ZYNAPTIQ Intensity

[pumafred]

I would like to thank Zynaptiq for continuing to participate constructively in this discussion, however harsh it may sometimes become. Thanks!
We’re all discussing because we love music and audio, and are trying to discover what you came up with this time, and if it is of use or interest to us.
BTW, I am fascinated by the growing application of AI and machine learning in this field (and irritated when the words are abused, as I think Izotope has done a few times). Of course, not all attempts will work!

[Nihilith]

We amplify the "Small structures" for the INTENSITY effect – it's similar to some image sharpeners.

Isn't the sharpen of an image enhancing the edges and thus making the image "sharp"? In order for a hair to be visible the effect is to darken the hair while simultaneusly lighten the surrounding area. A sharp image hase more separation visually.

If I were to compare sharpness with an audio property that would be the transients. A sharp sound is a a sound with audible transients and clear separation between elements. To me, Intensity is rather blurring things: bringing the background to foreground, reducing the transients of the foreground, basically it blends everything together.

This would be a smart and useful plugin: analyze the sound, fingerprint/identify the instruments and apply spectral processing to enhance the separation between them. That would worth €350.

i was really hoping it would be this kind of plugin. ->with the help of a.i. a better version of INA GRMs Contrast plugin.

but my experience with Intensity on individual tracks, subgroups and masterbus with a wide variety of settings was an increase of density, and not an increase of contrast. now one could think of density and contrast to be both facets of intensity, but i would have, in this plugin, preferred an increase of the latter.

[martinjuenke]

This would be a smart and useful plugin: analyze the sound, fingerprint/identify the instruments and apply spectral processing to enhance the separation between them. That would worth €350.

By the way: you described what iZotope Ozone and Neutron are delivering...

[jbarish]

i was really hoping it would be this kind of plugin. ->with the help of a.i. a better version of INA GRMs Contrast plugin. but my experience with Intensity on individual tracks, subgroups and masterbus with a wide variety of settings was an increase of density, and not an increase of contrast. now one could think of density and contrast to be both facets of intensity, but i would have, in this plugin, preferred an increase of the latter.

Indeed. On it's product page Intensity is billed as providing "Detail, Clarity, Density, and Loudness". I would like to see more demonstrations of the first 2 when what we've primarily been shown are more examples of the latter 2.

[DrMEM]

...which uses technologies also used in some types of facial recognition – assumes that there ARE broad strokes as well as details in the input.

Intensify uses facial recognition? What does the plugin 'see'? Does it recognize patterns in the waveform or the spectral graph?

Why not use the technologies in voice recognition?

What it "sees" are "big structures" and "small structures" in a special type of time-frequency domain representation; similar to a spectral graph in that it has the dimensions time, frequency, level, but not based on an FFT or similar.

We amplify the "Small structures" for the INTENSITY effect – it's similar to some image sharpeners.

It would be helpful to know what types of "structures" your plugin has been trained to identify to process. Since you call it an AI based plugin, I assume there is some training dataset and some specific set of identification targets.

Is it identifying instruments? Transients? Echoes? Reverbs? Different parts of an instruments sustain? What exactly?

[Caine123]

some Mastering (slight ) tests.

just a quick trap beat loop:

dry (with some eq afterward, soothe and limiter)
https://clyp.it/acts0oin



wet + INTENSITY at the first mastering spot
https://clyp.it/1qwxgzns


an old ebm demo:

dry (with some eq afterward, soothe and limiter)
https://clyp.it/3i5ob1ut

wet + INTENSITY at the first mastering spot
https://clyp.it/y3x3xlvs


newer ebm production, ONLY with INTENSITY on wet signal

dry
https://clyp.it/24yaylf1

wet
https://clyp.it/f2u3fshv

[SparkySpark]

By the way: you described what iZotope Ozone and Neutron are delivering...

I was thinking this too (and also mentioned Izotope above).

So when it comes to usage and result (rather than how they work technically), what is the difference between these plugins? I am still a bit confused...

[Fleer]

I would like to thank Zynaptiq for continuing to participate constructively in this discussion, however harsh it may sometimes become. Thanks!
We’re all discussing because we love music and audio, and are trying to discover what you came up with this time, and if it is of use or interest to us.
BTW, I am fascinated by the growing application of AI and machine learning in this field ...

Well said.
Zynaptiq is one of those very rare developers broadening and deepening our musical ideas. It’s simply mind boggling to see how exquisite their plugins are. I even consider them instruments in their own right. Sure, other developers have tried to make convincing “character” plugins (like Noveltech) but Zynaptiq follows its unique path. I really dig the face recognition reference, allowing me to best understand what Intensity is going for. Wonderful plugin.
Furthermore, it’s also wonderful to see how passionate its developer participates in threads like this one, going to extremes trying to explain, over and over again, what Intensity is about, particularly for people who don’t care to read the entire thread before posting their already answered questions or repeatedly rebutted remarks.
Kudos to other posters who also deeply care for this plugin and aspire to make it even better by focusing on what it does best, intensify any master or sub mix that hasn’t been overproduced yet.

[martinjuenke]

By the way: you described what iZotope Ozone and Neutron are delivering...

I was thinking this too (and also mentioned Izotope above).

So when it comes to usage and result (rather than how they work technically), what is the difference between these plugins? I am still a bit confused...

iZotope N2O8 is a complete (and I mean COMPLETE) suite for all track and bus finishing works. Very good and quick workflow, great results.
Intensity is another (innovative and effective) tool in youe box.
Let‘s phrase this way: I could live without Intensity, but not without N2O8.

[martinjuenke]

I would like to thank Zynaptiq for continuing to participate constructively in this discussion, however harsh it may sometimes become. Thanks!
We’re all discussing because we love music and audio, and are trying to discover what you came up with this time, and if it is of use or interest to us.
BTW, I am fascinated by the growing application of AI and machine learning in this field (and irritated when the words are abused, as I think Izotope has done a few times). Of course, not all attempts will work!

Me too!
By the way:
The analogy of the clash between Porsche drivers and fans of self-driving cars, which someone brought up earlier in this thread, is straight to the point. Some of us will NEVER let their car / plugins do the job itself, some (like me) are happy to have the best possible assistance just to concentrate on other (creative) things.

[zynaptiq]

I think inherently it will be less interesting to those who like to get forensic with mixes and masters and have the skillset to go with it, and more appealing to those who either need stuff done in a hurry or lack specialist knowledge / experience.

Yes! and.. No!

I do agree that for people who actually want to use a variety of tools – including their "secret weapon" and their "secret sauce recipe" - anything automatic, regardless of how good it works, will be something they will be less interested in, or highly skeptical of, for a variety of reasons. I get that, fair enough.

(For that clientele: try it with low amount of intensity, maybe 10%, and use the saturate without any extra gain, maybe even lowering it – gives a nice little "magic touch" as one part of the chain).

The self-driving car analogy is a great one, too – some people enjoy maintaining, tuning, and driving their car, some see it as a vehicle and want to get from A to B as efficiently and safely as possible.

In my personal experience though, that preference doesn't correlate with experience or skill level. It's more a question of preference or personality. I know some top end producers/engineers that don't really know or care what the buttons on that UREI thing are supposed to do but they DO know that it'll sound right on a particular sound when they press all of them (so they'll just leave them pressed and hard-wire that box to the channel that always has that particular type of sound playing) – and then go and mix a record that goes platinum and sounds like you wouldn't believe.
Then, of course, there's the opposite where someone thinks he knows-it-all and gets totally upset and disrespectful about any deviation from his "truth"...but has never released even one commercial record.

That's obviously anecdotal and simplified, but it does serve as reasoning for "no correlation between skill and whether one is interested in the journey or reaching the destination" IMHO



appealing to those who either need stuff done in a hurry or lack specialist knowledge

YES, it is great for a quick fix – and nowadays there's nearly never any time for anything as budgets shrink on an hourly basis – and YES, it is also a great way to get results even if you do NOT have a lot of specialized skills.

BUT NO, not JUST in those cases. It's also great if you have time and skill (and top end gear). It's great at the high end stuff, and can make good stuff great.
The thing is actually quite DOPE when you add 5% of intensity and maybe half a dB of saturation. We have some mastering engineers that we involve in some of our field testing and one of them commented "Aw MAN, cut it OUT will ya, I want to keep my **** job !" when he heard INTENSITY.

That said – if a track is already extremely loud and full-bandwidth – which means it's technically approaching white noise! – then there are clear limits to what INTENSITY can do. Which is kinda logical as there's pretty much nothing that can still do a lot in that kind of scenario. Which is why your friendly local mastering engineer will tell you to bring him the unprocessed 2-track.

For example, Skrillex or AVICII or AfroJack etc there really isn't a lot to be gained from adding INTENSITY on top.

[zynaptiq]

[
Isn't the sharpen of an image enhancing the edges and thus making the image "sharp"? In order for a hair to be visible the effect is to darken the hair while simultaneusly lighten the surrounding area. A sharp image hase more separation visually.

If I were to compare sharpness with an audio property that would be the transients. A sharp sound is a a sound with audible transients and clear separation between elements.

Actually, that's what I used to think before we made a sharpener/blur product. Intuitively, that's what you'd think - but you're describing CONTRAST, not SHARPNESS.

However, the actual audio analogy for a blur is a low-pass filter. It's basically an average across two dimensions (blur) or one dimension (LPF). A pixel "bleeds" into the neighboring pixels, basically. Conversely, the opposite is thus a high-pass filter – which, if you apply it to an RMS signal, would give you a "transient"-style envelope (weeeell...this is not 100% accurately worded but close enough for rock and roll); this is the principle on which transient designers as invented by my buddy Ruben Tilgner of Elysia are based – the derivate on the RMS.

More advanced image sharpeners use a different approach; the image is separated into features of different scales. These are then processed separately, for example smaller features could be boosted for an increase in detail. It's not entirely different as compared to the simpler "LPF"/"HPF" approach, but there's an extra step of breaking up the image intelligently and varying the processing based on the contents of the image rather than on "dumb" mechanics.

Now to your repeated claim that INTENSITY "blurs" stuff – I obviously completely disagree I'm a compressor nut and really ANAL about that kind of stuff. If you're ever in Hannover come and visit our studios and you'll see just HOW much of a gearslut and transient-tightness geek I am. We also have coffee and single malt scotch.
If you email me a before/after example of that effect – together with the setting you were using – then I can most likely make some more specific recommendations as to how to get better results.

That said - when you boost details as with INTENSITY – increasing density – then the level difference between the transient peaks and the "meat" will be reduced – this is the nature of increasing lower level stuff – which, depending on absolute monitoring levels, may seem as if the transients were lower in level relatively.

Same goes for the level compensation – if that is ON, and the signal is HOT, then the stronger transients may be attenuated relative to the detail stuff, giving the impression that they were weakened.

Basically: if you reduce dynamics they will be reduced afterwards.

Also, if you are combining sources that are somewhat loose in their micro-timing, then that looseness may become more obvious. Solution: get the stuff tight

[bbtr]

The thing is actually quite DOPE when you add 5% of intensity and maybe half a dB of saturation.

Why don't you increase this range? Add a /10 button. Where the 1-10 range becomes 1-100, so we could use the trackball to tune in more precisely. Also something similar for the saturation.

Several people complained that it is becoming too much too soon, so that would be an obvious solution. Just deceive them into thinking that they're dialing A LOT. Really MILKING IT. Pushing it into the RED. Sounds 1000 times better, right? RIGHT?

[zynaptiq]

Actually, the thought has come up to have a "zoom" option for the knobs, that's not a bad idea, yes.

Several people complained that it is becoming too much too soon, so that would be an obvious solution. Just deceive them into thinking that they're dialing A LOT. Really MILKING IT. Pushing it into the RED. Sounds 1000 times better, right? RIGHT?

We could also make the trackball itself go RED (blinking, too), add some soccer-match type horns to the sound and have them rise in pitch gradually, add a perpetually swelling snare roll, and start modulating the brightness of the computer screen in sync to the audio for a strobe effect.

Ah wait, that was the brief for the product called ECSTASY...duh. The one that would ship with a couple of p...

Hmm. I'll settle for a PRECISION button for the intensity and saturate dials.

[bbtr]

Hmm. I'll settle for a PRECISION button for the intensity and saturate dials.

That's a promise?