J
jfreeman373
New member
In response to alot of peoples statements towards me without quoting them all,...here is my reply.
I think alot of what is being argued comes down to is; "what represents a good mix ?". I think that changes, in regard to the type of music you ARE mixing. I am not here to slam one type of music or the other, but this is how I feel about it.
There are drastic different ways to mix, here is one example:
In most modern music like rock,pop, industrial, and alternative the goal is to create a particular "sound". That "sound" is as much an instrument as is the instruments being played by the musicians. In getting that sound you are mostly dealing with frequencies,dynamics and sound stage being last. You don't hear too much music anymore like Pink floyd did with ambient effects and delays.
As long as you balance out the frequency ranges of your instuments, smooth out unwanted dynamics, and pan your instruments where you want them, you should have a good start. It can be more complicated than trying to do a recording of say,... more "traditional" music. But most of the time the amount of delay isn't that critical in a pop recording between say 30 or 35ms delay on a snare. People aren't really listening to that. So... you just pick a delay you think sounds good. Most speakers have terrible phase response so you probably can't tell that much of a defference on a particular setting anyway.
Now to the other end of the spectrum... say jazz. The goal of our project is to make the recording sound like it was recorded in a hall or club using a pair of stereo mics, when in actuallity... it was multitracked.
THIS can be very difficult if almost impossible to do if you don't have excellent monitors and you try to mix by what you HEAR.
Most of the time you should't have to eq anything if it was recorded properly cause your not trying to make room for two heavily distorted guitars. The idea NOW is NOT compression or eq, but time domain issues. Transformation of a multitracked tune into a stereo live recording is our goal. The only way to accomplish this is in the world of time and the reflections of sound in time.
On crappy monitors this is an almost impossible thing to do because you just CANNOT hear "in any environment" the change that you are making, because the adjustments are so very small and intricate.
I have a technique for close micing a drum set and creating a virtual "stereo recording". Here it is.
Step one: Don't worry about phase considerations when micing the drums because we will correct for that later in our DAW. Just put them where it sounds good and DON'T EQ THEM, pick the right mic.
step 2: record the track and if you have the ability to record 8 tracks at once, do it...(And let the english see you do it!) Then assign each mic to its own seperate track. Next... align the attack and phase of all eight tracks so that they line up perfectly to the nth degree.
step 3:Make a copy of those tracks and place them on tracks 9-16. We will NOT be using any panning in our technique to create our sound stage because we will be using delays to simulate that. Normalize all your tracks, remove silence, so forth and so on.
We are just using the delay effect to move a particular track in time, as opposed to using it as an effect. Now assign tracks 9-16 to an aux bus and attach a delay to that. Don't use any feedback or predelays or anything.
In this effect we will be creating the "delay of the distance between our VIRTUAL XY Stereo Coincident mics".I hope you are all familiar with that technique.
Now how far apart are those 2 mics? A couple inches... which is equivalent to a couple milleseconds at most. Sound would of course reach those 2 mics at different (very small) times. Pick a delay of say 15ms and set it to the delay on the auxillary bus. Make sure the time you choose keeps the coherent phase response you created earlier, or you will get comb filtering at output.
step 4: Pan the first eight tracks hard left, and pan the next eight tracks hard right.Send the output of the aux bus of tracks 9-16 and the outputs of tracks 1-8 to the main output.
Now you have a true stereo representation of the 2 virtual mics and the distance betwween them and the sound "effect" equivalent of thier spacing.
Step five: Optional. Now you need to calulate the distance the drum kit is from those 2 virtual mics 10 ft, 20 ft however far back you want it sound like the live recording was made. We are not done though.
Think about this though. isn't the floor tom further away than say the kick or the toms? In this... wouldn't the sound from the kick and toms reach the stereo mics BEFORE the sound from the floor tom? It maybe a short distance but it matters.
Now assign a delay to all sixteen tracks and set them to full wet and 0 dry signal... with all DELAY "effects" OFF.
Now... which ever drums are the same distance from the mic cluster, you can remove the delay from that track and it's corresponding COPY track.
Don't mess with the delay that is on that aux bus. That has already been set for our stereo mic simulation.Only on the individual delays of the individual tracks are we making adjustments now. But what ever change you do to the snare on track one, you should be doing to the snare track on track 9 and so forth for all the tracks.
Now... set different delay times for the individual drums. Set the snare to 3ms, the cymbals to 4 or 5 ms, the floor tom to 8 ms and make sure that all the delays are smaller than 15 ms. The further the drum is from the mic, the larger the delay for its track should be and vice versa.
Now just ad whatever type of reverb you want on the master output and run the final stage through that to simulate your room environment.Here you can do anything that you want. Just make sure that all tracks are getting to the rever POST individual track delays and POST the auxillary bus delay that tracks 9-16 should be going through.
You could even go a step further and assign an eq to each individual delay to accomodate for high frequency roll offs of certain frequencies in regard of the distance to the VIRTUAL xy stereo mics.
This technique wields excellent results.
To reproduce reality... you must simulate reality. To do that you must use the laws of physics in which sound is subject. This culmination of minor adjustments can actually trick the ear!
It is the subconscious aspects of sound that most of us don't think about when we mix. It's new for me but I can tell you it works. This kind of thing you just can't notice on monitors of low caliber because they smear the image so much.
If you have good monitors you can hear the differnce between a 3 and 5ms delay. That's all I have been saying. If you are recording modern music you probably aren't concerned with such a technique, you are more concerned with getting the kick to thump with compression and eq.I realize most people in this forum probably don't have an 8 track sound card but if you do.... Try it out if your computer can handle it. Your gonna need a fast one to hear it in real time because of all the delays. If it's not fast... just make the settings that I said and render it down. You will most likely be happy with the results.You probably won't be able to pinpoint WHY it sounds better but it just will, and you will know it.
This can all be done with 8 mics, a sound card with 8 inputs, an 8 channel mixer, and cakewalk Pro 8 or 9 with cakewalks standard effects. For your reverb effect try using Sonic foundry's acoustic mirror plugin. (That runs like shit in cakewalk by the way.) Or a good outboard reverb.
If anyone tries it,... let me know the results you had. It's not that difficult and should only take a couple hours to do. Removing the silence in the tracks is the worst part for me cause I do it by hand.
I know this was long but i tried not to leave anything out.
On a previous note, if the frequency response of your monitor is out of flat by more than 3db I believe it is OK to eq it out, as long as you have a good eq, but note that you will be changing the phase response of the speaker ALSO by doing that. So I guess It's up to you to decide which you want, flat response, or proper phase. But if they aren't flat to begin with, the phase is probably terrible anyways so you can't hurt much by eqing it. You shouldn't make changes greater than 3db with your graphic though. Use some pink noise and an analyzer but make the minimal corrections you can. EQ CAN tear your speakers response apart.
It is always better to cut a rogue frequency than to gain a low one. So In that regard I think Boray AND Blue Bear are right at the same time. It comes down to what you need at a given time. If you know what your doing and what your trying to get and you know what techniques you should be using, there are times that both issues have thier valid use.If you know your equipment you know what you can get away with and when to do it.
Finally,
BYE NOW
Jfreeman
I think alot of what is being argued comes down to is; "what represents a good mix ?". I think that changes, in regard to the type of music you ARE mixing. I am not here to slam one type of music or the other, but this is how I feel about it.
There are drastic different ways to mix, here is one example:
In most modern music like rock,pop, industrial, and alternative the goal is to create a particular "sound". That "sound" is as much an instrument as is the instruments being played by the musicians. In getting that sound you are mostly dealing with frequencies,dynamics and sound stage being last. You don't hear too much music anymore like Pink floyd did with ambient effects and delays.
As long as you balance out the frequency ranges of your instuments, smooth out unwanted dynamics, and pan your instruments where you want them, you should have a good start. It can be more complicated than trying to do a recording of say,... more "traditional" music. But most of the time the amount of delay isn't that critical in a pop recording between say 30 or 35ms delay on a snare. People aren't really listening to that. So... you just pick a delay you think sounds good. Most speakers have terrible phase response so you probably can't tell that much of a defference on a particular setting anyway.
Now to the other end of the spectrum... say jazz. The goal of our project is to make the recording sound like it was recorded in a hall or club using a pair of stereo mics, when in actuallity... it was multitracked.
THIS can be very difficult if almost impossible to do if you don't have excellent monitors and you try to mix by what you HEAR.
Most of the time you should't have to eq anything if it was recorded properly cause your not trying to make room for two heavily distorted guitars. The idea NOW is NOT compression or eq, but time domain issues. Transformation of a multitracked tune into a stereo live recording is our goal. The only way to accomplish this is in the world of time and the reflections of sound in time.
On crappy monitors this is an almost impossible thing to do because you just CANNOT hear "in any environment" the change that you are making, because the adjustments are so very small and intricate.
I have a technique for close micing a drum set and creating a virtual "stereo recording". Here it is.
Step one: Don't worry about phase considerations when micing the drums because we will correct for that later in our DAW. Just put them where it sounds good and DON'T EQ THEM, pick the right mic.
step 2: record the track and if you have the ability to record 8 tracks at once, do it...(And let the english see you do it!) Then assign each mic to its own seperate track. Next... align the attack and phase of all eight tracks so that they line up perfectly to the nth degree.
step 3:Make a copy of those tracks and place them on tracks 9-16. We will NOT be using any panning in our technique to create our sound stage because we will be using delays to simulate that. Normalize all your tracks, remove silence, so forth and so on.
We are just using the delay effect to move a particular track in time, as opposed to using it as an effect. Now assign tracks 9-16 to an aux bus and attach a delay to that. Don't use any feedback or predelays or anything.
In this effect we will be creating the "delay of the distance between our VIRTUAL XY Stereo Coincident mics".I hope you are all familiar with that technique.
Now how far apart are those 2 mics? A couple inches... which is equivalent to a couple milleseconds at most. Sound would of course reach those 2 mics at different (very small) times. Pick a delay of say 15ms and set it to the delay on the auxillary bus. Make sure the time you choose keeps the coherent phase response you created earlier, or you will get comb filtering at output.
step 4: Pan the first eight tracks hard left, and pan the next eight tracks hard right.Send the output of the aux bus of tracks 9-16 and the outputs of tracks 1-8 to the main output.
Now you have a true stereo representation of the 2 virtual mics and the distance betwween them and the sound "effect" equivalent of thier spacing.
Step five: Optional. Now you need to calulate the distance the drum kit is from those 2 virtual mics 10 ft, 20 ft however far back you want it sound like the live recording was made. We are not done though.
Think about this though. isn't the floor tom further away than say the kick or the toms? In this... wouldn't the sound from the kick and toms reach the stereo mics BEFORE the sound from the floor tom? It maybe a short distance but it matters.
Now assign a delay to all sixteen tracks and set them to full wet and 0 dry signal... with all DELAY "effects" OFF.
Now... which ever drums are the same distance from the mic cluster, you can remove the delay from that track and it's corresponding COPY track.
Don't mess with the delay that is on that aux bus. That has already been set for our stereo mic simulation.Only on the individual delays of the individual tracks are we making adjustments now. But what ever change you do to the snare on track one, you should be doing to the snare track on track 9 and so forth for all the tracks.
Now... set different delay times for the individual drums. Set the snare to 3ms, the cymbals to 4 or 5 ms, the floor tom to 8 ms and make sure that all the delays are smaller than 15 ms. The further the drum is from the mic, the larger the delay for its track should be and vice versa.
Now just ad whatever type of reverb you want on the master output and run the final stage through that to simulate your room environment.Here you can do anything that you want. Just make sure that all tracks are getting to the rever POST individual track delays and POST the auxillary bus delay that tracks 9-16 should be going through.
You could even go a step further and assign an eq to each individual delay to accomodate for high frequency roll offs of certain frequencies in regard of the distance to the VIRTUAL xy stereo mics.
This technique wields excellent results.
To reproduce reality... you must simulate reality. To do that you must use the laws of physics in which sound is subject. This culmination of minor adjustments can actually trick the ear!
It is the subconscious aspects of sound that most of us don't think about when we mix. It's new for me but I can tell you it works. This kind of thing you just can't notice on monitors of low caliber because they smear the image so much.
If you have good monitors you can hear the differnce between a 3 and 5ms delay. That's all I have been saying. If you are recording modern music you probably aren't concerned with such a technique, you are more concerned with getting the kick to thump with compression and eq.I realize most people in this forum probably don't have an 8 track sound card but if you do.... Try it out if your computer can handle it. Your gonna need a fast one to hear it in real time because of all the delays. If it's not fast... just make the settings that I said and render it down. You will most likely be happy with the results.You probably won't be able to pinpoint WHY it sounds better but it just will, and you will know it.
This can all be done with 8 mics, a sound card with 8 inputs, an 8 channel mixer, and cakewalk Pro 8 or 9 with cakewalks standard effects. For your reverb effect try using Sonic foundry's acoustic mirror plugin. (That runs like shit in cakewalk by the way.) Or a good outboard reverb.
If anyone tries it,... let me know the results you had. It's not that difficult and should only take a couple hours to do. Removing the silence in the tracks is the worst part for me cause I do it by hand.
I know this was long but i tried not to leave anything out.
On a previous note, if the frequency response of your monitor is out of flat by more than 3db I believe it is OK to eq it out, as long as you have a good eq, but note that you will be changing the phase response of the speaker ALSO by doing that. So I guess It's up to you to decide which you want, flat response, or proper phase. But if they aren't flat to begin with, the phase is probably terrible anyways so you can't hurt much by eqing it. You shouldn't make changes greater than 3db with your graphic though. Use some pink noise and an analyzer but make the minimal corrections you can. EQ CAN tear your speakers response apart.
It is always better to cut a rogue frequency than to gain a low one. So In that regard I think Boray AND Blue Bear are right at the same time. It comes down to what you need at a given time. If you know what your doing and what your trying to get and you know what techniques you should be using, there are times that both issues have thier valid use.If you know your equipment you know what you can get away with and when to do it.
Finally,
BYE NOW
Jfreeman
