![]() |
![]() ![]() ![]() ![]() ![]()
|
next newest topic | next oldest topic |
Author | Topic: Regenerative Phasing Question |
Mark Member |
![]() ![]() ![]() Hi- I've been having difficulty making a good "phaser" from the allpass filter modules. In particular I'm trying to have a dependable, re-usable module that I could use with a variety of inputs. The problem is that at moderate to heavy regeneration levels (where IMO the best sonic coloration occurs), the filter tends to "blows-up". I've tried lowering input levels and compressing the input. I've also tried bypassing the filters built-in feedback and using a memory writer to read the phasing, then feedback it thru a function generator and compressor into the same mixer as the input. Though this worked better in some ways, the dreaded crackling and crunching creeps back in depending on the input. Additional compressing on the input doesn't reliably help. I was hoping the Kyma community could help me with this, is there a better technique to use that I'm missing? Could the allpass filter be modified the same way the lowpass filter was to make the (very nice) VCF? Thanks- Mark IP: Logged |
SSC Administrator |
![]() ![]() ![]() Hi Mark, Could you please describe the configuration? Are you mixing the output of the AllPass with the straight signal? Are you using a high order number (e.g. 48)? How does it compare, for example, to the one in Effects/Phasing flanging chorusing (the example called Phasor)? Thanks! IP: Logged |
Mark Member |
![]() ![]() ![]() I've used several different configurations; most are one to two serial allpass filters, with orders ranging from 4 (old fashioned stomp box) up to 48 (really unusual coloration!), the output of the filters is summed with the original signal in a mixer. The allpass filters are set very close to the Phasor example, except I'm using Gaussian as a modulation wavefom instead of Sine. I've removed the level shifting formula ( lfoL * 0.5 + 1.5) * 5000 hz + 50 hz to leave lfoL * 5000hz + 50 hz since Gaussian is positive going. What I find is when I set them up with a sample (good ol' "bucket2.aif") they'll sound fine, changing samples (say to "now I know" or "msdos") the filters will break up with higher regeneration. As a real beginner w/ Kyma, I hope I'm not missing something too simple here, but I've noted the filter module (whether set to lowpass, highpass) will break up with higher feedback at various input levels and frequencies, my guess was the same thing was happening when set to allpass. Ideally I'd enjoy all the filter variants to be as well-behaved as the delay modules are with high feedback levels. Thanks for your help! best- Mark IP: Logged |
All times are CT (US) | next newest topic | next oldest topic |
![]() ![]() |
This forum is provided solely for the support and edification of the customers of Symbolic Sound Corporation.