Native UAD Spark VST3 Plugins Out for PC+Mac (UAD Hardware Not Required to Purchase)

[Scotty]

If you can share some of your favourite presets that would be great. It sounds like you’ve got your favourites. I've always liked the Plexi but I am not a heavy distortion player and tend to use it for more slightly overdriven tones. I have my share of guitars here and not a single Les Paul. I had one when I was in my 20s and that thing was like a boat anchor around my neck, I've had back issues since my University days. Nice tones though.

You didn’t address the mic modelling or the room sounds captured with the Softube Plexi, safe to assume you can live without it. To me it is part of the sound I like.

A single tonex capture certainly doesn't relegate the Softube Plexi to the scrap heap.

Nothing you said is wrong. In theory.

The Softube Plexi isn't great sounding IMO. It's good. Not excellent or exceptional in any way.

But...there are a LOT of incredible TONEX Plexi's that are excellent and exceptional IMO. Some of those captures come with TONEX, others are in the tone.net library captured by users. I can get a better Marshall Plexi with TONEX just by browsing a few presets than I can with the Softube Marshall Plexi. It's just like, "oh, yeah, that's the sound...Les Paul, roll off the neck tone a bit...hey, Clapton in Cream in 30 seconds or less". It's truly great stuff. I'm still over the moon with it.

[Scotty]

--- double deleted ----

[jamcat]

I am open to the possiblity that a specific tonex capture might do it well or even better for a specific application but I wouldn't take it on face value that component modelling is inferior to AI machine modelling (Tonex) . A Tonex capture is said to be highly accurate at specific knob positions but one capture doesn't fully represent the amp nor does IK claim it will.

In the case of the Softube emulation they component modelled the amp across its FULL operational capabilities AND also added microphones and an actual physical space as well. This is something that Tonex doesn't do on its own and nor does it make that claim.

Traditional hand-modeling is inferior to machine modeling, though. A neural network will always beat a human at finding the most objectively accurate algorithm to model a complex system.

All you have to do is listen and compare.

This is particularly true when it comes to static speaker convolution vs. dynamic machine modeled cabinets. Convolution sounds lifeless, grainy and veiled. Machine learned cabinet responses don't suffer from these shortcomings.

Of course with Machine Modeling, you only get a snapshot, so you only get one mic position in front of the cab. But, IRs are only one snapshot each, too. So IK could just use their mic robot to make 600 cabinet Tone Models per speaker, just as they did for the cabinet IRs. Then just swap the IRs out with the cab Tone Models.

And machine learned amp models don't have to be fixed at one setting either. Just look at the machine learned models from Fazertone. They have complete behavioural models with adjustable knobs of the Klon Centaur and MXR Dyna Comp pedals. And even a Phase 90. And they're FREE.

Even IK indicates that traditional component modelling and AI Machine modelling (Tonex) have reasons to coexist.

Of course they do. Because they sell both.

[Scotty]

I am speaking specifically of Tonex and not the others that you mention. I am not moving the goal posts here.

If you find that specific capture(s) that nails it for you , you are going to happy. Absolutely thrilled. I am not quibbling about that.

These are complimentary technologies. The Kemper is similar (not identical) in concept and didn’t obviate component modelled amp sims. IK is not making this claim and it seems odd to me that you are doing it on their behalf.

Nowhere have we seen IK readying plans to release a model that incorporates 600 or more captures representing a comprehensive cross section of the full amp with various speakers, mics and rooms. Perhaps someone will do it but we aren’t there yet. Maybe in Tonex MK 5 we’ll have reached a milestone where 10 captures will do most of it. Who knows?

Tonex is certainly impressive. The price alone puts it in a class by itself and the results compared to the much more expensive standalone Kemper appears to be as good or better in most cases except perhaps for the “does it chug” crowd but a solution to that issue seems to be within reach.

I’m not doing away with my component modelled amp sims just yet. I’ll never say never though. For now, moving the mics around, swapping mic models on the fly and dynamically changing speaker setups and the size of the room the amp sits in, works really well for me. Not to mention toggling the switches and sweeping the controls to dial in the sound. Tonex doesn’t do that. At least not yet.

I own Tonex and I will buy the capture device. The possibilities are as real as its limitations. That’s all I’ve got to say on this topic. Let’s see where it takes us, exciting times!

I am open to the possiblity that a specific tonex capture might do it well or even better for a specific application but I wouldn't take it on face value that component modelling is inferior to AI machine modelling (Tonex) . A Tonex capture is said to be highly accurate at specific knob positions but one capture doesn't fully represent the amp nor does IK claim it will.

In the case of the Softube emulation they component modelled the amp across its FULL operational capabilities AND also added microphones and an actual physical space as well. This is something that Tonex doesn't do on its own and nor does it make that claim.

Traditional hand-modeling is inferior to machine modeling, though. A neural network will always beat a human at finding the most objectively accurate algorithm to model a complex system.

All you have to do is listen and compare.

This is particularly true when it comes to static speaker convolution vs. dynamic machine modeled cabinets. Convolution sounds lifeless, grainy and veiled. Machine learned cabinet responses don't suffer from these shortcomings.

Of course with Machine Modeling, you only get a snapshot, so you only get one mic position in front of the cab. But, IRs are only one snapshot each, too. So IK could just use their mic robot to make 600 cabinet Tone Models per speaker, just as they did for the cabinet IRs. Then just swap the IRs out with the cab Tone Models.

And machine learned amp models don't have to be fixed at one setting either. Just look at the machine learned models from Fazertone. They have complete behavioural models with adjustable knobs of the Klon Centaur and MXR Dyna Comp pedals. And even a Phase 90. And they're FREE.

Even IK indicates that traditional component modelling and AI Machine modelling (Tonex) have reasons to coexist.

Of course they do. Because they sell both.

[jamcat]

For now moving the mics around, swapping mic models on the fly and dynamically changing speaker setups and the size of the room the amp sits in , works really well for me. Not to mention toggling the switches and sweeping the controls to dial in the sound. Tonex doesn’t do that. At least not yet.

I also really like being able to move mics, and tweak amp settings.

But, as it's been said many times, there are limited settings that you really use on an amp, and there are limited mic positions that sound good. Getting a good sound from both is a bit of work and a small victory in itself. So if a Tone Model gets the sweet spot for both, then you're already in a good position.

And in practice, it's just going to sound better and more convincing right out of the gate than whatever amp model+IR you might load up and tweak.


As for swapping mics and speaker configurations, yeah, it's really great... in theory. But in order to do that, deconvolution is used. And it gets really ugly, really fast.

You can do some experiments to hear how bad it really is in practice:

In AmpliTube, select the 2x12 Z Wreck or the 2x12 Silver Alnico cab.
Both cabs have mismatched speakers in them.

Swap the left speaker with the right speaker, and the right speaker with the left speaker, and flip the stereo image. In theory, the amp should sound the same. But it doesn't. Instead, it will sound very bad by comparison.

You can do the same thing with the mics. Only the SM57 and U87 were actually used when capturing the cab impulses. Just like the speakers, the mics are swapped by a process of deconvolving an impulse of the original mic or speaker, then convolving the result with an impulse of what's being swapped in. And this is also how TR Mic Room works.

So to test the mic convolution, follow AmpliTube with TR Mic Room. Keep the AT5 mic on SM57 and set Mic Room up with SM57 as the SOURCE mic, and U87 as the TARGET mic. Now disable Mic Room and set the mic in AT5 to the U87. This should sound significantly better than the U87 through Mic Room, because it's an unmanipulated U87 in AmpliTube.

Now you can select a different mic as the TARGET, perhaps the MD421, and repeat the process. This time, it will sound the same whether you're swapping the SM57 with the MD421 in AmpliTube or with Mic Room. Because you're artificially swapping with convolution either way.

[Scotty]

EDIT: It looks as if the new Tonex Pedal announced about 14 hours ago is using both Tonex Captures and various IR type loaders to get the cabinet, speaker and mic models under one roof in a stand alone plugin. The Amplitube inpired bits that I like so much are there. It looks very compelling to for live work especially.

[dionenoid]

I have no plugins showing in UA connect
Does anybody know if the Native plugins is available when I have the old UAD-1 card?

Native versions only work with UAD2.

[olepro]

I have no plugins showing in UA connect
Does anybody know if the Native plugins is available when I have the old UAD-1 card?

Native versions only work with UAD2.

Found out by looking thoroughly on the UAD site, but thanks

[taoyoyo]

Is it really only ilok 3 or is the ilok key 2 also supported?

iLok 2 is supported.

[martinjuenke]

Is it really only ilok 3 or is the ilok key 2 also supported?

iLok 2 is supported.

Only Cloud and USB, no Machine Activation, right?

[Gosh]

Is it really only ilok 3 or is the ilok key 2 also supported?

iLok 2 is supported.

Thank you

[taoyoyo]

Is it really only ilok 3 or is the ilok key 2 also supported?

iLok 2 is supported.

Only Cloud and USB, no Machine Activation, right?

Sadly yes.

Would really like to have these available on a laptop, without needing to have an iLok or replying on an internet connection.

[nusound mind]

I have a question, I took the .99cent for 3 months Spark and now that they're available to purchase can I buy one of those bundles, cancel my sub and still keep all the work I've done? Or will I have to print all that work w/ the Spark versions of the plugins and the permanent licenses are like different files altogether?

[nusound mind]

I demand answers! Immediately! *bangs table with fist*

That's it I'd like to speak to the manager please.

[Funkybot's Evil Twin]

I have a question, I took the .99cent for 3 months Spark and now that they're available to purchase can I buy one of those bundles, cancel my sub and still keep all the work I've done? Or will I have to print all that work w/ the Spark versions of the plugins and the permanent licenses are like different files altogether?

Yes. You can buy now and cancel the sub.