View Single Post
Old 01-16-2019, 03:17 AM   #97
sai'ke
Human being with feelings
 
sai'ke's Avatar
 
Join Date: Aug 2009
Location: Germoney
Posts: 334
Default

Quote:
Originally Posted by foxAsteria View Post
Er well, I had one more idea...when you say you implemented some of these filters "from papers," do you mean schematics?
Yes, but it was painful and I still don't have any guarantee how close it got because I don't have the real hardware.

Quote:
Originally Posted by foxAsteria View Post
In which case, how hard would it be to add a schematic from a vintage wah pedal? They look pretty simple, only a lot of variations.
I've been trying to use Filther for this, since there are not a lot of good wah vst's out there and with this you could do abnormal things with it.
Narsty and Modulator are working pretty good though...
It may be possible. I'll look in literature first, someone probably already did the heavy lifting for such a thing.
I always thought a wah pedal was just a pretty generic resonant lowpass or bandpass filter and it was the modulation that gave it that wahwah-eyness.
EDIT: Yes found one. It looks like it'll take a while to implement, but maybe when I find a bigger chunk of time I'll give it a shot.

Quote:
Originally Posted by foxAsteria View Post
On that note, when I try to learn one of the sliders with mod range to my MIDI controller, moving the control always makes the mod % parameter show up as last touched, regardless of whether I move it with RMB or LMB.
Ah, that sounds annoying. I will see what I can do about this. I'm not sure how reaper identifies what was last touched.

Quote:
Originally Posted by foxAsteria View Post
There's also a lot of "zipper" action when modulating the cutoff via MIDI. Maybe needs a little inertia?
Hmmmm. I really don't want inertia as a default. There was inertia before, but this limits the resolution by a lot. It also caused the sweeping artefacts that were in Eliseat's earlier demos.
The zipper artefacts should not be there once you record it and play it back. They should also not be there when you operate the sliders in the GUI.
That said, it does sound like an issue worth investigating. I wonder if there's a way to differentiate between slider being controlled by the user and being controlled by automation (as the former needs inertia of 1 block and the latter should have none).
I don't have any sliders left to store a variable inertia. I might be able to stick a toggle in there somewhere.

Quote:
Originally Posted by Sju View Post
Another couple of bugs:
- Tempo synced modulation rate parameter doesn't recall properly
- Tempo synced modulation rate slider is weird in the rightmost values (doesn't always reach the fastest rates)
Is this on 1.54 that I uploaded yesterday night? Because I did fix something with the rightmost value and recall yesterday afaik.

Quote:
Originally Posted by Sju View Post
- When moving a cutoff so that the modulation range goes out of bounds, just cap it at the min/max, instead of changing the amount, so that when you bring it up/down again, the range is retained.
There would be a small performance penalty to this (because then I have to start checking bounds at the sample level), but this could in principle be done.

Quote:
Originally Posted by Sju View Post
- Would be neat if you could move the min and max values for modulation separately of each other, and then move both around by dragging the bar between them. When looking for a sweet spot, I find myself constantly going back and forth with cutoff and range adjustment, because changing one alters the other.
Yeah, I know. I want this too, but insufficient sliders.

Quote:
Originally Posted by Sju View Post
- Reset modulation range to zero with double right-click?
This can be done.

Quote:
Originally Posted by Eliseat View Post
- The LFO modulation should be set to be centered at the cutoff frequency. At the moment it moves inside the modulation range + or - what doesn't feel right. A sinus for example travels thru null to positive and negative values. Hence the LFO modulation should act like this and without the need of a positive or negative bias at the cutoff slider. LFO modulation range should just change how far the LFO travels out of the center of the actual cutoff frequency even without setting a modulation bias.
Afaik, if there is no dynamic value, the modulation is centered at the cutoff value. The total range when modulating is 2x that of the dynamic range.
Initially I felt the same, in that the added ranges are a bit confusing. But seeing the full extent of the dynamics seemed more important than the modulation alone.
I also tried making the dynamic variable 0.5 of the total range when modulation is active, but this made it much harder to do A/B testing with and without modulation.

Quote:
Originally Posted by Eliseat View Post
- The LFO frequency doesn't keep its values. If you save a preset at 1/64 speed and recall it, the slider is still at the far right position but the value is only 1/3.
Is this 1.54? It is possible that due to the introduction of the new sliders, some options shifted. This is a bit of a nuisance in jsfx that the sliders are not handled by name, but by order of the active ones. May have to resave some presets after setting them. If I'd make another jsfx, I'd make sure to occupy all 64 sliders from the get go to never have this issue again.

Quote:
Originally Posted by Eliseat View Post
- Though it is already possible to manipulate the cutoff via velocity thru cc it would be awesome to have this ability implemented in Filther. This would be awesome as it would complete this synth like feeling.
Hm. So you mean velocity modulating dynamic variable? This should be doable.

Quote:
Originally Posted by Eliseat View Post
- Some of the unstable filters just mess things up at a certain frequency or resonance. Wouldn't it be possible to decrease the slider (value) range of those filters rather then looking for a limiter kind of solution? (Have no idea. But if I take the Diode Ladder, which really sounds amazing - it stays (now) stable all the way of cutoff frequency if I don't move the resonance above 0.78. So for me this would make sense to cap the resonance at this point and restretch it over the slider or just make the value stop at 0.78 while the slider keeps its course without affecting anymore.)
Yeah, this had occurred to me too. And I did this for all of them without waveshaper. But stability depends on the waveshaper, amount of modulation, input signal and a few other factors.
I also don't really like the idea of ranges changing when you're changing the waveshaper (not to mention that I'd have to somehow store that info outside the slider values with all the potential bugs that brings with).
Which filters are you having issues with? If I have reproducible blowups, I can study what is going on exactly and maybe solve the problem at the core, rather than limit the functionality. With the SWRs for instance, this limiting had virtually no effect on the sound, but made the filter a lot more stable.

Quote:
Originally Posted by Eliseat View Post
And finally I can't repeat how crazy and cool your Filther thing is. I made a lot of presets which still feels like its only 0.1 % of whats possible. Swapping the presets is like clicking thru synth presets. Such an impact is just stunning. Its amazing! I LOVE IT! (,)_\,,/
Ghehe, thanks. I would also recommend pairing it with a convolutional cabinet simulator. This is what I often do.
__________________
[Tunes] | [Tracker Plugin: Thread|Github|Reapack] | [Routing Plugin: Thread|Github|Reapack] | [Filther: Thread|Github|Reapack]

Last edited by sai'ke; 01-16-2019 at 03:35 AM.
sai'ke is offline   Reply With Quote