Menu
What's new
By:
Filters

Negative Track Delay Database / Spreadsheet

Akarin said:
That would be hard to do. Many libraries have a different delay depending on the dynamic of the articulation (like a mp staccato with -40ms and a ff staccato with -20ms.)
Click to expand...
This is my first reaction to the graph as well. It is missing the dynamics component.
 
A.Heppelmann said:
I would absolutely love to have these for the Berlin Series. Especially for staccatos -- it's like every instrument's attack is slightly different (which i suppose is realistic, but that doesn't make it easy to program).
Click to expand...
The Clarinet staccato patch on BWW revive is atrociously late. There is no reason one should have to make adjustments to just this instrument while the other winds are (more or less) fine. Ok if they are late but at least be consistent within your own library.
 
David Kudell said:
That’s bonkers, why would they do that? Well, in that case you just have to pick the shorter number and use that.
Click to expand...
If this is something you haven't noticed, or could not deduce from the methods by which various dynamics are produced on many instruments, I have some doubts of the veracity of timings displayed on your graphs before even trying them.

You have a good idea here in charting timing differences, but you are going to need a bit of technical help to refine the accuracy. One thing that will really throw a wrench in here is all of the various settings available from plugin to plugin. Many different forms of attack, release, transition times, round robin effects, legato adjustments, scripting that contains time-dependent variables. There are many things that can change the timing of just one repeated note of one articulation.
 
This is somewhat related to the topic, though not quite the same: BBCSO has “Tightness” control which I believe trims a small portion of the front of the samples for a faster attack. The manual states that the idea is to use this when playing the line in, and then turn it down/off when playing it back.

I usually just use my ears when it comes to this kind of thing. I feel like this issue is something all sample library developers need to take on board and actually have a switch right there in the GUI or in the settings to turn on and off with the exact delay already set.
 
audioimperia said:


All of the patches in our libraries have the exact same pre-padding default value of -125 ms and you can control the amount of pre-padding with our sample start feature (between -250 ms all the way to 0 ms) :D
Click to expand...

I applaud the work that went into this effort! Such a clever system.
Zedcars said:
This is somewhat related to the topic, though not quite the same: BBCSO has “Tightness” control which I believe trims a small portion of the front of the samples for a faster attack. The manual states that the idea is to use this when playing the line in, and then turn it down/off when playing it back.

I usually just use my ears when it comes to this kind of thing. I feel like this issue is something all sample library developers need to take on board and actually have a switch right there in the GUI or in the settings to turn on and off with the exact delay already set.
Click to expand...
Even better, I think all predelays and track delays should be automatically turned off at recording, and automatically turned on and sync values at playback.
 
Thanks for the spreadsheet! This will definitely help me save some time in the future. I wonder how difficult it would be to make something similar like the CSS control panel for other plugins.

Tbh I'm surprised that other companies haven't standardized their delays or at least made the information more accessible.
 
Zedcars said:
This is somewhat related to the topic, though not quite the same: BBCSO has “Tightness” control which I believe trims a small portion of the front of the samples for a faster attack. The manual states that the idea is to use this when playing the line in, and then turn it down/off when playing it back.

I usually just use my ears when it comes to this kind of thing. I feel like this issue is something all sample library developers need to take on board and actually have a switch right there in the GUI or in the settings to turn on and off with the exact delay already set.
Click to expand...
Basically turn tightness all the way left cuts into the sample to point where the note is heard, turn to the right to have all the sample including the pre note sound.

Now spitfire have 2 issues here, firstly tightness as far as I have found is not available in all instruments/articulations. Secondly the sample sizes different and the amount of prenote sound differs

Audio Imperia have it sorted all samples have 250ms of sound pre note. The sample start control allows you to dial in how many milliseconds you want and then set the corresponding track delay. To get around playing they have a tight button which overrides sample start for recording then turn tight off afterwards
 
Yeah it doesn't make much sense since if you're messing around with track delays you've probably set tightness to 0, not some sweet spot.
 
Hi David , this will be very helpful.

Do you have a link to the document so I can see it in a larger format if possible?

It's really small on here.

Many thanks
 
Last edited:
This is a somewhat new concept to me. I have used it in the past with CSS, but I was using Cubase then. In Logic, I can set it to milliseconds, but it still only allows me to set ticks based on note division, i.e 1/16, 1/8, etc. Not sure why this is. However, even setting this value to the closest option immediately improved the BWW oboe. This is brilliant!

What I don't completely understand is how this number is not a moving target based on tempo. It seems to me that you would need half the value at 60 bpm as 120 bpm. Did I miss this somewhere?
 
Jett Hitt said:
What I don't completely understand is how this number is not a moving target based on tempo. It seems to me that you would need half the value at 60 bpm as 120 bpm. Did I miss this somewhere?
Click to expand...
Yeah... that is one of my concerns as well. Maybe someone can address this.

In my testing (which is very limited) I have found a great deal of timing inconsistencies; not just between tempos and articulations, but even using identical articulations at consistent tempos.
 
Last edited:
Jett Hitt said:
This is a somewhat new concept to me. I have used it in the past with CSS, but I was using Cubase then. In Logic, I can set it to milliseconds, but it still only allows me to set ticks based on note division, i.e 1/16, 1/8, etc. Not sure why this is. However, even setting this value to the closest option immediately improved the BWW oboe. This is brilliant!

What I don't completely understand is how this number is not a moving target based on tempo. It seems to me that you would need half the value at 60 bpm as 120 bpm. Did I miss this somewhere?
Click to expand...
Tempo doesn’t matter because you’re just compensating for the delay between the start of the sample audio file and the beginning of the note.

And the inconsistencies are due to the fact that human beings are editing all of these audio files to make sample libraries and there’s some human judgement involved as to where the file should be edited. Some companies do a better job of this than others.

This is also the reason why the “don’t quantize or it will sound too artificial” argument isn’t really accurate in most cases, since there’s already imperfections baked into not just the speed in which the musician played each note, but where the sample start was edited. (Exception to this is of course samples with really tight start times such as drums.)
 
Jett Hitt said:
In Logic, I can set it to milliseconds, but it still only allows me to set ticks based on note division, i.e 1/16, 1/8, etc. Not sure why this is.
Click to expand...
Yeah I remember there being two fields as well...one is per region and one is the entire track, it’s easy to get those 2 mixed up. And you might also have to change a setting or something in Logic to go from note values to ms.
 
Jett Hitt said:
This is a somewhat new concept to me. I have used it in the past with CSS, but I was using Cubase then. In Logic, I can set it to milliseconds, but it still only allows me to set ticks based on note division, i.e 1/16, 1/8, etc. Not sure why this is. However, even setting this value to the closest option immediately improved the BWW oboe. This is brilliant!

What I don't completely understand is how this number is not a moving target based on tempo. It seems to me that you would need half the value at 60 bpm as 120 bpm. Did I miss this somewhere?
Click to expand...
David Kudell said:
Yeah I remember there being two fields as well...one is per region and one is the entire track, it’s easy to get those 2 mixed up. And you might also have to change a setting or something in Logic to go from note values to ms.
Click to expand...


Under track header (not region header, as correctly stated above) you want to go to the line that says "Delay" and enter your numeric delay time there.

Just was doing that this morning getting Jeager Brass to line up with my NSS.

So using Jeager as an example, enter -125 and it will align with the default settings in the player, if you adjust the default settings you will have to adjust the track delay by a parallel amount.
 
You must log in or register to reply here.
Top Bottom