A thread for asking the things you should know by now but don't

[Savernake]

Originally Posted by MarkRB;1924305.

This is also known as ducking and radio stations use it all the time for the music and the DJ's voice.

Just so people know, this is an excellent example when explaining sidechaining, but as am actual use its not so good. No way as effective as riding a fader. BBC Radio certainly dot sidechain the voice like this.

As for the left and right thing, try thinking of it like this: what kind of listener do you have? If its musical '******y' (see some prog!) then map it to a drummer perspective. If your audience is more likely a listener than a drummer/performer then map it the other way. Of course, there are no rules.

[Francis Vaughan]

Originally Posted by inthemix

I mainly record 1 track at a time through an apogee AD then into my computer card.. SPDIF in

I have a choice in my computer to select the clock:

Internal
SPDIF or
Wordclock

Would there be a difference in the SPDIF and Wordclock? Is one better than the other?

Right now - no - there is no difference. This is because the digital input into the computer is not implicitly clocked. By which I mean that the system knows the sample rate, and that is held as meta data that describes the input data stream. There is no recovery of the precise moment of each sample from the data stream.

It is when you need to ensure that clocks are distributed to devices that need to be both kept in sync and maintain good clock accuracy that the different clock distributions matter. Sending a clock down S/PDIF versus a word clock cable is done in different ways. The word clock simply contains a digital pulse on each sample instant, whereas the S/PDIF signal can occur in two ways. Either the S/PDIF sends digital zero, in which case clock recovery can proceed reasonably cleanly, or in many cases people attempt to send audio data, as well as recover the clock, over the same cable. This is subject to degradation due to flaws in the way in which the link is specified. The audio data degrades the clock recovery. Very nasty audible degradation potentially.

In general you would use word clock cables to reticulate clocks to any device that needs it.

Where you current setup might break is if you wanted to add further devices. Then you would want to think very carefully about where the master clock comes from (usually the ADC, it there is only one) and ensure that everything else gets as clean a word clock as possible if it needs to maintain good sample timing (for instance a separate DAC used for monitoring.)

[SamSites]

A thing I always wondered about is ...

say you have an AD/DA Convertor. (8in, 8out) How do you connect it to your computer? You have to have some sort of 'firewire' convertor between? I know nothing about more expensive convertors. (Apogee, Lynx, ...)

A small question about phase with mics. I know more or less how to bring my mics into phase, but does the positioning have to be to the mm precise? OR do other mics have les volnurability to be out of phase.

Probably stupid questions ... but whatever.

samsites

[Francis Vaughan]

Originally Posted by unsung

What is a discrete amplifier?

Simply the antonym of "integrated".

So in a audio gear, not just amplifiers, but lots of gear, if something is made out of a pile of individual (i.e. "discrete") components - like resistors, capacitors, transistors etc., it is a discrete device, whereas if it contains devices that are fabricated as a single "integrated" circuit - all one one die of silicon, it contains integrated circuits and ceases to be discrete.

There is of course an element of snobbishness attached to an all discrete design. Some value too, but like all these things - the devil is in the detail.

The most common example of the distinction is in the use of op-amps. You get the expensive and revered discrete op-amps (990 etc.) in competition with the little eight legged device, which can vary from the very cheap (TL07x) the cheap (5532) to the relatively expensive (say OPA627). One important difference between most integrated op-amps and the discrete versions is that the discrete versions can run with quite high voltage power supplies (+/- 35 volts) whereas very few integrated op-amps can manage more than +/- 15 volts. On the other hand it is possible to create superbly matched input stages in an integrated device - something that is much harder in a discrete one.

[Kimakaz]

What do people mean when they say "comping vocals" for example?

I have a clue but need confirmation ?

[Francis Vaughan]

Originally Posted by Chapiro

Whats an opto compressor and how it differs from a vari mu or other types of compressor?

To expand on an earlier post.

Any compressor seeks to vary the gain in a dynamic manner. So you need to be able to control the gain electronically. The input to the gain control is derived from either detecting the level of the signal or from your side chain. A compressor is set up to reduce the gain in response to high input signals (whereas an expander does the opposite.) The exact nature of the control system (feedback, feed-forward, time constants) determines the majority of the sound of the compressor.

But the actual variable gain stage is an interesting beast. The two "vintage" ways of doing it have been opto and vari-mu.

With opto you have a light dependant resistor. (LDR) Place the LDR in the signal path and use the control signal to modulate the intensity of a light source. Instant compressor. (More complex possibilities as well, but the principle remains.) Neither the LDR nor a light bulb are either linear or have infinitely fast response. So you get a certain aural signature. You can use other light sources for faster response, and different transfer curves. Neon, LED, all sorts.

Vari mu uses the characteristics of a particular type of vacuum tube. If you make an amplifier tube with a varying pitch of grid wires it turns out that the amplification the tube provides can be varied. The amplification of a tube circuit is derived from its transconductance (one of the parameters that describes a tube) which is denoted by the Greek letter mu. By varying the bias voltage on the grid of a variable mu tube you can vary the amplification you get. Variable mu tubes were designed initially to allow automatic gain control in radios. Funnily enough, an AGC is nothing more than a compressor.

Variable mu tubes are not especially linear (they can't be - the variable gain is welded into the transfer function) so they have a certain aural signature as well. They can respond faster than an opto circuit, but are less clean.

Later circuit designs can include voltage controlled amplifiers (which in a way is just a newer fancy sort of transistorised vari mu - but they don't have the non-linearities) or chopper circuits that vary the gain with fast pulse width modulation of internal elements. Lots of other possibilities.

[Francis Vaughan]

Originally Posted by Kimakaz

What do people mean when they say "comping vocals" for example?

I have a clue but need confirmation ?

A word I have never grown to like. "Comping" is simply a corruption of "accompanying".

The verb "comp" similarly is a corruption of "accompany."

The noun "comp" on the other hand is a corruption of "complementary", as in free.

[nativeaudio]

Originally Posted by Kimakaz

What do people mean when they say "comping vocals" for example?

In a recording studio/music software, it normally means compiling a 'master' vocal track based on the best parts of the existing takes. There are different ways of dealing with this, here's one of them.

[Kimakaz]

Thanks nativeaudio! So i've been comping tracks all this time, and never knew...

[Francis Vaughan]

Ah, well that's me showing a musical background. Another meaning.

[TheRealRoach]

Originally Posted by bojan_1

Here´s one, that´s been bothering me:

When we see a waveform (e.g. in logic), what are we actually seeing?

The sound is a complicated superposition of many frequencies at every moment, so is this the sum? If so, why do we have "null points" quite frequently, as the odds, that the sum is zero are much lower?

and last but not least: in logic there is only one waveform representing left and right in a stereo file, so what does that suppose to mean?

enlighten me

bojan

This was partially explained but I'll elaborate. Logic represents stereo interleaved files on the arrange page as a 'mono-looking' waveform. What the cheese!? After doing a bit of experimenting I found that if there is ONLY signal in the left channel of a stereo interleaved file, then only the "top" part of the waveform shows any signal (as if someone just lopped off all of the "-" information of the waveform).

I don't know why Logic chooses this form of graphical stereo interleaved representation and not the standard Pro Tools/Nuendo/etc dual waveforms within one region. Another illogical design for Logic.

[FireMoon]

Comping tracks surely means compressing a bunch of takes into one take and the word comping is used so as not to confuse the term with compression?

[macc]

Originally Posted by FireMoon

Comping tracks surely means compressing a bunch of takes into one take and the word comping is used so as not to confuse the term with compression?

I've always known it to mean compiling, which is perfectly accurate as a description of the process 'Compressing the takes' is soooort of accurate, but not as correct as compiling

[tonymission]

- Why do I hear "bounce" so often in reference to printing audio? Because that's what the button in Logic says? :D

- What's up with these "clocks"? Big Ben etc ... I never understood that? I guess I don't really know what a clock is? I associate that word with improved timing but um... that's hardly applicable to the context I read it in. Why would I want a big ben with my motu for example? Besides "better quality" ... I wanna know why :D

thanks!

[Spud]

Originally Posted by Kimakaz

What do people mean when they say "comping vocals" for example?

I have a clue but need confirmation ?

Originally Posted by macc

I've always known it to mean compiling, which is perfectly accurate as a description of the process 'Compressing the takes' is soooort of accurate, but not as correct as compiling

Yeah...compiling sounds good. I have it my li'l haid that it derives from "composite"...as in a track that is "composed" of elements from various takes

[matt thomas]

a couple of answers and a question

every DAW that I know of represents the sound waves with a latitudinal wave, wheras sound in the air is a longitudinal wave. What this means is that a sound wave is not moving up and down like the picture in your daw, but rather it is a series of compressions and decompressions. The wave in your daw is a graph of these compressions over time.

which brings me to the first question which I don't think was adequatley answered. Why does a wave sometimes appear lop sided? The answer is there is no reason to suppose that a recorded wave should be evenly sided. This is because a compression is not necessarily followed by an equal decompression. If it were, then the whole process would repeat ad infinitum. ie. a compression would have to be followed by an equal decompression which would be followed by an equal compression which would etc..


and lastly, my question is: who uses expanders for anything other than gating?

narco

[matt thomas]

Originally Posted by tonymission

-
- What's up with these "clocks"? Big Ben etc ... I never understood that? I guess I don't really know what a clock is? I associate that word with improved timing but um... that's hardly applicable to the context I read it in. Why would I want a big ben with my motu for example? Besides "better quality" ... I wanna know why :D
!

a clock just tells digital devices when to spit out (or record) each sample (eg. 44100 times a second)

A good clock should do this more accurately than an inferior clock. More precise timing of samples means more accurate representation of sound.

Personally I have found any decent converter to have a decent clock built in.

narco

[Savernake]

xpandeds can be used to 'turn up' quiet parts of audio, without having to compress the louder signals to turn the whole thing up. An kind of backward compressor I guess. If that's not the proper use, try it anyway, sounds good to me sometimes - keeps dynamics of loud bits while getting the low bits more audible.

[eric maltz]

Originally Posted by tonymission

- Why do I hear "bounce" so often in reference to printing audio? Because that's what the button in Logic says? :D


thanks!

im pretty sure the term bouncing comes from multi track tape recorders. if your using a 4 track, youd "bounce" ( record) the three used tracks onto the empty 4th. thus creating 3 new empty tracks.

[Savernake]

Expanders, sorry... My iPhone makes me spell bad sometimes.

[eric maltz]

Originally Posted by Savernake

xpandeds can be used to 'turn up' quiet parts of audio, without having to compress the louder signals to turn the whole thing up. An kind of backward compressor I guess. If that's not the proper use, try it anyway, sounds good to me sometimes - keeps dynamics of loud bits while getting the low bits more audible.

yeah like reverse ducking, can be a pretty cool effect

[TheRealRoach]

Originally Posted by narco

a couple of answers and a question

every DAW that I know of represents the sound waves with a latitudinal wave, wheras sound in the air is a longitudinal wave. What this means is that a sound wave is not moving up and down like the picture in your daw, but rather it is a series of compressions and decompressions. The wave in your daw is a graph of these compressions over time.

Here's how i think of it:

When you watch the movement of a waveform in slow motion - alongside a speaker moving in and out - the outward movement of the speaker coinsides with the movement of the wave moving upward above the line (positive). As the wave then moves back down and underneath the line (negative), the speaker moves back into the cabinet.

If you counted the up and down cycles of a 1khz tone, you would count 1,000 for every second. If you slowed down video playback of a speaker playing a 1khz tone back, you would count 1,000 movements in and out of the same second, and -hopefully- in perfect synchronization with one another.

Obviously we don't go to the store and buy CDs of pure sinewaves. Sound is much more complicated than that. And speaker vibrations are reproducing thousands upon thousands of different frequencies all at once and the higher (faster oscillating) frequencies essentially ride on the backs of the slower/lower frequencies. You can see this very easily by zooming in on any waveform... especially on a monophonic source like a vocalist. The largest up/down trend that you will spot is the fundamental tone... the "note." And all of the other squiggles on that fundamental are various harmonics that make up the unique sound to that person's voice.

[andsonic]

Here I am going way out.

1) It is true that the 1176 is an opto compressor? (I'm having brain freeze)

2) Given that: Are the LA-3 & LA-4 opto processors?

-different subject- What in the world!?!

I print my mixes at 88.2, but most of my source multitrack material is 48. Last night I recorded band practice at 2496. The producer was awe struck. Then I set up a blind taste test between the 2496 mix master & the 1644 cd master. His jaw just dropped.

Now I want to listen at 96k in my car! We were wondering...

Why are we made to suffer mp3 when there is *clearly* a better format out there?!? (I know, I know...)

[PlugHead]

Originally Posted by bassistmarty

I did song today that sounded fine through my headphones but when the tracks are separated and I play back the vocal through my speakers my vocal sounds pitchy and flat...Do headphones give a false sound to my ears?

IMO,

Headphones can often skew one's perspective of pitch, esp, when monitoring at fairly loud levels (same issues with speakers at loud volume, tho significantly less than with cans). Closed-ear designs tend to lessen one's ability to accurately hear faulty pitch. Maybe try with open-eared phones, monitoring at a softer volume, or in a pinch, one ear out of the cans.

IME, one of the best methods in catching "pitchiness" (my hillbilly term!) is to listen to your monitoring off-axis - this is a tried and true method when avoiding correction tools like AutoTune, etc.

Not a "tech" explanation, but my experiences - anyone accurately know why this phenomenon occurs?

Absolutely awesome thread - will be my first rating at GS!!!

[nine99nine]

Originally Posted by andsonic

.
Why are we made to suffer mp3 when there is *clearly* a better format out there?!? (I know, I know...)

I have noticed that the general public wants a few things out of artist's music, quality songwriting with a large quantity of music. Basically they want to carry ALL of their FAVORITE tunes with them at any given time to listen to. Having recordings in 96/24 wouldnt allow them to carry all their songs with them, and possibly the price of the unit would go up because it would have to have 96k d/a. (imagine ipods more expensive than they are now, for even less storage...people would buy the competitor with more storage and mp3 quality). If you haven't heard of FLAC, it is a step in the right direction as far as offering lossless quality with small filesizes, but is still yet to be majorly supported by any large manufacturer/developer.

Here is my question to your question...can the general public hear the difference between a 320kbps mp3 and 96/24 lossless in their ipod earbuds with its what i believe to be under performing headphone amp? (mostly talking about iPods or mp3s in general)

[nine99nine]

Originally Posted by narco

a couple of answers and a question

and lastly, my question is: who uses expanders for anything other than gating?

narco

I believe mastering engineers use expanders a little more often than do recording/mixing engineers...but that is a speculation.

[loujudson]

Originally Posted by macc

I think - while you're quite correct in pointing out my error, you missed the general point of my post.

The point was that whether you mike from player or audience perspective, you can't sit in front of/behind all the instruments at once. That is, in your example with the piano, high to low, left to right. For a drum kit (assuming the drummer isn't producing ), the snare will be a little left, with the cymbals all round, left to right. So in effect you're sitting in front of both the piano AND the drum kit.

Sorry fopr miscommunicating - My example with the piano was a solo tune. And the drum one was for drums centered in stereo.

How I like to mix is the full stereo tracks, panned so that it is realistic - stereo piano from left to center, stereo drums from center to right, and so on. Only sometimes would I make everything stereo and wide, mostly only by artist request... or in some of my really old work! Would that I could go back and change it!

I didn't miss your point, only gave an incomplete response! Thanks for making me aware of it.

Lou

[loujudson]

Originally Posted by nine99nine

I believe mastering engineers use expanders a little more often than do recording/mixing engineers...but that is a speculation.

Expanding is wonderful for reducing room tone in spoken word, and reducing breaths in singing. Sometimes for making more dynamic performances when the player seems to always play loud, but one must be extremely careful to not make it obvious (especially to the player!)

In my live rig I have a couple of DBX 1066 expander/comps, and there are times when you need to lower hum or buzz in a bass, or use it as a subtle "gate" to clean up kick or tom drums sound. The 1066 is wonderful as the expander/gate has continuous control over threshold and ratio. Very versatile!

Lou

[andsonic]

Here is my question to your question...can the general public hear the difference between a 320kbps mp3 and 96/24 lossless in their ipod earbuds with its what i believe to be under performing headphone amp? (mostly talking about iPods or mp3s in general)

That's a good one. And that has been the argument against hi res. My answer is that I would notice a difference listening on earbuds.

I look at it this way. When I was a teenager I would record my fave tunes from the radio on to cassette to be able to rock out on my Walkman. I could enjoy my faves (for free!), but I had to deal with the inherant noise of a cheapo cassette and all the compression artifacts from that lovely fm signal. Now if I really dug the song, I'd go pick up a copy on LP. The very first thing I'd do is tranfer it to metal cassette on the first play. Those copies always rocked and, even on earbuds, were clearly superior to the "FM" copies when played on my Sears Walkman clone.
Does my analogy make sense.

And these days, for my personal use, I always encode mp3's at the highest possible quality. The "standard" setting always drives me nuts.

One last point to my absurdly long post: I think with iPod & hard drive capacity these days, one could hold 4-5 albums worth of music at 24/96. the 5 minute tune I mixed last night clocked in at 127 mbs. It's doable.

[Kadden Heart]

new question:

how the hell do i trim my automation in pro tools?

this sounds stupid, i've been rewriting all of my automation for a long time now because i simply can't figure out how to bring up the overall automation of a track.