UADs Paradise Guitar Studio

[DerKastellan]

I tried it yesterday and my impression increasingly was "this preset sounds similar to that preset" or "this preset is heavily slathered in effects, but not necessarily in a good way." The tonal palette seems to be limited unless you're into specifically that, then it might be a smorgasbord of variety, can't tell.

I selected "Aggressive" to get a bit of a sample of how heavy this can sound and plugged in my G&L Comanche Tribute. This guitar can be set to quite a bit of tonal variety as it has its share of low and top end you can dial out. But the tones didn't sound aggressive - they sounded muddy, fuzzy, undefined. Some tones didn't seem "aggressive" in the least, making me wonder who tagged this. I started to get the impression everything sounds like fuzz. I picked "crunch" and "overdrive" as well, a naming that just adds to the confusion that the terms aren't exactly agreed on. Actually, the preset browser was mostly confusing.

No longer focusing only on gain, I ended up playing two different varieties of strat-likes through the amp sim. Varying the input volume, different pickup combinations, and I was getting more and more dispirited. At some point when trying out amp sims I would find something I like. But I felt rather more confused and befuddled by finding anything. Some nice stuff occasionally, but the wow factor wasn't really setting in. (And I own too many amp sims to claim I'm not easily wowed.)

There was a heavily processed preset that was meant to show off the modulation capabilities of the thing. It was impressive in that regard only. It worked neither on chords nor single notes. When I go to Guitar Rig, there are some heavily processed, modulated presets in there - and you often find something bizarre, fun, or in some way sonically interesting. Here this preset seemed like a mere demo with no application.

Then there were two "edge of breakup" presets side-to-side. I dial the guitar in to get the first one to sound nice. And it did sound nice! And then I switch and I have to dial it in all over again. Why would I need to do that? Why would the edge of breakup be different with the same guitar? I mean, yes, the amps have different points where they are breaking up, but in a preset I'd expect somebody dialed it in with the same input sample and then tuned for that. And yet they were wildly different in their expected input gain.

Then there's the topic of variety. You get two types of vintage fuzz, but for most everything else it's one option, so the pedals won't turn this into a playground, either. You have no filter type - no wah wah. (If this were a vintage emulation amp sim, it should have wah and some early vibes, but no micro pitch shifter.)

One bizarre thing I just noticed is that you can't use the same effect twice - it will be marked "in use" and grayed out. (You can use it again in post.) But let's say you want to use an EQ/clean boost twice in gain staging - doesn't let you. This is a bizarre choice, emulating the limitations of a real-life pedalboard (if you only own each pedal once) instead of giving you options that pure digital makes possible.

So, let's say some of the limited selection is that they packed in what they have in their plugin repertoire. Then okay, no wah pedal. But the absences more noticeable and glaring are for post. I mean, UA has a big selection of rack-mounted effects. Yet only the FET 1176 is present. No opto compressor, the classic go to for clean strum sounds. UA sells it, doesn't include it.

Which tells me... they probably plan that for a future release. And this is where I don't trust them, because I believe this is where the paid extensions will come. (I would be all too glad to be proven wrong on this.) Their pricing in the past has already been notably unforgiving. I own, for example, the Lion and the Ruby, but no good deal specifically for completing your collection. By the way, the offer I've been emailed for the studio was 99 as well. (Of course, this price does not include tax, a trick also IK Multimedia pulls every time, but quite a few plugin vendors actually do this correctly when telling you prices in email. I know it's not required in the US...)

As for drive pedals, if I see it correctly for distortion they took models they had for Apollo (but not native) and put them in ("Distortion Essentials Bundle"). They also added three not available before. The question is, how many will they add in the future - for example in a high gain pack - to round out the picture.

I know they have two pedals for high-gain that they haven't modeled yet, so one could expect that these will drop eventually - but I really think they will charge for them. Pretty much like the Softube Amp Room ecosystem works - the masters of making you pay through the nose. (And these would be actually stuff I'd want in there for variety.) And the same I see given the fact that the king of emulating rack FX has chosen to include only one of their compressors. Basically they could be saying "well, for post-processing you could actually use our own fine other plugins" but then they are not part of your preset anymore. They live inside your DAW only. (For comparison, if you own something with Softube that is "Amp Room capable", it will show up inside the plugin. Where possible you get both, at least. Still expensive, though.)

For comparison, Line 6 has been constantly updating the Helix Native for a long time, lots of variety in there, it still keeps blowing me away. NI Guitar Rig 7 is a hidden gem because it contains all the separate plugin effects NI has made (like reverbs, compressors, the Raum delay, some crazy distortion units) besides its own, GR only stuff - you could use them through the plugin without ever buying them separately. (To be fair, NI has been super-lazy about adding new stuff and does paid upgrades that they then have little long-term love for.)

I also take note that the Enigmatic has not been released as a separate plugin, anymore. (Please correct me if I overlooked it.)

I actually will continue to test this plugin, but start ignoring the presets and just see where I get and if that's more fun.

[Uncle E]

There was a heavily processed preset that was meant to show off the modulation capabilities of the thing. It was impressive in that regard only. It worked neither on chords nor single notes. When I go to Guitar Rig, there are some heavily processed, modulated presets in there - and you often find something bizarre, fun, or in some way sonically interesting. Here this preset seemed like a mere demo with no application.

Yes, I wouldn't expect UA to do anything as far out as NI.

Then there's the topic of variety. You get two types of vintage fuzz, but for most everything else it's one option, so the pedals won't turn this into a playground, either. You have no filter type - no wah wah. (If this were a vintage emulation amp sim, it should have wah and some early vibes, but no micro pitch shifter.)

The Studio EQ has 4 filters. Agreed that it needs a wah, though.

Which tells me... they probably plan that for a future release. And this is where I don't trust them, because I believe this is where the paid extensions will come.

Seems likely. It wouldn't be the first time.

As for drive pedals, if I see it correctly for distortion they took models they had for Apollo (but not native) and put them in ("Distortion Essentials Bundle"). They also added three not available before. The question is, how many will they add in the future - for example in a high gain pack - to round out the picture.

They're all taken from the UAFX pedals. The amps not included, Anti and Knuckles, both have Tube Screamer overdrive pedals, so there are no new pedals to add from those. These are different from the Distortion Essentials Bundle.

I know they have two pedals for high-gain that they haven't modeled yet, so one could expect that these will drop eventually - but I really think they will charge for them. Pretty much like the Softube Amp Room ecosystem works - the masters of making you pay through the nose.

UA's been pretty decent about upgrades. The upgrade to Paradise is $79. The Guitar Amp Bundle Upgrade is $25. I could imagine them instead doing a separate high gain plugin but I would expect them to have a crossgrade offer available in that case.

For comparison, Line 6 has been constantly updating the Helix Native for a long time, lots of variety in there, it still keeps blowing me away.

Yes, based on your comments, Helix sounds like the perfect fit for you. Hopefully they'll be releasing the Stadium plugin soon.

I also take note that the Enigmatic has not been released as a separate plugin, anymore.

Correct. I don't expect it to ever show up.

[jens]

They are doing a bit of modeling, capturing and Machine Learning... I beta tested Guitar Rig 6 and 7, when Julian Parker was at NI.. and from my memory it involved modeling, some capture and definitely a lot of machine learning too..

Machine learning yes, as you can read in the article I linked to - however they use it to basically model the circuit in its entirety. As they explain they make a whole lot of measurements for that at all kinds of different points within the circuit and then they let their machines (back when that article was written) run for several weeks. How long takes a Tonex capture any user can do themselves? 30 minutes?

It's hardly surprising that most (if not all) Tonex videos out there showcase full-on distortion stuff that is completely void of even just the slightest hint of dynamics/nuances. It's all the-right-betwixt the-eyes testosterone-plagued mettel stuff for the eternally adolescent. No wonder they call their guitars "axes".
Edge of breakup? What's that?

[tonycore]

How are people getting $79? The lowest I can see in my cart at checkout is $99 with my loyalty offer, and a discount coupon I have will not apply to this offer.

[Uncle E]

Edge of breakup? What's that?

Tonex captures respond well to input gain. Many heavier patches do nice edge of breakup tones when the gain is turned down. Regarding Paradise, Ruby and Woodrow are particularly good at edge of breakup tones.

[jens]

How are people getting $79? The lowest I can see in my cart at checkout is $99 with my loyalty offer, and a discount coupon I have will not apply to this offer.

Maybe contact their support if you assume that you should qualify for the $79 offer - might be some glitch in your case.

[Uncle E]

How are people getting $79? The lowest I can see in my cart at checkout is $99 with my loyalty offer, and a discount coupon I have will not apply to this offer.

That’s my upgrade price but I own a lot.

[jens]

Edge of breakup? What's that?

Tonex captures respond well to input gain. Many heavier patches do nice edge of breakup tones when the gain is turned down.

Thanks to the recent group buy I own them all (I think) and beg to differ - I found in all cases I tried the algorithmic Amplitude counterparts (or similar models) to be significantly better in that regard.
The best explanation I can come up with for all the Tonex praise: "faith can move mountains."

Regarding Paradise, Ruby and Woodrow are particularly good at edge of breakup tones.

Yes, they're okay - the insane amount of CPU they use to achieve that however is not - but if the emperor needs new clothes he needs new clothes, regardless of their price.

[zvenx]

...
Machine learning yes, as you can read in the article I linked to - however they use it to basically model the circuit in its entirety. As they explain they make a whole lot of measurements for that at all kinds of different points within the circuit and then they let their machines (back when that article was written) run for several weeks...

What do you think their ML was comparing its model to in the process and tweaking its model?
My recollection is that there was capturing as part of that process...

But yes, if your real point is NI's method is very different from the Tonex method (though i am not sure if I even know what the Tonex method is behind the scenes), then indeed I was wrong to just label NI as capture (and I have since corrected it on youtube to where I am comfortable).

rsp

[jens]

What do you think their ML was comparing its model to in the process and tweaking its model?
My recollection is that there was capturing as part of that process...

From the article I linked to above:

Boris explains some more about the process: “You can think of it like getting from point A to point B on a map. You know where you’re starting from, and where you want to go, but you’re not yet sure of the places you’ll pass on the journey. So as well as the input and output of an amp, we also make electrical measurements at various key points inside it, which are called ‘states’. These states are a bit like those missing points on the map, and they tell us everything we need to know about the dynamics of the overall electrical system.”

“We put probes and measuring devices into high voltage amplifiers to collect lots and lots of data – input data, state data, and output data,” Fabian adds. “Once we have all those measurements, we give them to the artificial neural network that we’ve designed.

So again: they measure states at various points within the actual circuit, so that's nothing the average user could do themselves; even if we had their sofware and the required processing power.
You need a) an electro-engineering background and b) a good knowledge of what it is you exactly want to measure and why (i.e you need to know what exact information their neural network requires and be able to obtain it and subsequently where to feed it to).
The Tonex captures on the other hand everyone can do at home - and in a rather short span of time on average machines. And I don't think the NI folks would do it the way they do if they'd think the Tonex method would be adequate.
But then again I'm not taking their word for it. For me playing through the ICM-models vs playing through Tonex reveals that the NI teach is indeed lightyears ahead of the IKMM one (and the latter absolutely nothing worth writing home about at all - regardless of what everyone else says).

[zvenx]

I hear you. Like I said my memory was capture was part of the process they told us whilst Beta testing, but then again that was 5 or so years ago, so I might be misremembering part of it. Unfortunately all this communication that I no longer have access to, and I couldn't reveal anyway.

I actually have both and use almost Tonex exclusively.. so different strokes for different folks.
rsp

[zerocrossing]

Edge of breakup? What's that?

Tonex captures respond well to input gain. Many heavier patches do nice edge of breakup tones when the gain is turned down.

Thanks to the recent group buy I own them all (I think) and beg to differ - I found in all cases I tried the algorithmic Amplitude counterparts (or similar models) to be significantly better in that regard.
The best explanation I can come up with for all the Tonex praise: "faith can move mountains."

Regarding Paradise, Ruby and Woodrow are particularly good at edge of breakup tones.

Yes, they're okay - the insane amount of CPU they use to achieve that however is not - but if the emperor needs new clothes he needs new clothes, regardless of their price.

I agree. Regardless as to what you think of the UAD models, they're awfully CPU intensive. Even if I did a rigorous a/b and decided they were better, they would have to be substituted for something lighter during the recording phase.

[zerocrossing]

I hear you. Like I said my memory was capture was part of the process they told us whilst Beta testing, but then again that was 5 or so years ago, so I might be misremembering part of it. Unfortunately all this communication that I no longer have access to, and I couldn't reveal anyway.

I actually have both and use almost Tonex exclusively.. so different strokes for different folks.
rsp

Tonex can sound spectacular... and craptacular. Garbage in, garbage out, as they say. I was messing about with different alternatives to Lion and found a Tonex model called Plæxi Driver (nice ligature!) that's based on a Marshall Super Lead MKII that sounded quite similar and beautiful. Sometimes I am a bit amazed at how some captures sound terrible, but then again, I checked out a Neural DSP model (Misha Mansoor) of some metal dude's wet dream and I thought it was the most annoying sound I've ever heard, and I love high gain sounds. So, to each their own. I know Neural DSP can sound great, so this must just be what someone's idea of great is.

I think the problem we have is that almost everything is both great, and also requires a bit of finesse to get sounding like you like it. I know I praise Amplitube a lot, but while I do think it sounds great, a lot of what I like about it is how it is as a total package. Just the room convolution is fantastic, and I've never found anything as good as it on any other amp simulation. It can run stereo IRs, which if you like that sort of thing, is an amazing sound, plus you can run up to 3 paths. Not everything in it is amazing, but most of it is good and some of it is great. I've also used it the longest, so I don't really have to think about how to get what I'm after out of it.

I'm going to skip Paradise Guitar Studio, mainly because I have most of what's in it in some other form and it's a bit crazy on the ol' CPU.

[abrasivesounds]

I bought their amp sims, and think they are more than useable. However, I prefer tonex with amalgam audio captures now.
This really only interests me as an effects bundle. I suppose something akin to soundtoy's effects rack? IDK.
I have their space echo / ce-1 clone, and both sound great. UAD is good at modeling, so I imagine the effects are decent. The videos that are out mostly focus on the amps, not effects. Would like a good walkthrough on those, individuaslly

[Uncle E]

Thanks to the recent group buy I own them all (I think) and beg to differ - I found in all cases I tried the algorithmic Amplitude counterparts (or similar models) to be significantly better in that regard.
The best explanation I can come up with for all the Tonex praise: "faith can move mountains."

Satriani, ODS, and Hyde Street have a bunch of captures that clean up nicely. The captures I've made of Fender amps, including Vibrolux, Deluxe, Princeton, Vibro-King, Twin, and a Low Powered Tweed Twin, are all on the edge of breakup.

Yes, they're okay - the insane amount of CPU they use to achieve that however is not - but if the emperor needs new clothes he needs new clothes, regardless of their price.

The allegory doesn't apply. No one here is claiming that they're low CPU.