how do you time delays...

  • Thread starter Thread starter bmorris
  • Start date Start date
B

bmorris

New member
How do you actually time delays and other effects as such to fit the tempo of a song? Is there a general measurement unit and conversion that enables the average joe to still operate in this arena? Just a little topic on the brain.
B.
 
BPM = Beats Per Minute
1000ms = One Second

(60/BPM)*1000= Whole note value in MS.

For example

(60/150)*1000=400

at 150BPM a whole note is 400 MS. Cut in half for half notes, quarter for quarter notes, etc.
 
thanx...

that is very simple to follow and I think that this will certainly put me on the right path. I have a little background in theory but this is a different ballgame so thanx again...

B.
 
Thanks Guys!

It's amazing how sometimes you don't think of the simple stuff ...
 
Go to www.analogx.com for a free delay calculator...

However, let me add that when using delays (particularly multiple or looped delays) you should use prime numbers...(Many thanks to George Massenburg for sharing that)...


Now then, having said that, let your ears (and/or the client's ears) tell you if it's right! Exact beat matching is not always the proper mode for delay usage...In many instances you may want a delay on (for example) a vocal track to fit between the tempo "pocket" for effect... (Did that sentence make any sense???)....

I suppose I'm trying to say, more often than not, trust your ears more than you trust a calculator...A calculator can get you in the general "vicinity", but your ears should be the final judge...
 
marsmgr1 said:
... you should use prime numbers...(Many thanks to George Massenburg for sharing that)...
prime numbers?? A number that can only be divided by itself and 1? Why is that? :confused:
 
Giganova said:
prime numbers?? A number that can only be divided by itself and 1? Why is that? :confused:

For a list of the first 1,000 prime numbers:

http://www.utm.edu/research/primes/lists/small/1000.txt


George Massenburg's Explanation (I'm semi-paraphrasing but I've tested it and he's correct!)!

"I do use prime numbers for short, multiple delays, I do this because:

1) individual delays, when looped, invariably result in "pseudo-modes" that are inharmonic, and...

2) multiple delays (again, when looped) are somewhat less likely to have flutter."

-- -- -- -- -- -- -- -- -- -- --

Now then, that's MUCH too scientific for me to totally comprehend but I've experimented with it and it seems to hold true so far...(Go figure that someone as prominent as George Massenburg would figure that out...After all, isn't he supposed to be pioneering the technology of something called a Parametric EQ?...*LOL*)
 
Use prime numbers for what parameter? The number of repeats? I don't even know of many effects units that let you define it with a number.
 
TexRoadkill said:
Use prime numbers for what parameter? The number of repeats? I don't even know of many effects units that let you define it with a number.

No, not for number of repeats...The actual delay(s) amount(s)...

The "looping" number of repeats (feedback) will be directly correlated to the initial delay time, ergo remaining in the "prime number" domain...

Now that I've said that, I should add that the "prime number" scheme is not a steadfast rule...There will be songs/genres that the inharmonic structures may not be affected by "non-prime" number delays...But having A/B'ed a lot of my stuff I can definitely hear a difference...BUT, can John or Jane Doe hear it...Who knows but why not let 'em have the chance?
 
Now that I've said that, I should add that the "prime number" scheme is not a steadfast rule...There will be songs/genres that the inharmonic structures may not be affected by "non-prime" number delays...But having A/B'ed a lot of my stuff I can definitely hear a difference...BUT, can John or Jane Doe hear it...Who knows but why not let 'em have the chance?
Are u saying that instead of using 400 in the example "(60/150)*1000=400" you would use 401 instead since it is the closest prime #?
 
alter the parameters??

I've noticed that on many, almost all, effect processors, including those made for specifically one effect (i.e. a digital delay processor) that there really isn;t any way to alter the parameters of the effect. there is also no parameter to tell the processor how many 'tap's you want to hear. i have a Digitech 400 multi effect valve processor, and all of the delays in it are all preset. you can't alter they're parameters. theres just like, 400ms delay, or 700ms ping pong, stuff like that. and if you don;t like it the way it is, ur screwed. i wonder is there any way to custom taylor the tempo and the amount of taps using this processor. it is a very good processor, and the delays that it has are really nice. i want to control them now, so i can make it fit my music.
this processor resembles a Digitech 2120, it seems to be set up like it, as far as the look and feel, as well as the architecture of the processor.
how can i tell this thing what to do???
Mike...
 
bmorris said:
How do you actually time delays and other effects as such to fit the tempo of a song? Is there a general measurement unit and conversion that enables the average joe to still operate in this arena? Just a little topic on the brain.
B.

60000/BPM

6000/150BPM = 400ms

You can divide or multiply the result number for different settings.
 
Re: alter the parameters??

dcptnsdcvd said:
I've noticed that on many, almost all, effect processors, including those made for specifically one effect (i.e. a digital delay processor) that there really isn;t any way to alter the parameters of the effect. there is also no parameter to tell the processor how many 'tap's you want to hear.


I must disagree with the "blanket statement" above. I have several different digital delay devices and they *ALL* allow me to alter numerous parameters...Even the effects processors on my keyboard synth/samplers have adjustable parameters...

dcptnsdcvd said:
i have a Digitech 400 multi effect valve processor, and all of the delays in it are all preset. you can't alter they're parameters. theres just like, 400ms delay, or 700ms ping pong, stuff like that. and if you don;t like it the way it is, ur screwed. i wonder is there any way to custom taylor the tempo and the amount of taps using this processor. it is a very good processor, and the delays that it has are really nice. i want to control them now, so i can make it fit my music.
this processor resembles a Digitech 2120, it seems to be set up like it, as far as the look and feel, as well as the architecture of the processor.
how can i tell this thing what to do???
Mike...

Mike,

As I am not familiar with those DigiTech devices mentioned, I can't speak from experience...However, I seem to recall using a DigiTech delay several years ago (a guitar processor) that was lying around a studio and I'm pretty sure the delay parameters were user-programmable...

I really suggest you read the manual and look into user-programmable parameter settings...As is not uncommon with some units, factory presets may not be adjustable, therefore, you must create a user preset...I'm fairly sure that can be accomplished with the DigiTech devices...

Let me know how it works out!
 
marsmgr1 said:
George Massenburg's Explanation (I'm semi-paraphrasing but I've tested it and he's correct!)!

"I do use prime numbers for short, multiple delays, I do this because:

1) individual delays, when looped, invariably result in "pseudo-modes" that are inharmonic, and...

2) multiple delays (again, when looped) are somewhat less likely to have flutter."

-- -- -- -- -- -- -- -- -- -- --

Now then, that's MUCH too scientific for me to totally comprehend but I've experimented with it and it seems to hold true so far...(Go figure that someone as prominent as George Massenburg would figure that out...After all, isn't he supposed to be pioneering the technology of something called a Parametric EQ?...*LOL*) [/B]

Are you sure you've got that right? With all due respect to GM, how could choosing prime numbers possibly make any difference? No matter what delay time you choose, prime number or not, it's possible for it to create "pseudo-modes".

And "less likely to have flutter"? WTF?
 
I think the idea behind the prime numbers is quite interesting, and I can see how it would make sense... although I'm finding it a little difficult to explain.

Think of a very simple sine wave to represent the instrument (we'll go simple here to see the theory). Now, think of a sine wave moving exactly twice as fast to represent the Delay wave. Now, overlap them.

Every other delay wave peak is pushing the sine wave, and would in theory create a slight pulsing sound.

I imagine that in theory if you were to choose parameters of prime intervals, this theoretic pulsing pattern would become so random, it would not be probable for it to continuously line up with the wave patterns of an instrument, or even a mix of instruments to create that pulse.

OR, maybe i got up too early and I can't think straight.
 
Ryan,

Sounds like a logical explanation to me! Actually, it makes tons o' sense...

I'm no math expert, nor a physics theorist when it comes to audio, but GM's posture on the "prime number" situation relating to delays has worked for me...While the "flutter" may not, in many cases, be overtly discernable to most human ears, I can't help but think that the brain recognizes it and eventually responds accordingly...

This may, in some way, be indirectly (or directly?) related to the "scientific studies" (amongst serious pros, audiophiles, and John Doe consumers) that indicate lengthy periods of listening to digital recordings is more tiring and/or stressful than listening to the identical analog source...I will admit that I cannot personally validate that postulate, but I would be remiss in not giving some credence to the results of the findings in the respective studies...

Anyway, I must be going now...I have an appointment with my shrink (local pubkeeper) in 15 minutes ;)

Cheers!
 
MaxB got it right!

I used to do all of my delays by ear...and i did a pretty good job if I do say so myself. But then a friend of mine told me about the formula:

60,000 divided by BPM = Delay time in Miliseconds.
Then fine tune it by multiplying or dividing by 2 until you get the desired delay time.

This works 99% of the time...DAMN THAT 1%!!!
 
Back
Top