Mixing + Mastering For Only 5$?

[mykalallen]

http://fiverr.com/silentstudio

I'm willing to at least try it and see how they do, once I make my E.P.

How do you all feel about it?

[mykalallen]

Sorry about that - I have no affiliation at all with the link, just like you assumed. I was thinking I'd let them do one song.. but I felt like I wouldnt be able to sit there and listen to it with them and what not, so how would they even know what I liked? I guess its only 5$..

What are better options for mixing and mastering then? I make extremely simple songs, and have looked a lot into all of that stuff myself, but its all pretty overwhelming.

[mykalallen]

Well, let's try to find a logical place to begin this discussion.

• Your inquiry was about "mixing & mastering". If you're not already aware, those are both finishing touches. Only when your song is to the point where each of the individual tracks (vocals, guitar, etc) sound exactly the way you want them, do you even consider those two finishing steps. In other words, every individual track has to be played cleanly, sung cleanly, contain reasonably low noise levels, no distortion, etc. To reach that point, you need the musical ablility, proper equipment and a fundamental knowledge of how to use that equipment and a reasonably quite recording enviroment. So my question to you is this......does my description sound like you?
  
• If the answer is yes, then it may be reasonable for you to be thinking about topics like mixing & mastering.
  
• But if the answer is no, you're getting way ahead of yourself and probably need to back up a bit.
  
• As I'm unfamiliar with your work, what are you currently using to record your songs (mics, recording software/device, guitars, etc)?
  

These questions should help us get going in an appropriate direction.

Tom

Thank you very much for that - it's good to get a touch of reality.

I have medium diaphram condenser, and a few sm 57s.

I dont have any physical compressors.. or anything else really besides an interface.

I also have a house that I'll be recording in.. in the middle of nowhere, literally, so noise isnt an issue.

I have all the songs ready, and a decent idea of how to record each instrument. Just trial and error though, I suppose.

Most of the songs are just going to be acoustic guitar and voice, but there will be some electric guitar and bass every once in a while as well.

Edited December 9, 2012 by mykalallen

[Prometheus]

You most definitely can't master a track without a compressor, it would be out of the question. Stone wall compression (i.e. limiting) is one of the maintstays of the mastering process. Multiband Compression is also beneficial sometimes, not to mention de-essing. None of this can be done without compressors.

Mastering a track is easy for someone with the right experience and doesn't take long. It shouldn't be hard to find a pro engineer who will master a few tracks for a very reasonable rate in the current climate where pro sound engineers are not having a good time of it.

Just trial and error though, I suppose.

You should never apply any change to a voltage analogue or digital information that your brain hasn't applied five minutes earlier. Rule number one of producing records, never guess and press. If you don't know why you're doing something, you shouldn't be doing it.

[mykalallen]

You most definitely can't master a track without a compressor, it would be out of the question. Stone wall compression (i.e. limiting) is one of the maintstays of the mastering process. Multiband Compression is also beneficial sometimes, not to mention de-essing. None of this can be done without compressors.

Mastering a track is easy for someone with the right experience and doesn't take long. It shouldn't be hard to find a pro engineer who will master a few tracks for a very reasonable rate in the current climate where pro sound engineers are not having a good time of it.

You should never apply any change to a voltage analogue or digital information that your brain hasn't applied five minutes earlier. Rule number one of producing records, never guess and press. If you don't know why you're doing something, you shouldn't be doing it.

What I meant in reference to trial and error was mic placement, and room types and what not, not specifically those changes. By a compressor, do you mean a physical compressor? Can't you do said things digitally, with enough practice and/or research. I'm only recording a demo of myself, and of course I'd like to make it sound great, but I'm pretty much done buying recording equipment, at least for now, and if I can find a better digital alternative then I will. I'm just not sure what hardware you can substitute for plugins.. that's something I need to come to terms with.

[Prometheus]

What I meant in reference to trial and error was mic placement, and room types and what not, not specifically those changes. By a compressor, do you mean a physical compressor? Can't you do said things digitally, with enough practice and/or research. I'm only recording a demo of myself, and of course I'd like to make it sound great, but I'm pretty much done buying recording equipment, at least for now, and if I can find a better digital alternative then I will. I'm just not sure what hardware you can substitute for plugins.. that's something I need to come to terms with.

You can use plugin compressors. A high end one will be indistinguishable from an outboard one, provided it's used cleverly. Everyone does this in this day and age. Buying sixteen outboard compressors for a thirty two track studio would be a complete waste of money when one plugin compressor can do the same job.

[mykalallen]

I guess I just have to learn how to use that - that's the hardest part. Finding how much compression I would need for acoustic guitar/vocals.. I've looked around a lot, and there doesn't seem to be any standard, seeing as each mix has its own unique instruments and what not, so I guess the settings would be a trial and error type thing.

I've read a lot about limiting and finding buzz and nasty noise and cutting it out.. but one of the things I dont understand is which setting do I do to subtract a decibel or add one to a certain frequency? Is it the band one? Or notch.. or what?

I'm pretty confident I can get a pretty decent sound, and this is just a demo - I'd just like to be able to do some equalizing, compressing, mixing.. something to just add that extra crispness or "Ooomph" or what not.

Edited December 10, 2012 by mykalallen

[Prometheus]

I guess I just have to learn how to use that - that's the hardest part. Finding how much compression I would need for acoustic guitar/vocals.. I've looked around a lot, and there doesn't seem to be any standard, seeing as each mix has its own unique instruments and what not, so I guess the settings would be a trial and error type thing.

I've read a lot about limiting and finding buzz and nasty noise and cutting it out.. but one of the things I dont understand is which setting do I do to subtract a decibel or add one to a certain frequency? Is it the band one? Or notch.. or what?

I'm pretty confident I can get a pretty decent sound, and this is just a demo - I'd just like to be able to do some equalizing, compressing, mixing.. something to just add that extra crispness or "Ooomph" or what not.

It takes a clever man years to do that. You're right, you'll never apply the same settings twice. It's impossible to define a process that you can follow that tells you exactly what compressor settings to use. You have to learn what all the basic controls are about, attack, release, threshold, ratio and compensation, and then you have to practice incessantly until you get results. It shouldn't be done haphazardly. You should always have an idea why you're making a change.

Same with EQ. The thing to do with EQ is practice using subtractive EQ. Taking out unwanted frequencies is far more effective and transparent than additive EQ.

In answer to your question, it's all of them. Any EQ will subtract or add to certain frequencies. In very simplistic terms, the idea for the spectral domain is to make the mix fairly flat up to about 18KHz. EQ is also essential to achieve clarity by creating separation between mix elements. Contrary to a fairly popular misconception, that is not done by panning. Panning is actually used to widen the stereo field.

[mykalallen]

A lot of this sounds like another language to me, and I hope you dont feel too put out sitting here explaining it.

When you say "make a mix fairly flat up to about 18KHz" . What do you mean?

And.. what is widening, really? I know youre supposed to keep the drums center.. but yeah, thats pretty much sums up what I know about widths.

[Prometheus]

A lot of this sounds like another language to me, and I hope you dont feel too put out sitting here explaining it.

When you say "make a mix fairly flat up to about 18KHz" . What do you mean?

And.. what is widening, really? I know youre supposed to keep the drums center.. but yeah, thats pretty much sums up what I know about widths.

it's my bed time, but I'll go into it tomorrow when I get home from work.

[mykalallen]

it's my bed time, but I'll go into it tomorrow when I get home from work.

Thanks for the help - Completely understand that.

[Prometheus]

Okay... On the point about having a flat spectral domain. Here's a video, a very poor quality one, that I captured from a mixdown that I did.

http://www.youtube.com/watch?v=XVR7dSrQOB8&feature=youtu.be

As you can see, from 20Hz to 22KHz , the peaks graph roughly a straight line. There is slightly higher peaks at low frequencies, but it's a straight line all the way up to 22Khz where it's low pass filtered to keep within the standard Nyquist Limit, which is 44.1KHz. The reason for this figure of 44.1KHz is that to recreate a sine wave acurately you have to sample two points on it. This works because every sine wave is a predictable shape and the highest audio frequency a human can here is 20 KHz, and that's actually being kind. It's nearer 16 for most people by the time they reach adulthood. So, to sample two points in a 20Khz sine wave, you need a sample rate of 40KHz . Sampling at 44.1 adds a bit of attrition to this which is done because of other factors that aren't really important in this explanation.

Low pass filtering at 22Khz means that there will be no alaising errors. Not that you'd be able to hear them. More importantly, it means that you're not wasting disk space and processor time reproducing frequencies that only the cat could hear and no household stereo systems would reproduce.

On the other point about panning, that's already been dealt with on another thread. Here's a copy and paste of what I posted on it.

First of all, if a sound comes from a mono source, it should be mixed in mono, so it's far better to use panning than stereo enhancers to create the stereo field. Stereo enhancers are usually single band, so if used at the mastering stage they tend to widen the parts of the mix (low frequencies) that should actually be mono.

If you want a wide guitar sound, record the guitar part twice or double mic it with the two mics equidistant from the amp then pan one track to the left and one to the right. It's a mistake to do a panning hardover (panning all the way to eight o'clock and four o'clock) on both tracks since the field becomes too wide and leaves a "hole" in the centre of the mix. The trick is to pan out a little to the left and right until the spread sounds good. If there are going to be more double tracks, they can then be panned incrementally further and further out so they have their own space.

The kick drum and bass guitar (or any basso instrument) should be in the centre. Bass frequencies are omni-directional, which is to say that human beings are terrible at assigning direction to them. This is because the waves are bigger than the width of the human head and can therefore hit the timpanic membranes in both ears simultaneously.

Another thing to consider is that basso sounds have more energy than treble sounds, as is evidenced by the fact that woofers are much bigger, stiffer and more powerful than tweeters. This means that taking a basso sound off centre is going to seriously unbalance a mix. This used to actually wreck cutting needles in the old days of vinyl when mastering actually was as critical and difficult as mastering engineers still pretend it is.

Lead vocals should always be in the centre and backing vox double tracks can be panned using the method detailed above of starting near the centre and working to the outside.

Always listen to the mono sum of the mix as well to check if there are going to be problems with people playing the track over ghetto blasters or PC speakers or car stereos that are not really capable of generating an accurate stereo field.

As Darmin says, EQ is ususally the most effective way of achieving separation in a mix, not wide panning.

[Prometheus]

Additional:

The reason low frequency instruments and sounds should be panned centre is two fold. One reason is that they cause far worse phasing problems than high frequency sounds and the other reason is that bass sounds have long enough wavelengths to hit both ears simultaneously, which means that basso frequencies are omnidirectional to humans. That's why surround systems only need one sub woofer.

[mykalallen]

This is all extremely useful information.. Thank you.

I had read your post on panning before - but I far more understand what you're saying there.

What I really don't get is the whole "Sampling two points" bit, and are you saying, every song should peak between 20-22KHz?

I understand that the human ear can really only hear 20, and even that unsuccesfully, but what else did you mean by that top section? That low passfilter will make that the max of course, but the other stuff.. Yeah, I'm quite an novice.. so a lot of this lingo is lost with me..

Edited December 12, 2012 by mykalallen

[Prometheus]

This is all extremely useful information.. Thank you.

I had read your post on panning before - but I far more understand what you're saying there.

What I really don't get is the whole "Sampling two points" bit, and are you saying, every song should peak between 20-22KHz?

I understand that the human ear can really only hear 20, and even that unsuccesfully, but what else did you mean by that top section? That low passfilter will make that the max of course, but the other stuff.. Yeah, I'm quite an novice.. so a lot of this lingo is lost with me..

Bear in mind I'm explaining this in very simplistic terms. The maths behind this is very complicated, and I'm not much of a mathematician. Fortunately, computers handle the maths so sound engineers, unless they specialize in sampling theory, don't have to.

Every sound wave is a sine wave, and every sine wave looks like this:

You have zero amplitude at 0 degrees, a peak positive amplitude at 90 degrees, a peak negative amplitude at 270 degrees and zero amplitude at 360 degrees et cetera, all in increments of 90 degrees.

In order to have enough information to recreate this sine wave as digital information, you have to sample it twice within each 360 degrees. After that, the microprocessor in your analogue to digital converters can interpolate the rest of the information.

The reason that it's required, whether you're working this way or the old school way of working with voltage analogues, to have a spectral domain of 20Hz to 22KHz is because that encompasses the entire range of human hearing. In much the same way that if you want to visually simulate a rainbow you have to encompass the entire visual spectrum, the same goes if you want to acurately represent a sound.

To cut to the chase, the human hearing mechanism is geared towards amplifiying sounds at the key frequencies of the human voice, that is the midrange frequencies. In sound engineering, the mixing engineer and to a lesser extent the mastering engineer have to undo this feat of biomechanical engineering that goes on in the human ear. That is why a lot of midrange has to be subtractively EQ'd out at the mixing and mastering stage.