the red so it clips at a few dB. The last step in the mastering process is to apply a good 16-bit dither with noise shaping.
Then I’m ready to burn a disc.
I certainly don’t know much, never been schooled at it or anything. I’m sure I’m not doing something kosher. But it’s a step for me in the right direction.
** Normalizing probably isn't the most quality way to leveling out the two mono tracks to combine them into one stereo track. But that's where I'm at right now.
Usually I refer the mix to someone else to master, but in this particular case I was paid to do both.
Recording
The band recorded their own material with some type of portable studio recorder. So they took care of mic placement. The most important thing they did for me was to record in 44.1kHz at a 24-bit depth. I would have preferred 48kHz (that’s the maximum frequency my A/D converter works), but that’s ok.
They recorded all the instruments and vocals in stereo (two separate mics) which means each song had a folder with 20+ tracks once I imported them all. Multiply this by 12 songs and this is a lot of Gigabytes folks. My Pro Tools can only handle 32 tracks anyway, and I got really close to not being able to do this project.
Mixing
The first thing I did was take each mono track and normalize it**, so when I combined two tracks for an instrument they were equal signal strength pretty much. Next, I recorded those two tracks for every instrument and vocal into a stereo track. Finally, I made the old mono tracks inactive and hid them.
I start by fading all the volumes on each stereo track down to -10dB. Next I gently compressed each instrument and vocal, if it sounds like they need it. Next, I added eq only where an instrument or vocal is needing it. (This is using RTAS plug-ins in Pro Tools).
Then I add two stereo auxillary tracks – reverb and small delay. I add sends from each stereo track for both reverb and delay, and then work a little of each into the mix.
Next I panned the instruments. Then the fun and challenging part… mixing with the volume faders using the auto-volume controls. It’s not really easy to explain this part, it just gets done eventually.
So then I added a master fader and raised/lowered the volume to make the signal fall between -10 dB and -4 dB, approximately.
Then I bounced the tracks to disk. This is the mix for the first song. I repeated all this for each song until you’re out of songs. I bounced them to a folder called Mixes.
Mastering
Each program is different. I use Steinberg’s Wavelab Essential 6. I create a new session called Master. I imported each song, in order according to the band, onto the region. I looked to make sure each song looked about the same volume in general. I put track markers at the beginning and end, and between each song at the splits. Now the region sounds like the mixes made in Pro Tools, with the splits at each new song.
Then I added some VST plug-ins: eq, compression, stereo expander, and peak limiter. On the eq, I added a bit of treble at 7kHz, take away some mids at about 1.6kHz, add some warmth at about 350Hz, and then do a low frequency cut at 50Hz, and a high frequency cut at about 17kHz. Minimal soft opto compression is what I usually use, no more than 3 dB of reduction in the mix. Stereo imaging, like eq and compression is really up to the engineer, I make it suit my taste for separation. Finally, I use a peak limiter set to a soft clip at about -0.9dB. Important: I found I need to save this preset plug-in chain! You can create other saved chains and compare sounds.
Then once I’ve chosen a chain I like best, I push the Master up into the red so it clips at a few dB. The last step in the mastering process is to apply a good 16-bit dither with noise shaping.
Then I’m ready to burn a disc.
I certainly don’t know much, never been schooled at it or anything. I’m sure I’m not doing something kosher. But it’s a step for me in the right direction.
** Normalizing probably isn't the most quality way to leveling out the two mono tracks to combine them into one stereo track. But that's where I'm at right now.
Last edited: