[LAU] Electric pop music engineering history - Was: Mix feedback on a new track?

Ralf Mardorf ralf.mardorf at rocketmail.com
Sun Oct 5 03:46:03 UTC 2014

On Sat, 2014-10-04 at 13:00 -0700, Len Ovens wrote:
> So just keeping the release going will not give the same 
> souund, the release needs to be modified by the new hit.

I'm not thinking about making a drum machine sound equal to a real drum,
I only want to make drum machines sound nicer, so the release trick does
work, if -- the cymbal samples are long and release in a natural way, --
thy are multi-layered.

> I think just adding more drum sounds could make a big difference. A
> left and a right sound, a first hit, second hit sound. Even just using
> the second sound every second or third hit with different velocities
> as well could make a big difference.

I tried that several times in the past, it's too complicated, you can't
play it, you only can edit it. I guess the most important part is to
record drum loops that often, that they are working as a loop and at the
same time they are working without quantisation. Absolutely jitter free
MIDI devices are needed to provide exactly the optimal inexact loop,
this IMO works better than "human touch" features of drum machines (e.g.
provided by Hydrogen). Btw. when using Linux I quantize hi-hats etc.,
because of the time consuming work-flow, but when I used the Atari, the
work-flow was fast enough to "waste" time to make drums sound more
pleasant and there was no jitter working against the effort. IMO jitter
isn't the same kind of inaccuracy as done by a musician. There's a
difference between random inaccuracy and human inaccuracy under the
impact of groove.

More information about the Linux-audio-user mailing list