Compressor help needed

cyto · 13 Mar 2012

Hi mattje!

I didn't look too deep into the example, but I can offer a couple initial observations to get you going...

First: look inside your threshold and output gain modules (the db to lin conversions). You have some of the maths hooked up incorrectly (I think you just have the "pow" inputs backwards).

Second: you need to scale the knobs to have them output the correct range. You can do this simply by hitting the little "p" at the bottom of the module to bring up the properties panel. This is what the doc from your zip file recommends for the values...

slider1:0<-60,6,1>Threshold (dB) slider2:1<1,100,1>Ratio (n:1) slider3:20<0,250,1>Attack (ms) slider4:200<25,2500,1>Release (ms) slider5:0<0,30,1>Output (dB)

I assume the first value is the min and the middle of the bracketed values is the max. I don't know what the other values mean, but I'm sure they have something to do with the scaling. You can hone that in later. For now just put the min/max values in and see if you can't get this thing working.

-cyto

trogluddite · 13 Mar 2012

Welcome Mattje,

Your script is from a Reaper JS plugin, I believe. In which case the format is...

sliderN: Default <Min,Max,Step> description So the important values to put into your knob "properties" are the first two inside the brackets - the Min and Max. You could also include the Default too (it will be used by your plugin when you make a blank preset). SM's knob can do Steps too, but I would leave out that option - I think it is there just to make the numbers format better on Reaper's display, and unsetepped values will give you a little more control.

Happy SynthMaking 🙂

mattje · 15 Mar 2012

Thanks for the help. I've fixed the range of the knobs and the power calculation of the threshold/outputgain knobs. I'd also forgotten to calculate the exponential function of the attack/release coefficients in the code, but unfortunately it's still not working. Very frustrating. I've attached the updated schematic.

trogluddite · 15 Mar 2012

Hi again,

This line in the envelope follower looked a little suspect to me...

env=(det>=env) & det + (det<env) & det+envdec*(env-det);

...so I had a look at the compiler output, and my suspicion was confirmed. The '&' operation has precedence over the other arithmetic, so your code has the meaning... env=( (det>=env) & det ) + ( (det<env) & det ) +envdec*(env-det) So the two complementary comparisons are both 'anded' with 'det' - which is the same as saying... env = det + envdec* (env-dec)

I'm not totally sure what the correct precedence should be, but my guess is that you need an extra pair of parenthesis to make the second bunch of maths into a single 'unit'...

env=(det>=env) & det + (det<env) & ( det+envdec*(env-det) );