Welcome to the workshop. To try and prevent me from being overloaded in different threads, here's a little thread to track plugin progress. Note that for the alpha ones, beware that slider configuration may still change. If you really intend to use it already, I'd recommend making a backup copy of the JSFX for that specific project.
I also provided direct links to the files with the descriptions, but reapack is definitely preferable, since you can be be kept up to date with improvements that way.
It's become difficult for me to maintain this list of plugins in two places, so therefore, please find the list (with descriptions, images and whatnot) here:
If you find a nice workflow for some of them and happen to make a cool tutorial for any of the plugins, I would love to list it on the GitHub page as well.
Feel free to report bugs or feature requests, but I don't promise I will make any of them happen
__________________ Windows 10x64 | AMD Ryzen 3700X | ATI FirePro 2100 | Marian Seraph AD2, 4.3.8 | Yamaha Steinberg MR816x "If I can hear well, then everything I do is right" (Allen Sides)
Great! Finally a place for all the lovely little gems from Sai'ke.
I tested the Tone Stacks recently but for some reason they down level the volume really hard. Is that plugin meant to simulate the tone manipulation of their (Marshall, Fender) amps? (treble, presence, bass, mids etc?)
The nonlinear attack for Tight Compressor is very appreciated. But I struggle with the 50ms decay restriction. It was way better before to really push the release hard against the beat of a drum loop. And writing in those short times to get the effect back isn't really intuitive. I know that's useless in the case of less rhythmical material like basses, guitars or voices but it was part of its aggressive character which I really like. And short release times also offer great distortion effects. So ... (I know, I'm begging again!)
The MS20 cutoff slider would also benefit from nonlinear behavior because it has way to much way in the not so interesting range of high frequencies. While in Filther the 0.5 cutoff gets grip at about 800hz its way higher in MS20.
And of course many, many thanks for all these goodies. They are very useful, and they all seem kind of organic. You know I love them.
Thanks everyone. I'll post here whenever I develop stuff on one of 'em.
Quote:
Originally Posted by Eliseat
I tested the Tone Stacks recently but for some reason they down level the volume really hard. Is that plugin meant to simulate the tone manipulation of their (Marshall, Fender) amps? (treble, presence, bass, mids etc?)
Yes, with gain compensation off, they emulate what the real devices do (most eat up a lot of gain). Turn on normalize gain to peak to re-normalize the result (this was added in 0.03). Maybe the version you had didn't include this yet.
Quote:
Originally Posted by Eliseat
The nonlinear attack for Tight Compressor is very appreciated. But I struggle with the 50ms decay restriction. It was way better before to really push the release hard against the beat of a drum loop. And writing in those short times to get the effect back isn't really intuitive. I know that's useless in the case of less rhythmical material like basses, guitars or voices but it was part of its aggressive character which I really like. And short release times also offer great distortion effects. So ... (I know, I'm begging again!)
I'll make a better scaled version of it soonish. One thing that's a pity in JSFX is that you can't natively make a slider logarithmic, so I can't put [ms] on the axis anymore then. Both logarithmic would probably be best. Attack from 0 to 120 and release from 0 to 1000 or so. I'll display the actual time where the ratio is now then.
Quote:
Originally Posted by Eliseat
The MS20 cutoff slider would also benefit from nonlinear behavior because it has way to much way in the not so interesting range of high frequencies. While in Filther the 0.5 cutoff gets grip at about 800hz its way higher in MS20.
Yeah. I should probably copy over that new scaling from Filther. Have to add a legacy slider then too though, since this one has been out for a while.
I haven't tested Tone Stack yet, nor am I sure I have a use for it, but it occurred to me that it might be cool to have an invert or 'subtract' mode. Might be interesting.
Perhaps that idea could even be combined into one plugin by having a 'subtract' drop-down box (with no controls, no graph)?
I haven't tested Tone Stack yet, nor am I sure I have a use for it, but it occurred to me that it might be cool to have an invert or 'subtract' mode. Might be interesting.
Perhaps that idea could even be combined into one plugin by having a 'subtract' drop-down box (with no controls, no graph)?
Cheers!
Yeah I mean, basically they're just parametric EQ's with a bit of an unusual shape and phase distortion. I saw these diagrams and was just curious how they would sound. Besides, it was a nice practice to set up a little script that converted the analog filter expressions to BLT digital ones, and before I knew it I had a JSFX again.
Hmm, invert? You mean pass the audio through the tonestack, then subtract the filtered result from the incoming audio? I mean, I can do that, no problem. Might get some phase issues though (comb-filter like effects) as tone stacks result in phase distortion.
Quote:
Originally Posted by ReaMike
Cool. I'll be lurking.
OT: there is a ":" in front of your Author name for Hackey Trackey which prevents it from being grouped with the rest of your work.
Cool and thanks for the heads up. I will edit that asap.
Quote:
Originally Posted by Eliseat
Seems not very special. But it creates an amazing 3D placing effect. Its like pushing a signal back from you with a lot of clarity. Maybe its a psycho acoustic effect. I don't know.
It probably isn't as interesting for Filther as I thought but its nice you checked it with your tools. Did you hear what I meant about it? This strange atomized/pulverized/vaporized distortion? It sounds like 100 tiny layered amounts of distortion. Have no idea how to describe it better.
Anyway. Thanks for taking a look.
And now I will ping pong between your threads.
I heard some strong high-mid boosting and mild squashing, but I only ran it on a saturated 808 and then only test tones. Any input source I should go with? How much do you drive it in your audio sample? Maybe you have a short with/without sample for me?
I mean, it could just be that specific gain curve. I mean, what I can do is measure it, fit it with a little model and put it in a jsfx and see if it replicates that effect for you.
Short term plans:
- Change the compressor time axes to logarithmic, but still using a lower bound which prevents artefacts for low frequencies.
- Add a separate version named unsafe tight compressor here, which will not have the lower bound (that one you can use for your distortion Eliseat).
- Add the compressor to Squashman
It was more a thought-experiment: "could you change a Marshall tone to a Fender tone?" kinda of idea. Not even sure if that would be useful to anyone.
Ahhh, doubtful. There's usually non-linearities before and after the tone-stack which would add higher harmonics. You'd normally never hear the tone-stack in isolation.
...
- Add a separate version named unsafe tight compressor here, which will not have the lower bound (that one you can use for your distortion Eliseat)...
Please don't take my strange obsessions too seriously. I just asked for the 0 release/attack because I didn't get why it shouldn't reach the 0 in the first place. In my case I didn't notice any dangerous behavior. And if you mean that aggressive pumping and pushing ... - every compressor has obviously some constellations where it creates useless or distorted audio. Actually this can get useful in parallel mode or as unconventional distortion effect.
But a better way to give me that little satisfaction would be, to show me where the restrictions sit in the code so that I could change them on my own. That would be less work and distraction for you and more power and time for other cool updates and stuff for everyone.
But of course thanks for your offer. You are not only a blessed coder with lots of fantasy and creativity but also a shining example for a nice and kindhearted person. ヽ( ˘з˘ )ゝ <- Saluting!
Ah, and respective the ts808 "clean" sound. I use this a lot for drum loops and everything I want to be clear but also dirty. (I know, that's a paradoxon. )
Nothing special. But useful. At least for me. So it was a suggestion without much potential. Lets focus on the cool stuff again.
Ahhh, doubtful. There's usually non-linearities before and after the tone-stack which would add higher harmonics. You'd normally never hear the tone-stack in isolation.
Checked the new tight compressor update and made some tests: Still impressive how strong this thing gets into it. I love it. And every new drums will actually run thru it from now on.
No prob. Was a good question. Actually made me think for a bit whether what you mentioned is possible at all. I think it's not, since the waveshaping bits before and after act non-linearly and there is no good way of reconstructing harmonics that were created by this, but which started cancelling out frequencies that may have been there before. I may be wrong though.
Quote:
Originally Posted by Eliseat
Checked the new tight compressor update and made some tests: Still impressive how strong this thing gets into it. I love it. And every new drums will actually run thru it from now on.
I changed the time sliders to logarithmic and changed something to the rendering which should make the gradient less expensive (thank nitsuj for the rendering improvement, he wrote me with a way to do it at lower cost).
Glad it still works for you! In case you *do* want to hack around, I deliberately put MIN_ATTACK, MAX_ATTACK, MIN_DECAY and MAX_DECAY = 500 very visible somewhere near the top of the script.
Those values are in milliseconds.
Changelog: Squashman
- Added per band feedback.
- More visual feedback (gain and band number next to the waveshaper)
- Another waveshaper.
- Improved gradient rendering (lower performance penalty).
Tight compressor
- Logarithmic time sliders rather than linear, with more sensible ranges.
- Improved gradient rendering (lower performance penalty).
Just for the record, Squashman is not really ready for use yet. Here's me randomly dinking around with it:
Once again I have to say: You are crazy! Because my daughter got sick I'm at home and just tested the new Squashman update. Holy cow! Such an amazing distortion tool. I never ever saw something like this before and I'm surprised how useful it actually is. I'm very curious where this is going.
But - to be honest - it doesn't need more cowbell rather than an output level adjustment.
(´-`).。oO(He probably thinks I'm crazy!)
Many thanks again, ... and again!
(Edit: Your example video sounds like early 90ies techno. Nowadays Techno only uses 127-130 bpm! )
Once again I have to say: You are crazy! Because my daughter got sick I'm at home and just tested the new Squashman update. Holy cow! Such an amazing distortion tool. I never ever saw something like this before and I'm surprised how useful it actually is. I'm very curious where this is going.
But - to be honest - it doesn't need more cowbell rather than an output level adjustment.
(´-`).。oO(He probably thinks I'm crazy!)
Many thanks again, ... and again!
(Edit: Your example video sounds like early 90ies techno. Nowadays Techno only uses 127-130 bpm! )
'ello! Hope she gets well soon! And aren't we all a little crazy?
Cool. Glad you like it so far. There's more to come, but the code takes time and I'm bad at making decisions
Yes, gain levels, very important as this is once again a non-linear plug. I am actually still not sure how to best approach this, which is the main reason why it isn't in there yet.
So, the bands controls control drive which is just a pre-gain before the saturation component. One could do a compensating gain back down, but then the drive would act more like a ceiling which I think would be non-intuitive. Rather than make a band go up, the control would make it go down in volume, which would probably make a lot of people go "???".
Second option is total output gain compensation, but this is a tricky one for a different reason.
Each band has a certain slice of the frequency pie. So while the bands are limited by the saturator, it still means that if I'd push every band into saturation (-1 to 1), I'd end up with too much when adding them back. For five bands it'd be something like (-5 to 5). The problem is, I don't know how much one is nuking the band because I can never know the exact input level instantaneously. It's possible to make an estimate of this; but this would mean that the plugin would start acting as a compressor, which is not something I really want. I don't want it to start pumping just from the gain compensation...
The reason that this is not acceptable is because I want to have an optional compressor on each band as well. I want the dynamics to be user controlled rather than automatic. I think it would be very frustrating if you're "fighting" another compressor with the user set compressor.
Third option, which has my preference, would be to just leave it up to the user. Have the user control drive and such in the way it is currently implemented, and make a small post saturation EQ on the bottom bar where output level for each band can be adjusted post per-band-saturation. This small EQ would then also have a Master knob to manipulate the total gain. I could also make an AGC akin to Filther on this master gain, but this would be something that the user should explicitly turn off once settled on a sound (to avoid the dynamic compression effects).
I'm open to better suggestions; but I think I will first implement option 3 and see how that works out in practice.
Sai'ke, why don't you implement it like a post gain slider for every band similar to ReaXcomp? I mean if you want a compressor for every band anyway it would make sense to just use this as the solution.
I agree, that an automatic gain compensation wouldn't make sense because no one knows how the people want to make their sound. But without manual gain adjustment people have no choice between drive and gain, they always get gain. You know what I mean?
Seeing that you're pulling some parts of Filther and giving them their own focused plugin (a great idea), allow me to ask if you'd consider doing a similar dedicated wah js with the terrific wah filters you have. I can't seem to get them to do exactly what I want within Filther so having a wah plugin where the parameters go from only a little further than one end of a normal wah parameter to just a little past it on the other end (rather than from deep earthquaking lava pit insanity to outer space war meteors with a wah in there somewhere in the middle ) I would love it !
__________________
The reason rain dances work is because they don't stop dancing until it rains.
Sai'ke, I guess there is a bug in MS-20 Filter plugin. If I choose the HP Filter and increase the resonance, just a deep rumbling appears. The reason is, that the resonance doesn't fit the filter frequency but it is way lower. That gets obvious if I sweep the cutoff to a very high frequency where the signal gets into a thin sizzling because than the resonance appears as normal bass-low-mid sounding.
Also checked your other tools like the spectrum analyzers. Man, I can't find something useless!
Sai'ke, I guess there is a bug in MS-20 Filter plugin. If I choose the HP Filter and increase the resonance, just a deep rumbling appears. The reason is, that the resonance doesn't fit the filter frequency but it is way lower. That gets obvious if I sweep the cutoff to a very high frequency where the signal gets into a thin sizzling because than the resonance appears as normal bass-low-mid sounding.
Using atan for the integrator (tanh is safer)? How much oversampling?
Seeing that you're pulling some parts of Filther and giving them their own focused plugin (a great idea), allow me to ask if you'd consider doing a similar dedicated wah js with the terrific wah filters you have. I can't seem to get them to do exactly what I want within Filther so having a wah plugin where the parameters go from only a little further than one end of a normal wah parameter to just a little past it on the other end (rather than from deep earthquaking lava pit insanity to outer space war meteors with a wah in there somewhere in the middle ) I would love it !
It may be a while but it's on the to-do list (was also requested by FoxAsteria I think). I first want to get some better quality anti-aliasing going since I need it for squashman and the crybaby also depends on it. Don't expect the load to be lower than in Filther though. That one was just pretty expensive.
Squashman doesn't run properly. If I start it, no sounds get thru. After adjusting the gain once, the sound appears but only the first (active) band. I can repeat it with the second band and then it works like expected. It seems like there is an initiating effect with the gain control.
Greetings
Eli
Edit: After deleting the last separator to only get one band, Squashman just got knocked out. No sound at all. Even after rebuilding bands, changing parameters like gain, wetness, waveshaper etc.
Added another thingamabob for adding stereo to things that lack it. It converts the signal to mono, then uses two all-passes to get some comb filtering going. The comb filtered outputs are added in opposite polarity to each channel, such that when you go back to mono, the original signal is recovered without phase cancellation artifacts.
It sounds good when applied gently. Applying too much will result in something that sounds a bit like it's been flanged on stereo headphones, so be aware of this. The crossover is basically a HPF that cuts the bass of the widening to avoid widening the bass too much.
The last slider allows you to mix in the original side channel (which can optionally also be run through the 12-pole highpass).
Fixes:
Gain issue in Squashman should be fixed now.
HP issue in MS-20 should be fixed now.
Added another thingamabob for adding stereo to things that lack it. It converts the signal to mono, then uses two all-passes to get some comb filtering going. The comb filtered outputs are added in opposite polarity to each channel, such that when you go back to mono, the original signal is recovered without phase cancellation artifacts.
It sounds good when applied gently. Applying too much will result in something that sounds a bit like it's been flanged on stereo headphones, so be aware of this. The crossover is basically a HPF that cuts the bass of the widening to avoid widening the bass too much.
The last slider allows you to mix in the original side channel (which can optionally also be run through the 12-pole highpass).
Fixes:
Gain issue in Squashman should be fixed now.
HP issue in MS-20 should be fixed now.
Can't find it under "Sai'ke". There is only a different stereo manipulation tool. Maybe you forgot the prefix? Or its not in the Reapack yet? ¯\_(ツ)_/¯
But I checked the problems with Squashman and Tighty. Both work fine now. Well done, Sir!
Oh man, I'm a derp. I had given it the wrong extension. Check now
Remember, moderation is key. It can sound overly flangey if you go too crazy with it, but some sounds need more, others need less, hence the range is so big.
Edit: If you want more 'classical' widening of just whatever stereo is already in a sample, just boost the "Side" and leave Strength at zero.
I love Squashman already now, just airwindows TapeDust and Noise where there, playing around with Squashman I could hear beautiful noise bands. Take some seawave sounds, shape it to your desire or wind. Cool, thanks saike!
Btw. I see no gui like above for stereo manipulator. I have v1.0.