Audio Instruction

Mixing, EQ and Processing Questions

November 9, 2006 by Aaron

audio, audio questions, eq, equalization, mixing, processing, recording

I believe this is the last of the questions from years back that I have left…again I don’t find anything I said blatently dumb, but it’s interesting to see this old point of view…

Questions

I am a single acoustic act. Guitar and vocals. I use a Takamine FD360SC which I play through a Trace Elliot TA100R. For vocals I use a Shure Beta 58. Both are connected to a Mackie 1402VLZ. The Beta via channel 1 XLR and the Trace via a DI socket (pre or post eq). The Trace has a 5 band graphic eq, which I am currently leaving flat, it also has several onboard effects such as reverb and delay which I am not using. Both are eq’d at the board via a 3 band parametric eq. It never seems to me as though I am getting the correct mix! I generally boost the mid and the high on the vocal while leaving the low flat. I then boost the mid on the guitar and leave the high and low flat. Any golden rules ? Hopefully you will bear with me for a couple more questions, if not…I appreciate your time on the first. The next questions deal with effects. I have used reverb for years (Alesis Midiverb which is a 1/2 rack unit which I generally leave set to factory preset 21 “Medium/Warm” 1.4 sec delay). I am now using an Alesis Nonoverb for delay (because I don’t like ANY of the reverbs). Input and output set 3/4 of the way up and mix set 1/2. I don’t feel as though I am getting the most out of these effects. Any suggestions ? (Please don’t say “buy a Lexicon!”). Finally, I am considering adding the following: Compressor Rack mount Eq Chorus Noise gate Is there any particular order in which I should daisy chain them ? I want to run two loops one for guitar and one for vocals. Which effects should I dedicate to vocals and which to guitar ? My plan was: Vocals: Reverb+Delay+Eq+Noise gate Guitar: Compressor+Reverb+Delay+req+Noise gate

Answers

the golden rule is really really tweak a LOT until it sounds right 🙂 …vocals and guitars can clash, so you might try eqing them in opposite directions, like maybe CUT the mid on the guitar a little…usually to start with on vocals I cut the high a little, cut the low a little and boost the mid a little. Sometimes I totally cut the high and low and boost the mid all the way. I call that telephoning, although with that three band eq, you can’t really make the actual telephone effect exactly. but as a matter of fact, that extreme eq on the mackie sounds surprisingly normal on some mixes – unfortunately I doubt an accoustic guitar/vox mix would be one of those mixes 🙂 Sounds like you need to take a little more time with the mixes, and tweak everything in every direction just a little more. Also, the mixes might actually be pretty good, but your ears are tired – get 2nd opinions – also, if you’re going for a sound that sounds like you’re used to hearing on records, you’re probably missing some ingredients (like reverb and compression)
Don’t buy a lexicon, I’m really against buying the biggest baddest stuff without first pushing what you’ve got to the total limit. Well first off, learn the ins and outs of editing the effects (again the golden rule is play more) instead of just using presets….uhm, use them in layers, use them in extremes, use them very subtly – basically just spend more time running the gamut and experimenting – uhm…I would set the mix all the way wet, then feed the reverb returns to a channel on the board and mix the effect with the dry signal that way (and also, you can use the eq on the channel if you need to on the effect) – also if you’ve got a natural type sound (ie: guitar and voice) maybe you don’t want to use electronic reverbs so much – maybe you should find a great hall and record in there (I know I know, good luck on finding a hall 🙂 ) – then turn around and use your reverbs for some other kind of project – mostly though, just spend more time I think – every mix you do, go through every preset and see what’s the best, then if something isn’t PERFECT, start editing and tweaking – try everything under the sun

well some people other than me would probably know better and maybe tell you different, but here’s how I would chain them, assuming that this is to run an entire mix through: chorus, compressor, eq, noise gate although I wouldn’t run an entire mix through a chorus, and I wouldn’t neccesarily keep the setup the same for all projects – that’s another part of my little golden rule about playing – you should see how much repatching I do (I wish I had a magic digital auto-routing super patchbay 🙂 )

I would add compressor to vocals and maybe take it off guitar. you almost always need compressor for vox and I’m not sure I’ve ever compressed accoustic guitar. like I said though, for the effects, I wouldn’t have them in an insert loop (that is the delay and reverb) – what I do when in a studio where I can do this is, I have compression on the vocals on an insert loop, and effects like delay and reverb triggered with aux sends and returned to an actual channel on the board.

—
So yeah there’s those – if you want more, just email me a question at aarontrumm @ nquit . com and I’ll give you my opinion, which is mostly what it it is when you talk about audio and art. 🙂

Meanwhile, you can still grab free tracks over at my music download site. 🙂

— Aaron

Graphic EQ or Parametric?

November 8, 2006 by Aaron

audio, eq, equalizers, graphic eq, parametric eq, recording

This is another one of the questions that came through 15 years ago that I didn’t lose – I definitely have a better, more sophisticated view now, but I don’t see anything blatently wrong here 😉

Question

…I have been setting up a decent home studio over time now, and would like to pickup a rack mountable EQ for use at mixdown and to cut frequencies on certain tracks, especially guitar… Would I be better off getting a 15 or 31 band graphic, or will something like the ART dual channel 4 band parametric tube preamp be enough? Will the tube quality outweigh the lesser band control in terms of sound quality? Will 4 bands be enough?

Answer

First off, I very very rarely mixdown with eq on the whole mix, but it can be a useful technique.
Well, ok. Obviously 31 band graphic is probably better than 15 graphic, and probably more expensive. Parametric eq tends to give you a lot more control (parametric meaning you can change the frequencies you’re working with). It’s kind of a trade off in my opinion. Parametric is better, really. But you tend to get less bands with parametric, which means less frequencies to be effecting at once. I guess parametric eq is more expensive to make or something. It would be nice to have 31 bands of parametric! 🙂 As far as the tubes, I wouldn’t take that into consideration for gear like this until the end. I use tubes, and I like to, but frankly, I defy you to hear them working 🙂 I don’t know that I see the value of using tubes in an eq, I tend to just use tube preamps to track through and run mixes through. That having been said there are tube-preamp units that have great eq sections, and I also might lean toward the Art, they make some cool stuff. As far as graphic eq’s, Rain pretty much makes the best, I think. As far as four bands being enough, think about your applications, and your possible applications. Will you need to be effecting more than four bands all at once? Like I said, with parametric, you’ve got all the control for dialing in what frequencies you’re messing with, so that might be the ticket if you don’t need to effect more than four at once. As far as tube quality outweighing having less bands, no. I think you should never put tubes before functionality. If you need to get something without tubes so it’ll do what you need, get that, save up, and get a dual tube preamp to run the whole thing through. Also, by the way, you should know that quite a few engineers think the tube thing is crap (I don’t, like I said, i use ’em and love to). So it’s not necessarily a given that tubes are even better. And by the way, tubes burn out eventually.

—
And that was that question. Hmm – let me know if you have questions too – hopefully I have more and better info to answer with! aarontrumm @ nquit . com

And – get you some free music, while you’re at it, so you can see what I do with this “knowledge”.

A vocals mixing question

November 8, 2006 by Aaron

audio, mixing, recording, vocal mixing, vocals

Many (like 20) years ago, for a short time, I opened up to questions on my site and people would email me stuff…I have just a couple of those that haven’t been lost and here’s one…

Question

…you know how some CD’s are mixed so that the singer’s voice sounds “farther away” from your ears with the headphones on? and some sound very up close and in your face? well basically, my voice sounded way to up close and i was wondering how to get it to sort of blend in better with the rest of the mix. this may be a silly question but like i said, all of my recording so far has been instrumental so i havent had to worry about it so much, and of course i am self concious about my voice right now. i dont think its simply the recording level, because that just makes the sound quiet, but doesnt change the sound in the audio field. so is it a question of mic type or placement? or of how it is mixed? or some other professional equipment effect that i dont know about? i would love to hear your thoughts on this!

Answer

Actually this can be a tricky thing for me too. You’d be surprised that sometimes it really is level…try lowering the level on the vocals. Also yes mic placement is important. You definitely get a different sound if you stand farther away. 1,2,3 feet and even farther are all different distances I’ve used. Also, reverb can be what you’re looking for. If you try adding a little more reverb and and less of the dry vox in the mix, it tends to sound “farther” away. Sometimes maybe even eq will help you here (with eq on vox I almost ALWAYS roll a little low end off, give it a little extra hi-mids and a tiny bit more high end). Sometimes I try making the vox a little thinner. Sometimes I try the opposite of what I’m thinking, and that sometimes helps. Try doubling up the vocals. (doubling up sometimes cuts through my self consciousness too…don’t do it unless appropriate though)…tweak a lot, try all the things you can think of…also compression…technically, compression will do the opposite of what you’re asking, but compressing the vocal is so common, that to your ear, it’s gonna’ sound more like a finished recording and thus less like naked vocals…at least that was true in my mind on a number of occasions.. Watch out for your self consciousness though. You could be getting mixes that are perfect, but sound WAY out front to you, because you’re scared (let other people hear the mixes…I try to have one or two very select people in the room with me mixing, or at least have them listen later). I always take what the vocalist tells me about the mix with a grain of salt. They (vocalists) almost always bury their vox. You should definitely try things out…try way too much reverb and not enough dry signal, and see what that does, etc…on every mix I’ll try one extreme and then another, and everything in between, just to see what the stuff REALLY needs to sound like. First thing’s first though, get over that self consciousness! 🙂 If you’re a guitarist singing some vocals, become a vocalist too. Take lessons, if that’ll make you feel more confident (it probably will, since it’ll probably make you better). If you’ve already taken lessons, take some more. I don’t know how good a vocalist I am, but on my recordings, I am confident and do whatever I want (even when I shouldn’t probably 🙂 ); one of the reasons is I took a singing class and it scared me to death and that helped; also I ALWAYS sing in my truck driving around (break the speakers in your vehicle, you’ll see *grin*) Make it a point to sing in front of people. Not just at concerts, but always. I don’t know how much of this stuff you already do, but it helps your MIXES, trust me. Also, record your voice a LOT. every DAY make three or four acapella vocal recordings and maybe even some spoken word and listen to it a lot (they don’t have to be good recordings, in fact vary it a lot, listen to your voice in lots of different contexts). You know, get real used to hearing your voice on tape, ’cause it sounds different. I’ve recorded hundreds and hundreds of songs (way way way more than I’ve released, I mean some were on a freakin’ boom box, but hey, I got used to hearing my voice on tape)…all this so you won’t bury your vocals in the mix. 🙂 Now just watch out you don’t get cocky and crank ’em up too much!

An added ps: Don’t mix with the monitors loud. Vocals tend to get too hot in the mix this way. Also you burn your ears out, of course.

It’s quite alright with me, btw, if you want to send questions like this – I’ll post answers – I’m at aarontrumm @ nquit . com 🙂

Meanwhile, grab some free stuff over at aarontrumm.com 🙂

Basics Of Analog Audio Mixers

November 6, 2006 by Aaron

analog mixers, audio, basics, signal path

From a 1994 article for the University of New Mexico’s Intro to Electronic Music class:

Signal Flow:

Mixing boards contain multiple “channels”, each of which has separate input, and is controlled separately. These channels are then assigned to either a “buss” or a master output. Every channel assigned to a given buss or master is mixed together and output in it’s blended fashion through the output jacks corresponding to each buss or master output.
Channels on all mixers (with rare or no exception) have either an On switch or a Mute switch, or both. On switches simply turn that channel on. If no sound is being output of a channel you think should be outputting, the first thing to check is the On or Mute switch. Mute switches do the opposite. They turn the channel off. If an On switch is not engaged, no sound will be outputted. If a Mute switch is not engaged, sound will be outputted.
Signal fed to a given channel’s input procedes first to it’s trim section, which allows the input level (volume) to be adjusted; then to it’s EQ (equalization) section, then to the Pan-Pot, which simply adjusts to which side of the stereo spectrum (see stereo concepts) the channel is placed, then to that channel’s “fader” (or gain knob, depending on the board). A fader (or gain knob) is simply a volume control. It controls how much overall level is being output from that channel. This signal flow is represented somewhat by the physical layout of each channel. Input jacks are at the back, Trim knobs are next, then EQ section, then Pan-Pot, then Fader.
After flowing through Trim, EQ, Pan-Pot and Fader, signal is fed to all busses or master outputs that a given channel is assigned to. For example, if a channel is assigned to buss 1 (usually by pressing a button above or near the fader), then signal will be fed to buss 1, buss 1 will output that signal. Busses are simply outputs. All busses have a volume control of their own. That volume controls the overall level of all the sound being fed to that buss (meaning this sound is usually a mix of many channels).
Their are two basic configurations for output on mixing boards:
$nbsp$nbsp$nbsp 1) All busses are automatically fed to the mixer’s outputs (which usually consist of at least two sets of outputs: one pair for monitoring, and one pair to go directly to tape. “Monitor Out” and “Tape Out”).
$nbsp$nbsp$nbsp 2) The mixer has a “master left-right”, (which is simply the main master control buss) and individual busses can be assigned to that output. In this case busses are not automatically outputted to the master output jacks. In either situation, most boards have another set of output jacks, which correspond to each individual buss. (For example, if their are four busses, there are four “buss output” jacks, in addition to the master output pairs.)

Common Channel Characteristics:

INPUT – Channels on most boards consist of at least two input jacks (usually at the back): 1) Line input. This is usually an RCA (or “phono”) jack or Quarter-Inch (large “headphone”) jack (although really high-end professional equipment rarely uses anything except XLR connectors, even for line input…most good boards have a choice). This input accepts line-level input, that is, input that does not need to be boosted before proceding to the channel controls. Examples include cassette decks, CD players, Synths, etc. This input is usually labeled “line in” or “tape in”, or similar (sometimes there are both “line in” and “tape in” jacks. They are essentially the same).
2) Mic input. This is usually an XLR (three-pronged, or “cannon” input) connector, but it can be other types. This input accepts mic level input, that is level that DOES require boosting before it will be at an acceptable level before proceding. Examples include, of course, microphones, and assumably some other devices that ouput at mic level. This boosting is done by the Pre-Amp section of the channel. This input is usually labeled “mic” or “mic input” or simaler.
DIRECT OUT – Channels usually consist of a “Direct Out”. This is (as far as I know) always a line-level ouput. Direct out jacks consist simply of the signal that is being output of the given channel. Direct out jacks are not effected by other channels, busses, or master output levels. Plugging into a direct out does not effect the channel’s output into busses or master outputs, and does not interfere with the flow of that channel into a mix.
INSERT or ACCESS – Channels usually consist of an “insert”, or “access”. This can be accomplished in one of two ways: Either with one jack, that outputs the channel’s output, then returns it, within the same jack, or with one jack that consists of the channel’s output and another that returns it. Inserts interupt the signal, and thus allow you to “insert” another device into that path. (An example is a person may use the insert to take the output of the channel to a compressor or effects processor, then return the modified sound to the channel…the modified sound is then what is fed to any busses or master outputs.) Plugging into an insert DOES effect the channel’s output to any busses or master outputs. It does NOT effect other individual channels.
INPUT SELECTOR – Channels must have an input selector switch if they have more than one input jack. Typically this switch chooses either the mic or line input. IE: If the switch is switched to “mic”, then input for the channel will be accepted from the mic jack, and not from the line jack, and vise-versa.

Thanks for reading! This was quite some time ago but still applies. If you want some current flavor, I’ve got free downloads available!

— Aaron

Basics of Equalization

November 5, 2006 by Aaron

aaron j. trumm, audio, eq, equalization, music, nquit music

From a 1994 article for the University of New Mexico’s Intro to Electronic Music class:

Equalization is a process of enhancement of audio signal. It is one type of signal processor, because it processes an audio signal.

Equalization (EQ for short) does basically one of two things: Either it boosts or cuts a given frequency or range of frequencies by a chosen level.

Outboard EQ processors come in two basic flavors: Parametric and Graphic.
Parametric EQ uses a system that allows a user to select which frequencies they’ll be working with. Thus, it is very flexible. Fully Parametric EQ has completely selectable frequencies and bandwidth. Some EQ’s are semi-parametric, meaning they have some, but not complete, flexibility. Most mixing boards use some sort of semi-parametric EQ (although many use a system that is not parametric at all, but cannot be called graphic).
Graphic EQ has no method for changing any of the frequencies that are effected, but still can be quite flexible, if it has a wide variety of bands. Graphic EQ derives its name from the fact that its sliders are configured in such a way that when a user look at them, they graphically represent the configuration being used.

Mixing boards use an EQ system that is not usually truely parametric, but it is not a graphic system (since, of course, it does not use sliders that graphically show the configuration). Many boards use a fixed band EQ in which each knob effects multiple frequencies, such as high, mid and low (unlike graphic EQ in which each slider boosts or cuts only one frequency). Many boards use a semi-parametric system in which some sections of the EQ have a knob which will choose a frequency, which is then usually accompanied by a bandwidth selector, which chooses how wide an area of frequencies will be effected. (This is nearly true parametric EQ, but usually only one or two frequency ranges provide this option, so we don’t normally call this a truely parametric EQ system.) Some boards have sections of their EQ that are semi-parametric, because the frequency effected can be selected, but not the bandwidth.

EQ is used for many purposes, including enhancing an audio track, making a track more audible in a mix, fixing noise problems, helping audio tracks to sound more real or artistic effects such as “telephoning” (my word, meaning to cut low and some high frequencies, making a track sound as if from a telephone).

The Art Of Equalization

November 4, 2006 by Aaron

aaron trumm, art, eq, equalization, music, nquit music

From a 1994 article for University of New Mexico, and published in Cubase Newsletter:

EQ can be used in a variety of situations, from live sound to recording to tape to mixing down. Mainly, it should be used to enhance signals that have some problem. The golden rule of EQ is less is more. If something seems fine without it, I avoid EQing it at all. Then, if I do use it, I try to remain subtle. My personal golden rule is nearly never EQ signals going to tape (as in a multitracking situation). I always try to get the original sound on tape, then I can mess with it later. Putting EQ (or any other effect) on tape usually just leads to trouble. The other rule (the silver rule 🙂 ) is cutting is almost always better than boosting, especially when fixing problems. For example if a guitar sounds too thin, first try cutting high frequencies and boosting the gain a bit, instead of boosting the lows. The more clutter you can remove from a mix, the better. A better example is I very often cut a bit of high away from hats. Another example is, many times you may not hear something well in a mix…You might try cutting some frequencies in a different track that seems to be interfering, rather than boosting in the track you want to bring out. With these basic rules in mind, I’ll tell you my rules when I enter a mixdown session:

Rule Of Opposites: Usually, tracks with high sounds, (a high guitar, hats) need cutting in high frequencies and boosting in lower, and vice-versa. This is really only a starting guide, not a rule. Also, sounds that interfere with eachother can be separated in a mix by EQing them in opposite directions.
Bass usually needs a boost in the mid range somewhere and sometimes the high. This way it can cut through and be heard on smaller speakers.
Kick drums usually need that same mid and/or high boost on a subtle level so they too can cut through on smaller speakers. For hip-hop, kick needs a low end boost, but NOT TOO MUCH.
Snare drums always sound warmer with a boost in the low-mid range and some cut of the highs. An annoying CRACK can be softened with this high cut. Sometimes I boost the lows in snares to make them even fatter. But it really depends on the snare sound. The rule of opposites usually applies here. Snare sounds that were thin to begin with I usually warm up a bit, and heafty snare sounds I might thin out a bit.
Hats almost never need any EQ if they’re recorded clean. Usually an EQing for my hat tracks is to cut highs to get rid of an annoying hiss.
Guitars are simaler to snares for me. A thin original guitar might need boosting in mids and lows (depending on what the desired sound is, and what else is present in the mix) or a heafty guitar might need to be thinned out a little by cutting lows and low-mids.
Vocals usually like to have a boost in the mids or high-mids, but it depends on the voice. Vocals nearly always get lost amongst guitars…a good way to deal with this is the rule of opposites. Boost mids in the vocals and cut them in the guitar, or something similar. Vocals can also have annoying hiss or sibilance, and sometimes cutting high frequencies can help that.
Strings, and more specifically good string patches from a synth, usually need little EQ. If they are merely a support player, I may thin them out a tiny bit, or if they are meant to be present, I may thicken them in the mids a little (or sometimes the opposite…this stuff is highly subjective). But they usually work well left alone. Really clean piano or keyboard synth patches are the same way.
I like to leave reverb returns alone, but if the reverb becomes annoying and noisy, cutting some high can soften it up a bit…same with strings.
Extreme EQ setting create sounds of their own. Experiment. But for a non-novel track, be subtle.
AC hum from a track can almost always be fixed by cutting 60 Hz all the way off. (Sometimes this can take away from bass or kick sounds, but I believe that most frequencies audible in a song are above 60 Hz).
Play with EQ settings thoroughly to find appropriate settings.
I don’t mix horns too often, but when I do, I like to leave them alone. Clean horn tracks usually seem fine to me.
NEVER EVER EVER force yourself to EQ a track that sounds fine, just because you think you should use the full capabilities of the studio. NEVER NEVER NEVER!

Mixing, EQ and Processing Questions

Questions

Answers

Graphic EQ or Parametric?

Question

Answer

A vocals mixing question

Question

Answer

Basics Of Analog Audio Mixers

Basics of Equalization

The Art Of Equalization

Free Download and Newsletter