23-02-2015, 06:20 PM
I just have a few comments to expand on what Blitzzz has said. Blitzzz is one of the musicians playing here and the kind fellow who provided this track, so take him seriously. I'm not disagreeing at all, just want to add some things to support his points.
I agree! Even before processing the signal in question, it's been EQ'd and compressed indirectly by the microphone and preamp. Every microphone will passively resist louder signals very mildly and that has tonal consequences, and if tube gear is involved there will be increased amounts of saturation and color with increased level (and it usually sounds great.) So anything we do in the mix is in addition to what's already been done by the recording engineer.
I like to think of EQ and compression as something that "shapes" audio in different ways. You can radically change the sound with these tools, and in some genres that's appropriate and even industry standard, like electro, EDM, trance, dubstep, so on. Most often though, these tools are better used for shaping the tracks to help them lock together in a way that flatters the performance.
For instance, a compressor is a good tool for shaping an electric guitar to lock it in with the bass or vis versa. The attack control shapes the transient and the release control shapes the sustain, and the ratio obviously controls how much shaping is applied. A fast attack is great for emphasizing the body of a sound, and I often use a snappy attack like this to flatter the character of a sound when something else might be obscuring it. Often this is preferable to EQ as a means of enhancing the body of (for instance) a snare, rather than boosting the mids or cutting the highs, because both tone and dynamics can be shaped.
As far as EQ is concerned, lots of people wax lyrical about spectral separation, which is indeed a good and necessary use for EQ, but once again, you're simply shaping the signal to help everything blend... but instead of shaping the level, you're shaping the tone. There are corrective uses for EQ like HPF and LPF, and maybe notching out the odd resonant frequency, and then there are shaping purposes... for instance reducing the the upper mid content in a distorted electric guitar where the vocal is more characterful, consequently emphasizing the lower frequencies where the guitar may have more power. Often this can be preferable to compression because you may not want to alter the transient or sustain of the guitar.
Rides will be more transparent and pro mixes use tons of it, but don't think you should only automate the fader! Most of your "coarse" level automation (different static levels for each song section) should be done during gain staging IMHO, and quite often, tonal and dynamic shaping are only helpful for certain passages, or just as often are required in different amounts from moment to moment. For instance you may want to adjust the ratio of a compressor that you're using to enhance the fullness of a snare drum when the beefy electric guitar isn't obscuring it so much, or you may want to completely turn off the 2000khz to 4000 khz peaking dip on the electric guitar during an instrumental break where there's no lead vocal to worry about masking. And then you may want to simply ride the fader a db or 2 on the odd dropped lyric.
My main point is the automation isn't the only answer and it isn't always the best... it's a different tool for a different purpose. Different horses for different courses, yeah? Riding up every single vocal note when the guitar chugs are covering the sustain tails might not be the best strategy, because then you're might wind up having to automate a de-esser to keep the sibilants under control, and you may also wind up having to automate the guitar transients now the the vocal notes are covering them up... you can spend hours penciling this stuff in when a little light compression and EQ could have shaped the guitar and vocal around each other moresmoothly. Obviously you'll still likely need to ride the faders to avoid EXCESSIVE EQ and compression, which is where Blitzzz has an excellent point relevant to this particular mix, but using these tools in combination, each for their specific purpose with a purpose-driven goal in mind, is when you're really mixing audio.
And let's not forget... a little compression and (sometimes) EQ on the master buss can smoothly blend things such that much less compression, EQ and automation are needed on the individual tracks!
(23-02-2015, 02:50 PM)Blitzzz Wrote: - don't use EQ/compression if you don't have a clear goal and vision how this song should sound. use the reference song to check if you are still on track
I agree! Even before processing the signal in question, it's been EQ'd and compressed indirectly by the microphone and preamp. Every microphone will passively resist louder signals very mildly and that has tonal consequences, and if tube gear is involved there will be increased amounts of saturation and color with increased level (and it usually sounds great.) So anything we do in the mix is in addition to what's already been done by the recording engineer.
(23-02-2015, 02:50 PM)Blitzzz Wrote: - Volume automation is mixing in its purest form. start riding faders with automation and make smart pan decisions for every track. only use eq and compression to take care of the rest.
I like to think of EQ and compression as something that "shapes" audio in different ways. You can radically change the sound with these tools, and in some genres that's appropriate and even industry standard, like electro, EDM, trance, dubstep, so on. Most often though, these tools are better used for shaping the tracks to help them lock together in a way that flatters the performance.
For instance, a compressor is a good tool for shaping an electric guitar to lock it in with the bass or vis versa. The attack control shapes the transient and the release control shapes the sustain, and the ratio obviously controls how much shaping is applied. A fast attack is great for emphasizing the body of a sound, and I often use a snappy attack like this to flatter the character of a sound when something else might be obscuring it. Often this is preferable to EQ as a means of enhancing the body of (for instance) a snare, rather than boosting the mids or cutting the highs, because both tone and dynamics can be shaped.
As far as EQ is concerned, lots of people wax lyrical about spectral separation, which is indeed a good and necessary use for EQ, but once again, you're simply shaping the signal to help everything blend... but instead of shaping the level, you're shaping the tone. There are corrective uses for EQ like HPF and LPF, and maybe notching out the odd resonant frequency, and then there are shaping purposes... for instance reducing the the upper mid content in a distorted electric guitar where the vocal is more characterful, consequently emphasizing the lower frequencies where the guitar may have more power. Often this can be preferable to compression because you may not want to alter the transient or sustain of the guitar.
Rides will be more transparent and pro mixes use tons of it, but don't think you should only automate the fader! Most of your "coarse" level automation (different static levels for each song section) should be done during gain staging IMHO, and quite often, tonal and dynamic shaping are only helpful for certain passages, or just as often are required in different amounts from moment to moment. For instance you may want to adjust the ratio of a compressor that you're using to enhance the fullness of a snare drum when the beefy electric guitar isn't obscuring it so much, or you may want to completely turn off the 2000khz to 4000 khz peaking dip on the electric guitar during an instrumental break where there's no lead vocal to worry about masking. And then you may want to simply ride the fader a db or 2 on the odd dropped lyric.
My main point is the automation isn't the only answer and it isn't always the best... it's a different tool for a different purpose. Different horses for different courses, yeah? Riding up every single vocal note when the guitar chugs are covering the sustain tails might not be the best strategy, because then you're might wind up having to automate a de-esser to keep the sibilants under control, and you may also wind up having to automate the guitar transients now the the vocal notes are covering them up... you can spend hours penciling this stuff in when a little light compression and EQ could have shaped the guitar and vocal around each other moresmoothly. Obviously you'll still likely need to ride the faders to avoid EXCESSIVE EQ and compression, which is where Blitzzz has an excellent point relevant to this particular mix, but using these tools in combination, each for their specific purpose with a purpose-driven goal in mind, is when you're really mixing audio.
And let's not forget... a little compression and (sometimes) EQ on the master buss can smoothly blend things such that much less compression, EQ and automation are needed on the individual tracks!
I'm grateful for comments and suggestions. Thank you for listening!