|
Page 3 of 3 Recording Tips Part 3
Frequencies and Equalization (EQ)
Frequencies.
Let's start with a sine wave.
The signal starts at zero, rises through a parabolic curve for half a second and falls through zero to the negative half of the cycle till it gets back to zero, at the one second mark in time.
And so we have one cycle of the signal wave, per second.
This used to be called a cycle, but has changed and now represented by the symbol Hz which you may have seen for example 20Hz ~20kHz.
Outside of this range you can feel in your body, but you can't hear it.
Some would argue that if you can feel it, you can hear it and I would have to agree, so let's leave this in general terms for the moment.
The majority of frequencies that we will be dealing with, are in that range 20Hz~20kHz.
That's 20 cycles of our sine wave per second up to 20,000 cycles per second.
You may have heard the term A 440.
The A note when played at concert pitch resonates at 440 Hz per second.
An A note one octave below that note is 220Hz and one above is 880Hz.
At least that's what happens when you set a tone generator to a particular frequency. A pure signal.
Instruments are a completely different animal.
They produce multiple signals (sine waves) all at the same time.
The underlying fundamental note of an A note played on say a guitar, will certainly have a predominance of 440 cycles in what you hear, but it will also have numerous other tones (harmonics) coupled to it.
These additional tones are the character of the instrument being played.
The are often called the harmonics of a note.
Every instrumental has its own unique set of tones.
How a musician plays the instrument, can greatly affect the sound as well.
If you strike the strings of a guitar back near the bridge you will get a sharper higher emphasis in the frequencies. Play the strings over the sound hole you will get deeper bass tones or more mellow sounds. Even the way you hold your pick or the type of pick you use, can change the sound.
I remember sticking Drawing Pins into the strike pads on my mum's upright piano, to get it to sound more like a harpsichord, or at least a bit more Honky Tonk.
I think I got a hiding for that, but it worked.
The type of wood used, the size and design of the sound box etc etc.
All of these things can change the overall sound of the instrument.
When you are mixing, it is wise to keep these things in mind, especially with the faithful reproduction of acoustic instruments.
Faithful Reproduction being ... what is coming out of your speakers, sounds exactly the same as it does when you are siting in a room listening to the instrument, without all the electronics in between.
This is what I call transparency.
To a greater degree we are trying to do this, so that we don't put in any extra color in the sound.
From there it can be a matter of taste.
I bit more bass, middle, treble in the instrument is up to you, but I always go for transparency (as much as your gear will allow) and then start enhancing various frequencies to taste. Using an analogy with food... Over cook the meal and no amount of sauce will save it.
The cooking is in the actual recording in the first place.
We put the sauces on in the mix.
If you look at the accompanying chart (Instrument Frequencies.pdf) you will see that each instrument has it's own dominant frequency range, as well as it's own harmonic frequency range.
That's what we are dealing with in the mix.
Start off by listening to just one instrument at a time.
The frequencies I suggest are just a starting point, to put you in the ball park of the particular instrument.
If you can't take a day or 2 between recording and mixing, at least try to give yourself a good 30 minute break, so that you can attack the mix with 'fresh ears'.
Listening fatigue is very common in studios, especially after you heard the same song hundreds of times.
I've done plenty of marathon mixes that sound great on the night, but the next day, you realize they are swimming in reverb or the overall mix has way too much bass or treble or something.
Invariably this can come from mixes that are done at higher volumes as well.
Oooo they sounded so good last night.
When you get a good mix, write down your settings, because you will probably use those settings again and again to get a 'quick mix' together.
Saves doing all the work twice.
Once you have them saved, you can also charge off and experiment with different gizmos and know that if you get lost, you can always go back to the original settings quickly.
Practice makes perfect and the more you do it, the more you can train yourself to pick out one particular sound out of many. It doesn't come straight away, but it will come over time.
I would also like you to consider the proximity effects while listening.
Make sure you are sitting dead centre of left and right speakers, so that you don't get a skewed version of where centre is.
Remember it's all these details that make all the difference.
There is another detail, that's worth a moments thought as well.
Had you ears tested lately?? It's worth it.
You would be surprised how many people can't hear much over 8kHz.
What do they do when they are mixing?
They compensate the over 8kHz range to suit their own inability to hear those frequencies.
Normal to them, sizzling high frequencies to me. (otherwise known as a hot mix)
Get some liquid stuff from the chemist to clear the wax out of your ears at least 2 days before your recording session.
If you do get your ears tested for sound you can ask for a chart that will tell you what if any deficiencies you have in your hearing.
For example, Let's say you had a 10db loss at 10kHz then you could wind up the 10kHz band until it was sweet to you. Drop it by 10db and it should be perfect for someone with full hearing. It is not unusual for musos to lose some of the top end in their hearing. Same goes for people who wear headphones all day amd dare I say it prolonged, loud, ipod use.
One other tip for mixing .... Use subtractive methods rather than additive.
Try to bring frequencies and volumes down rather than bringing surrounding levels up to enhance. Solo the track, Set the frequencies first, then adjust the volume in the mix. Same goes for effects like Reverb etc. Remember ... if you add Reverb on one instrument channel and then put Reverb across the entire output the effect will be additive to the instrument channel Reverb. In most cases, less is more effective.
Mastering.
Mixing and Mastering sort of go hand in hand.
You can make a bad mix sound good with Mastering, but how much better it would have been, if the mix was good in the first place.
Many many moons ago, I hustled and hustled to go to the record company that was doing the mastering for a vynyl record we had just recorded.
The record company said OK, on the basis that I watch and say nothing.
Just let the guy get on with his job.
Me being me, my curiosity could stop me from asking questions about what he was doing. Fortunately for me the guy was more than happy to ramble on, as he went about mastering the tracks.
Fact is, nothing much has changed except for some fancy new gear that we could only dream of in those days.
Then along came the loudness maximizer.
Before I get into Loudness Maximization, let me go back a little.
Before you consider the overall Mastering of the track, first get your Mix sounding as close to perfect as you can. (it will never be perfect, so all I do is ask myself if I can live with it the way it is.)
In Mastering we are no longer considering things like the volume of the guitar, or vocals, or any other instrument.
That's part of the mixing process and if something is not right, go back and fix it best you can, before starting the mastering process.
In Mastering we are no longer listening to the song, we are listening to the overall frequency ranges and how they are responding.
It's not so much that the bass guitar is too loud anymore, it's .. Are the overall bass frequencies too loud/soft, or maybe there are other frequencies that need a push or a shove up or down.
We are aiming for a flat response across all the frequencies.
This gives us the best chance to hear each instrument clearly and also gives us the greatest dynamic range in the overall sound quality.
There are lots of different things you can do with all sorts of enhancement equipment.
They are wonderful gizmos if you know how to use them, but I still somehow prefer to use the least number of gizmos as possible, perhaps more so on an acoustic recording.
Syths require a different mindset and so I tend to just experiment with sounds rather than go for transparency.
Meanwhile, back at our acoustic / electric mix of 2 acoustic guitars, electric bass acoustic drums and vocals, we should now have a mix that is sitting just under 0db and none of the signals are going through zero.
You can achieve amazing results in a home studio with just a few mastering tools.
I use Cubase with a mastering plug in called iOzone.
I have already done my best to get a great dynamic range in my mix, but now I go mess with the EQ compression. You really shouldn't have to compress or expand any of the frequencies by more than shall we say mildly, if you've done a good mix. Once again we don't want to hear the electronics working hard, if at all.
I gave a mix of mine to someone to master for me, cause he had this whizz bang gear. I was curious to see what sort of result he could get.
When I got the mix back, I could plainly hear the quality of his equipment working on all the frequencies at once, but I couldn't listen to it for long.
It was drenched in electronics.
If he'd used it lightly, I can imagine it would have really given the whole track a lift, but because I could hear the electronics working overtime, it totally destroyed my mix. Gently does it, is the order of the day with frequency compression and expansion.
Once we are happy with the overall frequency balance we go for Loudness Maximization.
I set the limiter to -0.1db.
I then turn up the gain by no more than 5db.
This gives you around the best compromise between loudness and loss of dynamic range. You do lose a small amount of dynamic range, but it is negligible.
More than 5db will certainly make it louder, but in my opinion you start to hear those losses the louder you make it.
You can even set it 20 db louder than zero, but it sounds terrible to me.
Quite often I get a CD that has been Mastered properly and see if my mix is around the same volume.
It should be. If not try to match it, so long as you don't have to push it much further than 5db.
You are just polishing up the overall sound of the mix.
Getting the best possible frequency range and volume out of a track, is what Mastering is all about. Simple as that.
Exporting your Final Master
The next stage is exporting the Final Mastered file into various formats, ready for CD manufacture, mp3's for the web and a host of other mediums.
CD Manufactured tracks play back at a rate of 16bit / 44.1kHz. and at Red Book Standard. This is the International Audio Standard for CD's.
It is best to discuss the output requirements with your CD manufacturer, before exporting the files to them.
We can burn all sorts of file standards on our computers, but they need to be the correct type of files for manufacturing, or else the CD won't play properly in a standard consumer CD player.
The track is coded with an industry standard code and the CD player has built in Industry standard file decoders.
Codec is short for COde and DECode. The file code type is built into the file. The decode file is built into the CD player.
This is why some of the older model CD players won't play MP3 files you may burn to disc, off your computer. No built in decoder for MP3.
The MP3 codec wasn't around when they CD player was built although I notice Mp3 CD players seem to be more common in car builds these days.
Same goes for .WAV files although most CD manufacturers these days can convert these and many other types of files for you. You can also have trouble playing CD-RW discs in various CD players for similar reasons.
These days they include many more of the popular file type decoders when they manufacture the players.
Mac users can breath easy most of the time as they already use .aiff files as a standard for audio.
(aiff stands for Audio Interchange Format File) Even so, you can still have an aif file that does not comply with what is needed. CD-RW discs being one of them.
MP3 files are very popular these days, because you can achieve all sorts of standards with it and the file sizes are much smaller than a lot of other file types.
They are typically used for the internet, ipods, mobile phones etc., because of the small file size.
There is another advantage to Mp3's in that you can playback in all sorts of sample rates whereas CD's are only 16 bit / 44.1KHz.
Higher sample rates mean better quality audio and most professional studios use at least 24bit / 96kHz sample rate to record with. (One day this will probably become the standard for consumer gear. )
They then use a dither down program like UV22 or similar, to convert back to CD standard 16/44.1. This does make for a finer recording, but you can still get very acceptable results recording at 16/44.1.
After all that's what we hear all day long on the radio and TV.
Everything is going HD these days, so I don't think it will be long before the audio standard is lifted to go with it. TV stations already use higher standards and Radio can't be far behind as well as higher standard CD's. The only thing in the way is the millions of CD players that are still at 16bit/44.1kHz.
Methinks planned obsolescence .... hmmmm.
Digital sound has come of age in terms of quality.
Many moons ago you could have said that tape had a higher resolution than digital, but I think that they have gone way past tape resolution now, so we can't really make that claim any more.
It is similar to photographic resolution of 2700 dots per inch on film resolution.
Computers can easily exceed 2700 dots per inch these days and the printers are a far cry from the old dot matrix printer I used to own.
What to make of all this?
CD manufacture ........ 16bit / 44.1kHz (talk to the manufacturer)
Mp3's ..... for a CD to play in a Cd player with the Mp3 Codec. 16/24bit (bit rate), 8k mono to 192k stereo (sample rate). It will play them all but I'll stick with 16/44 stereo at the moment.
MP3's for the web are a slightly different story and I still provide small files at the best quality I can get. I do this because even a small file of only 2 Meg can take up to 10 minutes to download for the guy/gal on dial up. Although the situation is changing the majority of people using the net, are still on dial up. (please remember that Mr Webmaster)
This way I try to provide 2 files. A really small file for dial up, or a higher resolution file for those who want better quality.
The absolute smallest I've been able to go in mp3 file size without losing to much audio quality is at 96k/bits/44.1kHz stereo with a sample rate of 32kHz.
But of course, if you want to output a much higher resolution with sound you can do so, without the same file growing to much more than 10 Meg.
Your original stereo master file size, may well have been up around 40 or 50 Meg.
I have had fairly good results with the above rates for ipod and mobiles as well.
Some audio burning software (like Toast or Jam) have built in converters that automatically convert your file appropriately before the burn.
Other programs like WaveLab are propose built for Mastering.
I hope this information has been useful to you. If you have further questions give me a yell:
This e-mail address is being protected from spambots. You need JavaScript enabled to view it
Cheers Pete.
|
