GPO - volume mod ?

[Ricky B]

Just curious, but, when using the mod wheel to control volume/expression, does the sample switch to other samples as you fade or swell, or is the starting sample only used? If the one sample only is used, would one get the same results using automated volume control on the track once bounced to audio?

Thanks,

RickyB

[Nickie Fønshauge]

You would not get the same results, since CC#1 controls not only volume but also timbre.

[cptexas]

You would not get the same results, since CC#1 controls not only volume but also timbre.
What he said.
I'm pretty sure that there are a couple of sampled dynamic levels per note.

[Ricky B]

Thanks for the quick replies! Like I said, just curious, but, you guys have settled my approach from here.

RB

[JonFairhurst]

In general, GPO only has one sampled dynamic, and uses filtering to change the timbre with the mod-wheel. Similarly, it uses a synthetic attack control to control the attack with velocity, rather than a bunch of different sampled attacks.

It's a brilliant way to keep the sampling costs down, while providing smooth playability.

-JF

[cptexas]

In general, GPO only has one sampled dynamic, and uses filtering to change the timbre with the mod-wheel. Similarly, it uses a synthetic attack control to control the attack with velocity, rather than a bunch of different sampled attacks.

It's a brilliant way to keep the sampling costs down, while providing smooth playability.

-JF
WOW!! :eek:
Sure had me fooled!
Birilliant indeed!
Gary, got any more great concepts buried in GPO?
Care to share? :D

-Chris

[Tom Hopkins]

WOW!! http://www.northernsounds.com/forum/images/smilies/eek.gif
Sure had me fooled!
Birilliant indeed!
Gary, got any more great concepts buried in GPO?
Care to share? http://www.northernsounds.com/forum/images/smilies/biggrin.gif

-ChrisI've made no secret of the fact that most techniques in GPO rely on intricate programming rather than brute-force sample count. Dynamic changes in volume/timbre, attack control through velocity, legato/tongue/slur capabilities, bow stroke direction differences, variability of intonation and timbre, derived "plr" instruments, and many other things are accomplished through data manipulation rather than separate samples. This helped me keep the "footprint" of the library small enough to realize the concept of getting a rather large orchestra into the RAM available on one reasonably well-appointed computer. The downside is the need for a fairly quick CPU to handle all of the real-time calculations in the complex programming. Fortunately, most recent computers meet that requirement. Another advantage to this approach is the continuous nature of most of the expressive controls (no switches between audibly disparate sample layers as volume changes take place, for one.) This can result in smoother, more natural changes that can more closely emulate the behavior of real acoustic instruments. Finding ways to get intrinsically static samples to behave as if they are dynamic is one of the goals of the programming; keeping the library small makes it more practical for the average user.

Tom