Author
|
How to loop a stringsample ?
|
Conny
IsraTrance Senior Member
Started Topics :
224
Posts :
149
Posted : Mar 22, 2007 20:37
|
Hi i have a string sample i would like to add looppoints to so it sustains.
How can i get it to sustain smootly
so it doesn´t sound like it´s looped ? |
|
|
Colin OOOD
Moderator
Started Topics :
95
Posts :
5380
Posted : Mar 23, 2007 20:18
|
You'll need to crossfade the loop. This is the process of mixing in part of the beginning of the sample into the end of the looped portion so that a smooth loop can be achieved. There are several varieties of crossfade, and most samplers will do this for you automatically but I sometimes like doing it manually - it's a particularly rewarding experience for a diehard geek! If you want to know how to do it manually I can tell you but it would be a long and involved explanation... I'm up for it if you are though!
  Mastering - http://mastering.OOOD.net :: www.is.gd/mastering
OOOD 5th album 'You Think You Are' - www.is.gd/tobuyoood :: www.OOOD.net
www.facebook.com/OOOD.music :: www.soundcloud.com/oood
Contact for bookings/mastering - colin@oood.net |
|
|
talolard
IsraTrance Full Member
Started Topics :
64
Posts :
282
Posted : Mar 24, 2007 02:14
|
I'd like to hear it
  Work like you don't need the money.
Love like you've never been hurt.
Dance like nobody's watching.
natanofgaza@yahoo.com |
|
|
Colin OOOD
Moderator
Started Topics :
95
Posts :
5380
Posted : Mar 24, 2007 03:13
|
Ok... well don't say I didn't warn you!
In writing this description I'm visualising working in Sound Forge but any audio editor that lets you set loops and markers should work just as well. Strings are one of the hardest sounds to loop smoothly but it is possible with care.
First, get the best bog-standard loop-point you can, ie. a loop-point that glitches as little as possible. Don't worry if it's not quite perfect. Place markers at the loop start and end points (your editor may do this for you once you've chosen the loop points). The loop start point should ideally be about 1/4 of the way into the sample, and the loop end point can be anywhere before the end (but before the strings stop playing!) so that between 1/2 and 3/4 of the entire sample lies inside the loop. The area of the sound before the loop is used to crossfade so while the figure of 1/4 is not critical it's important that these areas aren't too short or the crossfade might be easy to hear.
So... the simplest kind of crossfade uses a portion of the beginning of the sample and mixes it into the end of the looped part. We will need to define 4 main areas of the sample:
'A' is from the sample start to the loop start. Let's say it's (A) samples long.
'B' is the area inside the loop.
'C' is from the loop end to the sample end. In this kind of crossfade, area C can be deleted as you'll never hear it.
'X' is the crossfade source area, and is (X) samples long. X always ends at the loop start point, and (X) < (A) so that X doesn't contain the attack portion of the sound. When choosing X, you don't need to worry about zero-crossings.
Firstly, copy X into a new audio file and fade it in over its entire length. Next, select the part of area B of the string sample which ends at the loop end point, and is (X) samples long; place a marker at the start of this area so you can find it again, and fade it out. Now copy the faded-in version of X into the clipboard, place the cursor at the marker you just created (ie. at the beginning of the faded-out part of B), and perform a 'mix paste' so that the faded-in X is mixed equally with the faded out portion of B.
Then you go - one manual crossfade You might need to give it a few goes with different lengths of X or different loop positions to get the best results. There are also more complicated types of crossfade which take multiple crossfade sources from different areas of the sample and combine them with the original sound in various different ways; if you want a description of these I can scan in the page from my Ensoniq sampler manual where they are described I'm not going to try it here
Hope the above post doesn't cause your brain to explode... manual sample editing and crossfading can be fun - honest!
  Mastering - http://mastering.OOOD.net :: www.is.gd/mastering
OOOD 5th album 'You Think You Are' - www.is.gd/tobuyoood :: www.OOOD.net
www.facebook.com/OOOD.music :: www.soundcloud.com/oood
Contact for bookings/mastering - colin@oood.net |
|
|
Elad
Tsabeat/Sattel Battle
Started Topics :
158
Posts :
5306
Posted : Mar 24, 2007 10:19
|
|
Colin OOOD
Moderator
Started Topics :
95
Posts :
5380
Posted : Mar 24, 2007 13:32
|
|
Seppa
Started Topics :
8
Posts :
485
Posted : Mar 25, 2007 06:54
|
Quote:
|
First, get the best bog-standard loop-point you can, ie. a loop-point that glitches as little as possible. Don't worry if it's not quite perfect. Place markers at the loop start and end points (your editor may do this for you once you've chosen the loop points). The loop start point should ideally be about 1/4 of the way into the sample, and the loop end point can be anywhere before the end (but before the strings stop playing!) so that between 1/2 and 3/4 of the entire sample lies inside the loop. The area of the sound before the loop is used to crossfade so while the figure of 1/4 is not critical it's important that these areas aren't too short or the crossfade might be easy to hear.
So... the simplest kind of crossfade uses a portion of the beginning of the sample and mixes it into the end of the looped part. We will need to define 4 main areas of the sample:
'A' is from the sample start to the loop start. Let's say it's (A) samples long.
'B' is the area inside the loop.
'C' is from the loop end to the sample end. In this kind of crossfade, area C can be deleted as you'll never hear it.
'X' is the crossfade source area, and is (X) samples long. X always ends at the loop start point, and (X) < (A) so that X doesn't contain the attack portion of the sound. When choosing X, you don't need to worry about zero-crossings.
Firstly, copy X into a new audio file and fade it in over its entire length. Next, select the part of area B of the string sample which ends at the loop end point, and is (X) samples long; place a marker at the start of this area so you can find it again, and fade it out. Now copy the faded-in version of X into the clipboard, place the cursor at the marker you just created (ie. at the beginning of the faded-out part of B), and perform a 'mix paste' so that the faded-in X is mixed equally with the faded out portion of B.
Then you go - one manual crossfade You might need to give it a few goes with different lengths of X or different loop positions to get the best results. There are also more complicated types of crossfade which take multiple crossfade sources from different areas of the sample and combine them with the original sound in various different ways; if you want a description of these I can scan in the page from my Ensoniq sampler manual where they are described I'm not going to try it here
Hope the above post doesn't cause your brain to explode... manual sample editing and crossfading can be fun - honest! |
|
well I'll be honest i just don't understand everything, is there a chance you actually show some snapshots of what you're talkin about. I have trouble understanding the X area..... and also when you refering to x samples, do you mean a lenght equal to the area X or do you mean (X) as a variable?
X has to be part of the original sample right ? but which part ? it seems to me that x should be taken around the center of the loop section so its not too different from the end and the begining of the loop. I' mquite confused as you can see.
cheers |
|
|
Boobytrip
IsraTrance Junior Member
Started Topics :
39
Posts :
988
Posted : Mar 26, 2007 08:21
|
Quote:
|
On 2007-03-24 13:32, Colin OOOD wrote:
Like I said, most samplers will do it for you Sometimes it's fun to get your hands dirty though.
|
|
Indeed, if you're using Kontakt 2, you can go into Loop-mode and put your cursor anywhere in the sample to select a loop start-point, next press the button called "Find Loop End" and most of the times it works without pops, clicks or subsonic rumble. You won't build any editing-muscle tho
|
|
|
Colin OOOD
Moderator
Started Topics :
95
Posts :
5380
Posted : Mar 26, 2007 14:26
|
Quote:
|
On 2007-03-25 06:54, Seppa wrote:
when you refering to x samples, do you mean a lenght equal to the area X |
|
"X samples" means a length equal to the length of area X.
Quote:
|
X has to be part of the original sample right ? but which part ? it seems to me that x should be taken around the center of the loop section so its not too different from the end and the begining of the loop. I' m quite confused as you can see.
|
|
That's ok Seppa, it's a bit of a confusing thing to try and visualise!
Here is where I define what I mean by 'area X':
"'X' is the crossfade source area, and is (X) samples long. X always ends at the loop start point, and (X) < (A) so that X doesn't contain the attack portion of the sound."
So: X always ends at the loop start point. That means it is not inside the loop, but ends exactly where the loop starts:
|________|______|______________|_______|
<-attack-> <-- X --> <- looped area -> <- end ->
This is important, as the end of area X is the only part of the sample guaranteed to play glitch-free before the start of the loop!
Hope this helps explain things a little better.
  Mastering - http://mastering.OOOD.net :: www.is.gd/mastering
OOOD 5th album 'You Think You Are' - www.is.gd/tobuyoood :: www.OOOD.net
www.facebook.com/OOOD.music :: www.soundcloud.com/oood
Contact for bookings/mastering - colin@oood.net |
|
|
Seppa
Started Topics :
8
Posts :
485
Posted : Mar 27, 2007 21:32
|
cheers
|
|
|