New Release: Wavefield v1.0.0 by Fine Increments - Wavetable-based Spectral Filter (VST3/AU/AAX)

[Funky40]

And just like a handful of other 1.0 releases here lately, this is barely 0.9.

I`m testing right now again. I wanted to be clearer in my feedback.
So far, this is exactly also my impression. To the wording.
There was no reason to attack him ( TheMaestro) for that. He is right.*


True, i patch within GP (host). it is niche. BUT: it IS the leading live play host.
Coded by a guy how actually has a masters degree in computer science, And IS a professional Musician.


This post is feedback for the Dev, FineIncrements:

So, the same problems i had occour again.
- Some parameters not working, doing nothing to the sound while beeing manipulated by CC control ( host automation)
- sometimes even not working when adjusted on the GUI with the mouse/Trackpad
- sometimes soundwise working, but not giving any visual feedback out.
This three same cases again. Some paramezters much more prone to that: Phase mostly and Floor. some less so but not always. Skew, Depth. So far was start/End and trim offset allways working i think.




ok, ik`ve also checked again vs. BlueCats "Patchwork" (Host). Used as an aU hosted in GP:
- same, just same, but completly diufefrent than zthe other day. Now some stuff works
- while yestreday late, some stuff was completly not working in GP

Also: to make the controls work while starting i have to chose a WT/preset first.
It`s hard to remeber all details and report here everything to detail and totally correctly. But you get the picture @the Dev . There is a main problem: the controls not working, or they do, or just somewhat, somehow.......thigs change. It`s not consistent
Thats my conclusion





I like the plugin as such. I had uses for it. I just try to give a clear honest feedback.....not fighting anything. Not the Dev and not the plugin.

Latency is one thing and completly "it uses FFT" related. I understand that context. We had that thema, case is all clear, topic closed.

The unfunctionality makes it -for me- impossible to fork out for this plugin in this state.
As much i had liked to add this to my FX arsenal


I have DAWs, but i not use them. I have right now no interest to counter check within a DAW


Ahh, to mention: i`m on modern Mac, OS 15.3 or so, AND: i only check/used the AU version. Not checked with a VST3


* and yes, there is a flood of "such" releases recently. it starts to concern me too.

to notes: i have spent here a fair amount of time. this was not just 3x 5 minutes of work.
All in all was i 3 times behind it.

[CinningBao]

I think you'll come across many devs who only have access to a few DAWs and OSs (possibly only 1), and with that, the literal impossibility to test on a broad range of hardware and software. Really, only the major manufacturers who have built a solid framework they can push their DSP into, can be confident that their plugin will work across the board.

I don't know how familiar you are with software development, but building plugins to work on mac/win/linux, as VST3/AU/<linux thing>, to work in all the hosts (who conform to the plugin standards in variable amounts) is pretty difficult.

Small developers also do not have access to a huge team of internal beta-testers (which a large company would) so ultimately even releasing the plugins can create negative attention if only the people who use a certain host are vocal about the failures of the plugin. In reality, as a user and consumer of these electronic software items, there will be situations where our systems sit outside the expectations of the developer, in which case the developer needs to know there's a problem. I don't think it's useful to immediately discard the software and give up on these new devs or the tools they create; I mean, they do want the software to work on all systems! And honestly, they are willing to take more risks than the bigger guys, so don't turn away too quickly!

With that in mind, I would try to consider presenting your issues, rather than "this is clearly unusable, I'm off", as bug reports the developer needs to look at. Like I said, most devs just want their thing to work on your system! You just have to let them know what was going wrong.

It sounds like GP conforms to the plugin standards differently to the hosts I've tried in (Logic and plogue bidule) so it's quite possible the dev has to make changes which are unique to how GP gets the plugin status.

[fineincrements]

I see, just bought a plugin thru a company using stripe with I guess the paypal link integration you mentioned. (https://www.cutclassic.com/)

Ok, for now it needs to stay card only but i'm looking at expanding the option in the future to include the full Stripe payment element which includes Paypal, Apple Pay, etc. This would be a very significant refactor of my website and backend, and isn't a light task.

[fineincrements]

* i understand that this is not how you see your plugin right now. I just mention it ( with no further understanding how much work it would take a Dev)

Honestly from the beginning I just considered the live performance aspect a bit of a niche use case, so didn't treat it as important, and I'm sorry for that. I did have a beta tester wishing it would play nice with GP, and I just sort of flagged it as "maybe down the line, if there's a lot of demand for that".

I'm struggling to see why those hosts aren't picking up the parameters because they're exposed standardly and correctly using the Audio Parameter Value Tree State, which all DAWs had no problem picking up in testing. I'm stumped on that one, but I've never tried opening Wavefield in any non-DAW host. I don't have access to these hosts right now, but if I can get my hands on them at some point I'll try to reproduce these issues on my end and have a closer look, get some logs.

I did not mean to dismiss the possibility that a live performance solution could be implemented on the plugin side so it plays nice with these tools, but they do evidently behave quite differently from DAWs, and I'll need to do a substantial amount of research and engineering work to try to solve, that may involve refactors requiring extensive re-testing in DAW environments.

I did include a note in the documentation upon release that Wavefield is NOT suitable for real time use. That may change. I was graciously given a copy of Unify from Plugin Guru yestereday, and had a chance to learn some of the differences. I made a 1.0.1 patch to Wavefield that makes it play nice with their software. So I want to make it clear that I hear your needs and wants and I'm willing to address them, but it may take some time. I'm sorry that it's not possible to use with GP/Patchwork right now. I'm sure they are excellent softwares. Wavefield has not been tested at all in any non-DAW hosts pre-release, and I am a solo dev without access to every tool out there, as another post here suggested.

I've added an explicit note in online documentation that Wavefield currently does not work with GP/Patchwork, so hopefully that helps in the meantime

[Funky40]

* i understand that this is not how you see your plugin right now. I just mention it ( with no further understanding how much work it would take a Dev)

Honestly from the beginning I just considered the live performance aspect a bit of a niche use case, so didn't treat it as important, and I'm sorry for that. I did have a beta tester wishing it would play nice with GP, and I just sort of flagged it as "maybe down the line, if there's a lot of demand for that".

Thanks for coming back to me.

I'm struggling to see why those hosts aren't picking up the parameters because they're exposed standardly and correctly using the Audio Parameter Value Tree State, which all DAWs had no problem picking up in testing. I'm stumped on that one, but I've never tried opening Wavefield in any non-DAW host.

What i saw from the GP Forum was, that many plugins are NOT necessarily coded that good.
also: there seem to be clearly "defined specs or protocols" out there in regards to Host+plugin communication and how it has to happen. While many DAWs themself seem to not take all of it into account, or not properly.

Over the time i have been under the impression that a live play host as GP has some more problems with plugins than a DAW. I saw many discussion, and the Dev of GP (who has a masters in Computer science) has been very present in said forum. I`ve seen alots of his detailed comments on plugins which would create problems in GP.

So from there was i always under the impression that GP is usually not the cause of such problems. They effectivly can usually track the problem down. It`s also worth to notes that the GP community itself is full of IT savy people, who are at same time pro level musicians. It`s in fact quite stunning.

The cause that some plugins have no problems in a DAW but have in GP "could" exactly be based on the fact that GP is taking the communication with the plugins more explicitly to the protocol than most DAWs do. I even could imagine "now that we speak" -new tought- that DAWs leave out some things not based on lazyness but maybe by real calculus, .....who knows ?

So, you saying your plugins works in DAWs does not mean "for me" it`s flawless.
But i *totally* see the point of how much work and hassle it could be to go thru all the odds, to take care of all eventualitys. Anyway for a one man show.

The whole topic seems much more complicated than even most Devs know. ( i say all that with NO real knowledge in regards to any form of coding, etc. ) But it`s becoming my impression.

another reason why DAWs might be a little less picky could be the aspect that they react inside differently. More buffer and what not ? DAWs *require* to run stable by whatever goes on ! DAWs might have their tricks to keep things as worky as possible. While the focus on a realtime play host has to be latency. But wait, i know for a fact that GP has more than anything else stability in mind. Latency 2nd.



I personally am definitly interested in your plugin, even with it`s given (FFT based) latency.

So, i leave it as that.
I would say we had a good exchange, and i thank you very much for that !
I just wanted to make that statement, but not meant to change anything on this situation. Not saying "you should look into GP" etc. / i totally see your point and stance. And totally accept that.

So, Thank You !


good Devs welcome. Good Tools anyway. Keep em coming

[mcbpete]

Is this kinda like Slink Filter - https://hypnusrecords.com/shop/p/slink
or UZU https://crql.works/uzu/

Or neither?
Or both (in two of the three modes)?

I'm trying my best to keep my new plugin purchase habit to a minimum this year so trying not to overlap what I don't need to !

[fineincrements]

I personally am definitly interested in your plugin, even with it`s given (FFT based) latency.

So, i leave it as that.
I would say we had a good exchange, and i thank you very much for that !
I just wanted to make that statement, but not meant to change anything on this situation. Not saying "you should look into GP" etc. / i totally see your point and stance. And totally accept that.

So, Thank You !


good Devs welcome. Good Tools anyway. Keep em coming

Thanks for understanding. The only standards I measured it by was whether it works in DAWS. The rest is unexplored. I know there are a ton of better devs out there and more expertly built products. It's quite common in software to release a minimum viable product and improve it from there as you learn how the market responds, and how compatible it is with other software. I am just taking the leap for the first time and I'm getting mostly very good feedback. A lot of people are finding value in it, but these are almost all obviously DAW users. I am committed to improving my standards, and there will be updates. I apprecicate your viewpoint - and if you ask any of my testers who challenged me on feature requests and so forth - I am responsive

[fineincrements]

Is this kinda like Slink Filter - https://hypnusrecords.com/shop/p/slink
or UZU https://crql.works/uzu/

Or neither?
Or both (in two of the three modes)?

I'm trying my best to keep my new plugin purchase habit to a minimum this year so trying not to overlap what I don't need to !

Hey, great question! On slink - same inspiration actually (water), totally different processing technology. Wavefield is able to "move like water" because of the complex organically evolving nature of wavetables. but not all wavetables are "water shaped" so there's a whole world of possibilities beyond water-inspired sounds. and the additional processing modes, especially Spectral IR, give alternative flavors of filter movement.

On UZU, phasers again are really totally different processors, and can produce interesting movement and cancellation "filtration" but operate on a fundamentally different and more limited architecture as to what shapes they can produce. A phaser produces a characteristic repetitive notch pattern. its shape and periodicity is constrained by its design. That being said, UZU really is a creative reimagining of a phaser in the spectral domain (where Wavefield resides), and I love the ingenuity. Wavefield can't do what UZU does, and vice versa.

Wavefield morphs through 128 spectral filter shapes derived from a wavetable, frame by frame, meaning you can sculpt completely asymmetric, non-periodic, musically-targeted shapes that a phaser physically cannot produce. You’re not going to easily get the depth of notching from a phaser either - Wavefield’s minimum FFT bin gains on the filter curves can extend effectively to silence. So if you loaded a square shaped wavetable frame, and tweaked the right parameter, you’d have one half of the spectrum boosted, and the other half silent.

There’s a tutorial video on the product page that explains everything. You’re welcome to download the free trial and A/B test it for yourself to hear the difference!

[suthnear]

Morning. Interesting plugin. I have a few questions. Is there a reason changing the speed control doesn't cause the sweep rate to change smoothly? Speeding up or slowing down a sweep is nice effect and does work smoothly if you modulate the WT position via a modulator (e.g. one of live's lfos). It even works discretely when stepping through the sync speeds. So it's not like this is a limitation of your processing engine. Speed doesn't go very fast either. At the highest speed it is considerably less fast than sync.

Also, am I correct in understanding that there is no store of wavetables and thus no way to change them other than via changing presets. I.e. each preset is - effectively - a wavetable

[RPH]

- This plugin reports a latency of 4336 samples in Bertom EQ Curve analyser 2.
- Filters used seem to phase a lot?
- There is a scaling issue. Positive signal is max 12dB, but the negative can go down to something like -132dB. Negative signal can't be shown fully in the main view, it stops at -40dB or so.
- Editor: What you draw isn't exactly what you get in the main view (db mode). Also checked in Voxengo Span and Analyser 2. This is also the case when I reset to a basic waveform e.g. Sine, and it's skewed in the main view (dB mode).
- Editor: Next to pencil and line a curves option would be nice.
- Editor: There are only a few lines to snap to, add more in between or make it user selectable.

[fineincrements]

Morning. Interesting plugin. I have a few questions. Is there a reason changing the speed control doesn't cause the sweep rate to change smoothly? Speeding up or slowing down a sweep is nice effect and does work smoothly if you modulate the WT position via a modulator (e.g. one of live's lfos). It even works discretely when stepping through the sync speeds. So it's not like this is a limitation of your processing engine. Speed doesn't go very fast either. At the highest speed it is considerably less fast than sync.

Also, am I correct in understanding that there is no store of wavetables and thus no way to change them other than via changing presets. I.e. each preset is - effectively - a wavetable

Hey, thanks for digging in, and great questions. On the speed thing it sounds like you have the transport link on. Transport link locks Wavefield’s processing to your DAWs transport for deterministic and repeatable playback. Modulating speed while linked causes abrupt jumps because your fighting against the continual effort to link to transport. Turn it off (the little button in the upper right corner) and you’ll get smooth speed modulation. I mentioned this in the training video on the product page because I knew it’d be a point of confusion.

It is possible for the lower beat divisions (1/32nd, etc) while tempo synced to produce extreme speeds because it cycles through the entire wavetable once per beat division, reaching speeds outside the range of the knob when unsynced. Another thing to note is that the feeling of speed is actually caused by how much the shape changes frame to frame. There are always 128 frames generated per wavetable. If you’ve used the editor at all, wavefield switches from using every frame of the original wavetable to a mode that interpolates between key frames, and you lose the original detail, but gain editing control. So like I point out in the video, you may need to change speed after editing to gain back the feeling of speed you had before editing.

There are 2 ways to change the wavetable while keeping the same settings. In any case, you have to toggle the “parameter lock “ button on the main panels visualization display ( upper left corner of the display). Then you can either flip through presets, or drag and drop a .wav file and Wavefield will convert it into a table. Hope that helps!

[fineincrements]

- This plugin reports a latency of 4336 samples in Bertom EQ Curve analyser 2.
- Filters used seem to phase a lot?
- There is a scaling issue. Positive signal is max 12dB, but the negative can go down to something like -132dB. Negative signal can't be shown fully in the main view, it stops at -40dB or so.
- Editor: What you draw isn't exactly what you get in the main view (db mode). Also checked in Voxengo Span and Analyser 2. This is also the case when I reset to a basic waveform e.g. Sine, and it's skewed in the main view (dB mode).
- Editor: Next to pencil and line a curves option would be nice.
- Editor: There are only a few lines to snap to, add more in between or make it user selectable.
- Editor: Would be nice to have single cycle waveform import.

That's the true reported PDC: Wavefield does spectral processing with a 2048-point FFT, so the pipeline itself is 2 × fftSize = 4096 samples, plus ~5ms of limiter lookahead (240 samples @ 48kHz, 221 @ 44.1, 480 @ 96). It's inherent to the spectral approach. the plugin reports it honestly so the DAW can compensate.

It sounds like you may be hearing the frequency dependent phase rotation that is the whole point of "phase warp" mode. try switching modes

The fact that the full dB negative range is out of view is a limitation of the window size and keeping a consistent scale positive and negative. It's not meant to be a hardcore analysis tool, it's a creative sound design tool. And I figured if you wanted the floor to drop below -40 db, you're approaching silence anyway, so the information is less useful from a creative standpoint. Open to suggestions if you have a better idea.

The reason what you draw isn't what you see in the dB view is that dB shows you the final result of processing. There are numerous controls that shape the sound. Smooth, Depth, Floor, Skew. if you switch out of dB mode, you'll see the wavetable view which is what you're drawing.

You can import small .wav files but Wavefield will always duplicate them in the keyframe system because there has to be 2 reference frames to create relative motion between.

I appreciate your suggestions and will add them to the tracker for future updates. Cheers!

[RPH]

It sounds like you may be hearing the frequency dependent phase rotation that is the whole point of "phase warp" mode. try switching modes

Ah, right. Magnitude mode sounds better indeed.

The fact that the full dB negative range is out of view is a limitation of the window size and keeping a consistent scale positive and negative. It's not meant to be a hardcore analysis tool, it's a creative sound design tool. And I figured if you wanted the floor to drop below -40 db, you're approaching silence anyway, so the information is less useful from a creative standpoint. Open to suggestions if you have a better idea.

Hmm, maybe let the user adjust the upper/lower visible range? Like in Span.
Agreed it's not a analysis tool, but it's always handy to see what the result is in the tool itself. Saves some screen space.

I appreciate your suggestions and will add them to the tracker for future updates. Cheers!

No problem, it's a fun plugin to play around with so far.

[billinder33]

Really digging this plugin. Definitely gonna buy.

I was messing with the 'Talkative' preset earlier, sounds really good. Makes me wish there were a few more formant presets in here. May see if I can export some of the ones from Phaseplant into Wavefield.

[Ah_Dziz]

Having a user adjustable fft size would be a great move. Keep your default as whatever you like most but the change in sound with different sized bins is huge and a useful tool. It also allows for much lower latency at the expense of frequency resolution.
This is the best implementation of this idea I've seen in a plugin.