Fuse Audio Labs releases the VCL-25A Vintage Vari-Mu Leveler

[fuseaudiolabs]

Anyway, this section was the reason I'm replying. I'm curious as to how you're testing this. I'm using Plugin Doctor and not seeing anytning like that. In fact, the aliasing performance is quite good. Try this:

1. Load up Plugin Doctor if you have it
2. Load the VCL-25A
3. Go to the Harmonic Analysis tab of Plugin Doctor
4. In VCL-25A, set the input gain knob to 12 o'clock (value of 10.0)
5. Adjust the threshold to 12 o'clock (50)
6. Now go to the Plugin Doctor settings and play with different input levels and frequencies for the harmonic analysis. I'll usually start with a signal as close to 1k as possible with an -18 or -12 db signal, then check higher frequencies (17.5k just randomly became my go-to) to really test for foldover.

Results: pretty fantastic aliasing performance throughout.

I'm really just curious why you're seeing terrible aliasing results and I'm not.

Sorry I couldn't resist:

Thank you so much for your elaborations, Funkybot's Evil Twin!

This is a perfect example of why it's a bad idea to try to judge a digital processor's aliasing performance merely by means of measuring its Nyquist gain...there is so much more to it than this.

Let me give an example: I can come up with interpolation/AA filter designs that have like 90 dB of stop-band attenuation around Nyquist, but bluntly put, they'll either have like 1 dB of passband ripple, 512 samples of latency or non-linear phase. And, what does that even mean in terms of how that would actually perform with a given non-linear chain? Correct: Not much, since it's always a trade-off between aliasing/audio transparency/CPU-load/latency for a given setting anyway.

Or here's another one: Imagine a static harmonic spectrum is implemented by means of e.g. Chebychev polynomials - that's not what I do here, but lets consider this for a minute. Converting your polynomial monomial basis you can design and apply your AA filters specifically for each power of the signal because the maximum bandwidth extension is known beforehand. E.g. x^2 will double the bandwidth so you have to apply a halfband filter before squaring your signal etc.. In this case you get around oversampling althogether and depending on the quality of the filters you use the response around Nyquist will reasonably flat.

It's just not a black and white sort of problem and I'm aware of the constraints. This may start a long discussion, I know, but from my POV I'd like to keep it at that

Best,
Ray

[fuseaudiolabs]

I've tried VCL-25A and it's not for me. I'm afraid I don't see what I can do with it. It doesn't sound bad, it just doesn't make anything great either.

And I'm perfectly fine with that

[cturner]

I don't understand what you mean by that sentence.

I'll stand to be corrected by Mr. Dratwa, but Fuse impresses me as a side project driven by interest and research, and given his talent, and that he's doing this alone, he's able to commodify the result quite inexpensively.

Now that we're here, perhaps you would expand on your judgment of the Waveshell as "an antiquated heap of garbage from the 90's" that also happens to be used by Fuse's sister company Black Rooster?

[fuseaudiolabs]

I'll stand to be corrected by Mr. Dratwa, but Fuse impresses me as a side project driven by interest and research, and given his talent, and that he's doing this alone, he's able to commodify the result quite inexpensively.

Well, thank you, I'm flattered - although I kind of disagree on the idea that Fuse is driven as a side project: Yes, it's pretty much a one man show at this point, but that's not really set into stone.

Now that we're here, perhaps you would expand on your judgment of the Waveshell as "an antiquated heap of garbage from the 90's" that also happens to be used by Fuse's sister company Black Rooster?

Imagine you have many many plugins in many different formats and at least two different platforms. From a company's/dev's POV the idea of having a shell is really appealing because your plugin only has to be coded once and the respective VST2/VST3/AU/AAX/whatever shell will take care of translating things into the respective host protocol at runtime - much less maintenance overhead, even if you use the wrappers provided by e.g. JUCE instead. The downside is that particularly the VST2 protocol added the "shell feature" in two different kinds of flavors (VST 2.2 or 2.3 and another 2.4 type of shell protocol iirc) after its initial specification. That's why to this day it's not too well supported in some VST2 environments (even in the Steinberg world). In AU/VST3/AAX I don't understand what all the fuss is about, since the idea of having one plugin expose multiple effects was part of the protocol specification right off the bat (if done right by both the host and the plugin implementation that is, but I think that's a pretty general matter).

Hope this helps.

Best,
Ray

[MogwaiBoy]

Wonder how it compares to VariComp, my favourite - or MJUC, everyone's favourite?

Agreed the intro price is almost too tempting...

I prefer it over MJUC. I love MJUC but this one is easier to get a great sound fast. I also think it sounds better than MJUC. Another poster mentioned TuCo which is another fav of mine that this will also likely replace. It feels weightier in the way it manages low end. TuCo is great for guitars but not so hot on kick and bass. This new one excels at both.

This thing is as smooth as butter on the windowsill! The highpass filter looks pretty gentle on Plugindoctor but it's there. I think VariComp retains low end better but haven't compared directly. The sound when you push this thing is pretty cool! The noise and mix controls are welcomed deviations from trying to emulate hardware to a fault (let the user choose where the noisefloor should be).

[cturner]

Well, thank you, I'm flattered - although I kind of disagree on the idea that Fuse is driven as a side project: Yes, it's pretty much a one man show at this point, but that's not really set into stone.

I was recalling this, which I'm lucky to have located:

Fuse is meant as a side project for my own stuff and as a means to work with other companies besides BRA.

from here: viewtopic.php?p=6984409#p6984409

Certainly it's fine to have changed your mind, or view the same circumstances differently over time. How else to stay brilliant?

[fuseaudiolabs]

Fuse is meant as a side project for my own stuff and as a means to work with other companies besides BRA.

from here: viewtopic.php?p=6984409#p6984409

Certainly it's fine to have changed your mind, or view the same circumstances differently over time. How else to stay brilliant?

Yeah, that's in fact how it started all out, but I don't see it as a side project right now. That's what I meant when I said, it's not really set into stone that this is a one man band.

Best,
Ray

[cturner]

Ray, best wishes for the Fall and into the future!

[fuseaudiolabs]

Ray, best wishes for the Fall and into the future!

Thanks man! Much appreciated.

[mcbpete]

I’ve no plugin coding experience so this is more a query that anything: What causes some compressor plugins to have latency and others not. The aforementioned MJUC for example is zero latency until you switch on the HQ mode and then it has 256 samples of latency, VLC-4 is zero latency and VCL-25A is 256 samples. I stick to zero latency plugins on my channels and mix bus whilst tracking/composing, then don’t really mind any kind of latency whilst mixing – For that reason I’ve MJUC (in non-HQ mode) on the mix bus but VLC-25A is super lovely sounding and would like to have it there instead!

[garfield78]

mcbpete => i think than vcl-25a have 32samples of latency if i"m good .

[fuseaudiolabs]

mcbpete => i think than vcl-25a have 32samples of latency if i"m good .

Yes, that's true it has only 32 samples of latency.

@mcbpete: Often times latency is caused when a plugin performs oversampling, particularly to attenuate audible aliasing artifacts that would otherwise occur when highly non-linear processes are involved or in order to "unwarp" the response of a discretized analog system that can't be matched by other means e.g. Orfanidis or similar filter designs. The oversampling process involves interpolation and/or anti-aliasing which both are typically implemented using linear phase FIR filters with reasonable quality and enough audio transparency. These filters cause latency.

Hope this helps.

Best,
Ray

[garfield78]

mcbpete => i think than vcl-25a have 32samples of latency if i"m good .

Yes, that's true it has only 32 samples of latency.

@mcbpete: Often times latency is caused when a plugin performs oversampling, particularly to attenuate audible aliasing artifacts that would otherwise occur when highly non-linear processes are involved or in order to "unwarp" the response of a discretized analog system that can't be matched by other means e.g. Orfanidis or similar filter designs. The oversampling process involves interpolation and/or anti-aliasing which both are typically implemented using linear phase FIR filters with reasonable quality and enough audio transparency. These filters cause latency.

Hope this helps.

Best,
Ray

Ok, Good, i'm not crazy ! héhé

I like the result of vcl-25a for mix bus , it alow to have almost a constant audio level for different kind source of music .It's very smooth .I like it .

[mcbpete]

@mcbpete: Often times latency is caused when a plugin performs oversampling, particularly to attenuate audible aliasing artifacts that would otherwise occur when highly non-linear processes are involved or in order to "unwarp" the response of a discretized analog system that can't be matched by other means e.g. Orfanidis or similar filter designs. The oversampling process involves interpolation and/or anti-aliasing which both are typically implemented using linear phase FIR filters with reasonable quality and enough audio transparency. These filters cause latency.

Hope this helps.

It does indeed, as I say it was more out of curiosity that anything else. So I'm guessing that's why MJUC increases in latency in HQ mode as it then switches on oversampling

[garfield78]

I'm not a specialist of vintage compressor , what is the function of "floor" feature ?