Re: gain-mapping - was Re: OSC next

On Sat, 14 May 2016, Len Ovens wrote:

> Only two kinds of math would be nicer than the four I have now... this at 
> least can be considered done. I think It should be no problem in this case to 
> allow a surface that is getting it's feed back in one math to be able to 
> control Ardour in both. This would allow a positional fader with the ability 
> to set the level in dB.

As there is a general agreement that there need only be at most two gain 
mappings with one being an integer and the other dB. I am thinking OSC 
mapping could be:

/strip/gain rid dB
/strip/fader rid int1024

0dB for int1024 would be either 800 (limits max gain to less than +6) or 
792 (overshoots +6db). The other possibility is to leave the math as is 
and special case 800 = 0 dB. (0 would be inf, 1023 would be +6 and 800 
would be 0) this would mean that 800 would be off position wise by 
0.006990539841353893 db or about .003% position. The advantage of the last 
option is that there would be a guarenteed 0dB with no thought to some 
weird math error somewhere else.  Both 800 and 792 work down to 7bits.

--
Len Ovens
www.ovenwerks.net