difference between applying delay and reverb

[mikeshead]

I understand that the difference between delay and reverb is just a question of the fed back echo times. But does anyone know how to apply each and what is the different effect each has on the sound?

At the moment (though ignorance) I tend to apply reverb to acoustic guitar (with no delay) and delay with a little reverb to vocals.

It would be interesting to know also how delay and reverb affect each other.

Currently (on my br900) I tend to add most of the delay on each primary track and then add a little reverb once I have bounced the tracks down and they are ready to be mastered.

Good idea or not ?

 

[SouthSIDE Glen]

I understand that the difference between delay and reverb is just a question of the fed back echo times. But does anyone know how to apply each and what is the different effect each has on the sound?
...
Currently (on my br900) I tend to add most of the delay on each primary track and then add a little reverb once I have bounced the tracks down and they are ready to be mastered.

Good idea or not ?

Mike, an entire book - or at least a several page chapter - could be written to answer that first question; and several have been . I'd suggest spending an afternoon or evening at the coffee shop of your local Border's, grab a few books on studio technique from the music section, get a coffee or soda and just browse the chapters on reverb and delay.

You have two fundamental choices: use them to try and model/simulate the sound of an actual space, or use them to create special sonic effects that don't necessarily resemble anything natural, but sound great anyway.

One tip I'd like to get you started with, though, is the idea of setting delay and reverb decay times to work *with* the tempo of the song instead of against it. This usually means setting the times to some multiple of the tempo. For example if you are working at 120 beats per minute (BPM), setting the times to something like 120, 60 or 30 BPM often has a more pleasing effect than just some random setting. If you set delay to the same a the tempo of the song on a rhythm track or two, it is a nice way to add some "punch" to the rhythm.

As far as your last question, whether it's good or bad is up to your ears to decide.

That said, there are times when you may want to apply reverb to the whole mix like that, like in trying to make it sound like you are playing in a live space. There are other times, though, when applying reverb to individual tracks will be different but equally or more pleasurable. It's good for (in combination with other volume and EQ adjustments) making something sound more distant, for example; if you want to move an instrument back a bit, it might get more 'verb than the up-front stuff. Or you might use it on a vocal to help disguise the vocalist's imperfect voice (the ol' singing in the shower effect) without soaking the entire mix in the same verb. And so forth.

Read up; there's a lot of stuff out there that isn't suited for quick answers via forum posts.

G.

 

[Daisy]

Currently (on my br900) I tend to add most of the delay on each primary track and then add a little reverb once I have bounced the tracks down and they are ready to be mastered.

Good idea or not ?

I guess if you really like a lot of echo.

 

[carlosguardia]

Glen,
So I was recording an "electronic musician" ok well basically this dude has a Roland groovebox and Ableton Live and does many gigs and get paid really well for them too! He plays parties, weddings, you name it. I really dig his tunes and he seems to know quite a bit about theory/harmony, physics et al. BUT, when I was recording his stuff, and playing around with delay, he asked me NOT TO PUT THE DELAY EXACTLY AT 120 BPM (if the song was at that tempo) but instead at 119 or 121 because of PHASE issues :S

I didn't believe there would be any phase issues, but didn't want to disagree with the famous dude that just came in to my small studio... Plus, I played the "customer's always right" card. So I agreed.

What do you think about this?

 

[SouthSIDE Glen]

he asked me NOT TO PUT THE DELAY EXACTLY AT 120 BPM (if the song was at that tempo) but instead at 119 or 121 because of PHASE issues :S

I swear, I have heard and read more hand-wringing, white-knuckling, and loss of sleep over "phase" in the past three years on this forum than I had in the quarter-century I have been messing with this stuff before that .

The whole idea behind setting the delay to the BPM is to reinforce the beat. This reinforcement in fact happens because the waveforms compliment each other. If the waves were not exactly coincident, then phase might start affecting them in slightly different ways, some good, some bad. It's when relative phase starts approaching 180° that it really stats to become a problem.

Frequency represents time (cycles PER SECOND). Therefore every tempo (also representing time) will be coherent at certain frequencies and incoherent at other frequencies.

The question is, for which frequency do the waveforms approach 180° out of phase? Some frequency somewhere is *always* going to be out of phase for any given delay time.

A little math:
119BPM = 504.2ms/beat
120BPM = 500ms/beat
121BPM = 495.87ms/beat

This means that sliding the delay by one beat either way at that tempo is the equivalent of changing the delay time by just over 4ms. To keep the math simple, let's say 4ms.

4ms also happens to be the length of a full cycle at 250Hz. A pretty important section of the mid-bass. This frequency/delay time is also an even multiple of one second, which means that any whole number BPM will be evenly divisible by this number as well. Which is all a fancy way of saying that at exactly 120BPM, the original wave and the delayed wave will be coherent (in phase) at 250Hz.

This coherency will, in fact, be true for every frequency that evenly divides one second.: 20Hz, 25Hz, 50Hz, 100Hz, 125Hz, 200hz, 250Hz, 500Hz, 1kHz, etc.

Slide the timing by 4ms (approx. one BPM at ths scale) and the coherency at 250Hz will not change, becuae all we are doing is moving things one full cycle at that frequency. However, because this is only a half-cycle change at 125Hz, 125Hz will experience 180° phase incoherency if you try sliding the timing of the delay off by a single beat. So that change will not affect 250Hz whatsoever, but will have maximum effect on 125Hz.

So, what it comes down to - as is usual with most things - is how does it actually sound. Some frequencies will be complimentary, others will not. This will always be the case; it's just a question of which frequencies and how important they are to that given sound. If a delay at exactly 120BPM sounds wrong, then slide it a bit or do something different altogether. If it doesn't sound bad (which it very well may not) then there's nothing to worry about.

This will be true regardless of the delay time; there is nothing "magical" about an exact BPM match in this regard.

G.

 

[mikeshead]

Many thanks Glen (and others),
And I thought I had discovered the “secret” to recording when I discovered the basics (and I do mean basics) of EQ. It feels a bit like when you’re studying for and exam and you feel OK about it, then you learn something new, which reveals just how much you don’t know.
I need some more reading. Any websites you can recommend?
By the way, why is there some much discussion on the forums about equipment, when there is all this stuff to learn about eq, reverb ect?

 

[SouthSIDE Glen]

I need some more reading. Any websites you can recommend?

Hahahaha, I know that by Internet standards I am considerd an old man, but I still find it humerous nonetheless that people actually equate websites with reading these days.

Get out of the house for a couple of hours. Head down to Borders or Barnes and Noble, grab a half-dozen books of the bookshelf, get yourself a cup of coffee or a soda at the cafe, flirt with the cute lady serving up the drink behind the counter (maybe get her phone number), and enjoy an hour or two of the pleasure of being able to browse information in a much more relaxed and workable format than trying to read something printed on a stationary lightbulb.

I'd start out with "The Mixing Engineer's Handbook" by Bobby Owsinski, "Understanding Audio" by Daniel Thompson, and "Modern Recording Techniques" by David Huber, all three of which have been in stock on the shelf in every Borders I have been in recently (a larger number than you might think.) There are usually at least three or four other good books in stock as well, along with a couple that may not be so good. If you decide you want more than an hour or two with a book that particularly connects with you, you can either buy it or have your library get it for you for free.

By the way, why is there some much discussion on the forums about equipment, when there is all this stuff to learn about eq, reverb ect?

It depends upon the forum and who is doing the discussing, but the easy way to sum it up IMHO is that it makes for relatively boring conversation to say that the path to a great recording requires patience, a trained ear, attention to detail and more practice. It's a lot more fun to talk about the latest gizmo with the big flashing lights that claims to cure the common cold and make one more attractive to women, as full of crap as those claims may be.

G.

 

[fraserhutch]

Makes me wonder if the only things to be found in the libraries of the future are interweb terminals.......

 

[Rokket]

It always amazes me how many people accept what they find on the internet at face value.

A lot of the information on the internet is good, don't get me wrong. But so much of it has no means of varifying the information. You don't know if the author of that web page is indeed an expert or just Joe Donut putting his own personal opinion on the web...

 

[starbuck26]

who you calling joe donut?

 

[mikeshead]

Glen seems to know what he’s talking about, so off I go to the local library. First I manage to find a parking spot...eventually.. Find I have no change for the Meter (for some reason the Mrs thinks she is saving money by collecting every bit of change I secrete around the car for just such purposes). Go into a shop to get change, meanwhile a traffic warden (thin pimply youth) appears out of another dimension and like an exorcet missile heads strait for my car. Like a SAM (surface to air guided missile) I manage to intercept him and say” I’m just getting change mate”. To which he replies in an adolescent voice reminiscent of that guy in Scooby doo “You had better”.
After mentally doing to him what Itchy does to Scratchy, or is it the other way around, (although from a more practical point of view I mentally compose a letter to the council... well I am British don’t you know) I get my change, get my ticket (eventually, the machine needed a good talking to before it finally, reluctantly succumbed and deemed part with my ticket after stealing two fifty pence pieces from me) and go into the library.
Under the section euphemistically marked ” music” in the reference section I note there are about half a dozen books (left eye brow arches ), I then note there are about a dozen other music books placed more or less at random along the selves. I say randomly, there may have been a complicated code which is an advance on the Dewy system most of the Libraries in the western world uses but I doubt it.
Left eye brow still in place, I approach the slightly rotund (British under statement) librarian trying desperately not to make eye contact with me as she is using all her energy to keep her size 16 body in her size 12 dress. I ask, in a hopeful manner, if perhaps I am looking in the wrong section. She checks on the computer. Err “Computer says “No” she says. They are a bit specialised for us she says. “But not for me” I say trying, but failing to keep the sarcasm from my voice as my right eye brow joins its brother. “Perhaps you could check the national library network?” Her eyebrows match mine, then a tight smile followed by finger flashing across the computer terminal key board. “Computer says No” we say together.
I consider asking if she would like to have sex with the thin traffic warden and make a vaguely humanoid offspring, then think better of it, pull my eyebrows down from the ceiling (where they have been hovering in a final defiant gesture) and leave.
Glen, you are a bright and knowledgeable guy. Please don’t make me go out into the outside world again. Please someone give me a website which will answer my questions.....computers says yes?

 

[SouthSIDE Glen]

Glen, you are a bright and knowledgeable guy. Please don’t make me go out into the outside world again. Please someone give me a website which will answer my questions.....computers says yes?

While I'm sorry about your travails, I do thank you for a very humerous, practically Douglas Adams-ish take on it in your post One of the few times when someone types "LOL" and they actually mean it. Glad you're apparently taking it wil a bit of a sense of humor .

Hoooowwwwwwever, if you'll notice, I did not recommend you go to the library. I sent you to either of the two big bookstore chains here across pond (I did not realize you were a Redcoat ). I don't know if they have those chains or their equivalent over in GMT-land, but here those stores have plenty of ample *free* parking, comfy little in-store cafes where you can even order a cup of Earl Grey and a biscuit, and actually have books in stock that are relevant to Human Life As We Know It; far more than your average library does over here. Don't blame me if your little improvisation wound up falling as flat as an American Guiness .

The fact/problem is, I personally cannot point to many quality sources for good, reliable, high-signal/low-noise general information on the Internet for anything other than porn. (If they removed all the porn sites from the Internet there would be only one website left with information you can trust; www.wherestheporn.com.)

As far as the topic at hand, I have heard references here to supposedly good information on tweakheadz.com, but I find it against my nature to spend any time at a website that spells the plural of a word with a "z", so I can't vouch for that. Might be worth a look though. Also TheRecordingProject (rp.com, I think?) has a boat load of articles online, they might have something worthwhile over there.

Additionally, check out the websites for many of the better manufacturers of this kind of equipment. Companies like Lexicon, Kurzweil, TC Electronics, etc. often have very informative how-to articles and documents online. And even more general companies like Alesis, Rane and JBL can be good sources for some real nuggets of gold (sorry, no porn, though.)

G.

 

[mixsit]

..By the way, why is there some much discussion on the forums about equipment, when there is all this stuff to learn about eq, reverb ect?

Easier to talk about the toys..

Here's an angle you can try. The loose connections between 'verb, being packed clusters of many discreet echos, and the roll of the effects- time (or width) spread- smooth' or defined, and the sources -smooth or percussive.
Echo vs verb- Some times the roll is depth, smear or width, with little tail or 'fill. Discreet echos are easier to hear' (relative to verb). Perhaps to say, they have the maximum audible effect with the least filling in the mix, or minimum return level-
Sharp sources 'show' discreet echos -and the 'details that make up verb.
Sharp sources like very dense verb. Smooth sources like/need verb exposed low density 'innards' , delays more level.

All this is very general of course -but connected and completely true.

Except when it isn't.
___________
Monitoring just fine at CathouseSound SP Tech Continuum AD

 

[mikeshead]

Err thanks Mixsit, I think.

Please forgive my slight reticence, I need to go and read up on quantum physics for some light relief.

With regard to your quote:

Sharp sources 'show' discreet echos -and the 'details that make up verb.
Sharp sources like very dense verb. Smooth sources like/need verb exposed low density 'innards' , delays more level.


Have I stumbled into an alternative universe where my audio forum has trans morphed into a cookery website ?

 

[SonicAlbert]

I did read that George Massenburg doesn't use even delay times. So that 500ms delay time for 120bpm would be 499ms or 501ms. Setting it off by a full bpm, like to 119 or 221, is too far off the tempo. But a smaller increment in the millisecond range would not be noticeable.

 

[mixsit]

Err thanks Mixsit, I think.

Please forgive my slight reticence, I need to go and read up on quantum physics for some light relief.

With regard to your quote:

Sharp sources 'show' discreet echos -and the 'details that make up verb.
Sharp sources like very dense verb. Smooth sources like/need verb exposed low density 'innards' , delays more level.


Have I stumbled into an alternative universe where my audio forum has trans morphed into a cookery website ?

Sorry. I picked up that 'assigning human traits to inanimate objects' thing from a loved one..
Stick a verb with low density and lots of interesting reflections (Lex 'Rich Voc Plate' is loaded with candy) on a vocal.
Now whack a nice crisp (sharp = spike) snare in there.
All those pieces of candy will pop up like bee bees in a can.

Ergo, the mix will not like it. Of course, the more accurate version is- Everyone will like us better if we fix it.

 

[SouthSIDE Glen]

I did read that George Massenburg doesn't use even delay times. So that 500ms delay time for 120bpm would be 499ms or 501ms. Setting it off by a full bpm, like to 119 or 221, is too far off the tempo. But a smaller increment in the millisecond range would not be noticeable.

Welcome back, Al...haven't seen you in a while.

While there certainly would be no harm in shifting things a millisecond to keep them at odd values, I'm tempted to believe that this is GM's version of knocking on wood or always putting his left sock on first . Like you say, a millisecond shift would probably be mostly inaudible as far as delay effect.

Plus, perhaps the most deciding factor here, whether one is talking BPM or ms, the timing against the actual beat is going to vary significantly in either range from beat to beat if there is a human being supplying the beat. Unless one is running a MIDI beat or digitally snapping drum tracks to a grid, even the best drummer will be good at averaging a pretty tight BPM, but individual hits are likely going to vary by as much as a couple milliseconds plus or minus anyway.

G.

 

[mikeshead]

Many tx for your “Obi-Wan Kenobi” like wisdom Glen. Yes, we do have Borders over here (and electric lights and everything). I didn’t mention it because the nearest one is in a town called Brighton (along the south coast for all you renegades who snook away from the empire.... British that is not the Star Wars one). Thing is in Brighton they don’t much like the car, in fact, Osama Bin Laden would get a better welcome in the White House that you get from traffic wardens in Brighton. Yes, it is that bad.

I have been to the Tweekheadz site and in my humble opinion very good it is too. I really like their attitude and if I can work out how to join, I will.

I will try the other websites you mention and I thank you for the suggestions. Did you know the porno website was defunked ? Not that I tried it of course, err someone just phoned me up out of the blue to tell me.

I am picking up quite a bit from discussions on this thread, so guys (and others) please keep it going.

One thing occurs to me, since, reverb times can be enhanced by being related to the bpm of the song, it would appear that in most cases it would be better to apply the reverb (or delay) to the initial tracks, rather than adding further reverb on the bounced tracks (as a general rule). Otherwise it would seem there is a danger of a ”smearing” effect between the two sets of reverb.True or false?

 

[SouthSIDE Glen]

One thing occurs to me, since, reverb times can be enhanced by being related to the bpm of the song, it would appear that in most cases it would be better to apply the reverb (or delay) to the initial tracks, rather than adding further reverb on the bounced tracks (as a general rule). Otherwise it would seem there is a danger of a ”smearing” effect between the two sets of reverb.True or false?

I hear there's a rumble in Brighton tonight...or so a local stray cat whiskered in my ear. Brighton. That's just down the road from Fawlty Towers, isn't it?

To your question, I hate to say this (I really do!), but the short answer is neither "true" nor "false", but "it all depends..." First, the question is whether one is using the effect to try to recreate or synthesize a real-sounding location, or are you just trying to create a cool-sounding special effect.

If it's a delay, you're right, *usually* (there are always exceptions to everything in this racket) that is something that is applied on a per-track basis (though sometimes it might be more than one track); usually that strong of a delay is not going to be part of a natural environment that sounds all that good, so usually it will not be applied to the entire band.

Reverb, though, can have any number of uses. Often for effect on a per-track basis (e.g. reverb on the vocals to make them sound good, but the drums and the gits are dry, or vice versa.) Sometimes it is used on a per-track basis, but not on the entire track. For example, you may not want to reverb the entire guitar sound because it sounds too muddy and mushy, but you do want to stick a pretty substantial reverb "tail" on that last tension note before the two-beat rest or leading into the bridge or something.

Then you have the kind of reverb that you apply to the overall mix, where you are trying to place the band or the mix within an actual (or virtual) room. Many stereo reverbs and room modelers are good for this purpose. This can also sometimes be used in moderation across the mix to act like a little bit of a "glue" bringing the tracks together.

But probably the most important things to keep in mind when considering all this stuff (IMHO) are: a) the old axiom "If the average listener conciously notices the reverb, you probably have too much", and 2) keep the mixing of different reverb/delay types across tracks and mixes to a minimum because (as you alluded to) they will build into a muddy mess real quick on you. As to a) the main exception is if you're purposely looking for an audible special "k3wl" effect, but otherwise, often less is more.

DOH, you're going to trick me into writing a book right here and now, aren't you? You think I'm from Barcelona or something?



G.

 

[SonicAlbert]

One thing occurs to me, since, reverb times can be enhanced by being related to the bpm of the song, it would appear that in most cases it would be better to apply the reverb (or delay) to the initial tracks, rather than adding further reverb on the bounced tracks (as a general rule). Otherwise it would seem there is a danger of a ”smearing” effect between the two sets of reverb.True or false?

I don't understand this question. All the tracks of the song are at the same tempo, right?

But in general you would apply reverb per track, not submixes of tracks. Unless there was a compelling reason to do so I suppose.

I personally think the way to apply reverb in a plugin setting is to mimic how you would do it with hardware. So you'd create an aux track that has the reverb effect, then send to that track from the other tracks. This approach also saves on CPU usage.