Gigedit workflow

Things that make the work with LinuxSampler & Co easier, better, faster, louder and more sexy. Including friends like JACK, ALSA, etc.
Post Reply
User avatar
dahnielson
Moderator
Posts: 632
Joined: Wed Jan 23, 2008 11:25 pm
Location: Linköping / Tranås, Sweden
Contact:

Gigedit workflow

Post by dahnielson » Sat Feb 09, 2008 2:33 pm

Stuff I've learned the hard way by starting all over several times:

1.) Add all regions you need.

2.) Add any default sample you want to each region unless you're satisfied with NULL.

3.) Add all dimensions you need to the regions.

4.) Add samples to the dimensions.

5.) Edit the settings for each region/dimension. Remember to check "all dimensions" before setting the unity note to save some time.

In short: Plan ahead what you're going to do and prototype it on a single region to see if it works.
Anders Dahnielson

Ardour2, Qtractor, Linuxsampler, M-AUDIO Delta 1010, Axiom 61, Korg D12, AKAI S2000, E-MU Proteus 2k, Roland R-5, Roland HP 1300e, Zoom RFX-1000, 4GB RAM x86_64 Intel Pentium Dual 1.80GHz Gentoo Linux

josander
Newbie
Posts: 8
Joined: Wed Jun 11, 2008 10:30 am
Contact:

Re: Gigedit workflow - Drums

Post by josander » Sat Nov 08, 2008 4:55 pm

Thanks for posting this workflow suggestion.

I'm a real Gigedit Newbie and want to make a drum set of one of Natural Drum's (NDK) kits. Do some one have any workflow suggestion for this?

Thanks, Jostein
Ardour, Rosegarden, Linuxsampler, misc. DSSI & LADSPA and LV2 plugins, M-AUDIO Delta 1010, Edirol PCR-800, 8GB RAM AMD Athlon 64 X2 Dual Core Processor 6400+, Ubuntu Studio

User avatar
konsumer
Newbie
Posts: 9
Joined: Sat Aug 01, 2009 12:31 pm

Re: Gigedit workflow

Post by konsumer » Sat Aug 01, 2009 12:33 pm

The thing I have the most trouble working with is loop points. Do you have any suggestions for finding loop points? Any plans for auto-detection, like new versions of swami? Any ideas on how I could write a little python script to do this with plain wav files (spit out list of possible loop points)? I looked at the swami source, and couldn't figure out how it works. I can't seem to get it to compile, currently, so this is from memory...

User avatar
dahnielson
Moderator
Posts: 632
Joined: Wed Jan 23, 2008 11:25 pm
Location: Linköping / Tranås, Sweden
Contact:

Re: Gigedit workflow

Post by dahnielson » Sun Aug 02, 2009 9:04 am

I haven't looped anything myself since my hardware years. But I wrote a little bit about it here.
Anders Dahnielson

Ardour2, Qtractor, Linuxsampler, M-AUDIO Delta 1010, Axiom 61, Korg D12, AKAI S2000, E-MU Proteus 2k, Roland R-5, Roland HP 1300e, Zoom RFX-1000, 4GB RAM x86_64 Intel Pentium Dual 1.80GHz Gentoo Linux

User avatar
konsumer
Newbie
Posts: 9
Joined: Sat Aug 01, 2009 12:31 pm

Re: Gigedit workflow

Post by konsumer » Sun Aug 02, 2009 9:30 am

Cool, thanks for the info. I knew about zero-crossing, but all the other info is new, and sounds really good. I guess I am looking for a way to replicate the "find loops" feature of swami, without using swami (can't get it compiled, and anyway it seems really geared towards SF2s.) For now, I will try your excellent suggestions.

If I can figure it out, I'd be happy to write an Audacity plugin, or just a command-line script.

I am specifically trying to recreate a SF2 I made of a Casio PT100 (toy keyboard) our band uses it in lots of songs, and I'd like to use higher quality samples of it, and not get any clicking or popping on loop point.

luisgarrido
Newbie
Posts: 22
Joined: Wed Mar 25, 2009 12:16 pm

Re: Gigedit workflow

Post by luisgarrido » Mon Aug 17, 2009 11:42 pm

I authored Swami's autolooper, although the current version is buggy.

The algorithm can't be any simpler, it is just autocorrelation analysis. Basically you make an exhaustive brute-force search looking for similar chunks and you pick up the best results. It is not evident from the actual code because I tried to optimize the search storing intermediate results, but be careful because that code is, as I said, buggy.

I will of course add an improved version including crossfades for loop smoothing to qgiged, but it will take some months until I am at that stage.

If you have any question just shoot.

Cheers,

Luis

Alex
Moderator
Posts: 316
Joined: Wed Jan 23, 2008 9:08 pm

Re: Gigedit workflow

Post by Alex » Thu Sep 17, 2009 10:15 am

luisgarrido wrote:I authored Swami's autolooper, although the current version is buggy.

The algorithm can't be any simpler, it is just autocorrelation analysis. Basically you make an exhaustive brute-force search looking for similar chunks and you pick up the best results. It is not evident from the actual code because I tried to optimize the search storing intermediate results, but be careful because that code is, as I said, buggy.

I will of course add an improved version including crossfades for loop smoothing to qgiged, but it will take some months until I am at that stage.

If you have any question just shoot.

Cheers,

Luis


Luis, please keep us posted on your progress with this, as i'm sure it will be useful for users.

Alex.

Post Reply