i watched a tutorial lately in which morph eq was used to carve out some vocal formants and formant shifts by being set to parallel and setting up those formant resonances with 3 bandpass filters while running a clean saw wave at it. i think that was pretty cool already as it perfectly emulated one of my favourite oldies, delay lama, and added another cool twist to it as well. however, i was wondering why the entire dry saw wave still goes through when the plugin is set to parallel. like… what’s the dry/wet parameter good for if 100% wet is just 50% wet irl? i get it, it’s probably useful with the other filter types and in other situations. nevertheless i think it might make sense to add another parallel alternative where you really only hear the filters’ outputs. imagine that. you could make formant stuff with your clean saw input, but actually use a smooth sound as the “floor” of that patch. do you think it would be possible to kinda fit it in there still? :>
Hey! I’m not sure I totally follow here – wouldn’t you expect the entire dry signal to go through each filter in parallel mode?
Unfortunately I’m still not understanding @ 100% wet you should only hear the filters outputs.
In in series – the last filters output is fed to the next, if in parallel, a copy of the dry signal is sent to each filter, then summed with some gain staging.
Have you seen another behavior?
J
Sry for taking a some time to answer. wasn’t logged in lately. i’m more of a discord person when it comes to communities etc.
I just meant that in addition to the parallel bandpasses the dry signal is also still played on top when using parallel bandpass filters. After I noticed it’s not possible yet to only hear the bandpasses’ output with morphEQ I made my own plugin for that, called Manta. Idk if you’ve seen that one yet, but that basically shows what I mean. I’d still like to see this feature in morphEQ as well, because Manta doesn’t have your pinch- and shift-parameters and this entire multi-node system and all. that would sometimes be useful in the context of parallel bandpass filters as well
Hmm – That shouldn’t be the case – I just gave it a shot, I think it might be that the rolloff on the bandpass are just fairly soft? I’ll double check in the code but I believe it’s correct. We’d like to add some steeper rolloffs to those filters for sure
nah, you’re right. I also double-checked and the rolloff really is just the thing here.
But then I’d add a gain parameter to the parallel bandpasses so that higher q-values are more desirable.