I concur. There is some voodoo in those cinematic rooms. I thought 7th heaven was amazing, then tried rooms. Wow.

Relab has a sale right now for their lite version of the 480 for $35. UAD also has their summer sale going on. Previous comment about the UAD AMS Neve DFC might throw the win over to getting the UAD 4 for 3 bundle though since I primarily do media music now.

I wouldn't miss that Relab deal. The developer is a Reverb expert who also coded the IKMultimedia CSR. The Delays on the UAD platform and Relab are the best vintage tools you can get ITB.

glenesis here in 2021.

This month, August, UAD have their 224 and 480L on sale for $139! Immediately upon noticing this, I rang up the 224, and found a $50 voucher living in my UA account. That brought the price if a DSP-accurate Lexicon 224 down to about an even hundred bucks for me after New York State robbed me for sales tax.

Considering I bought a hardware LXP-5 for $139 a week ago, I feel like buying these particular UAD plugs is pretty much a no-brainer.

Sale ends on 8/31/2021.

Cheers y'all!

The state charges 4%; not exactly robbery. Cali, by comparison, is 7.25%.

Anyway, for a legit 480-in-a-box, the cost of admission is intensely low. Imagine the tax on my $16,000 960L!

If we're talking about 480l sales, Reverb Foundry has their very good HD Cart 480l emulation on sale for $79 (regularly $199) until September 6th.

If you own any LiquidSonics plugins, you can add their discount codes too to get it for an even lower price.

So a question comes to my mind on this. I've been reading through various Lexicon threads as I am in the market for a 480 including this one:
Lexicon reverbs: a brief bestiary

And the one from Michael Carnes on the release of the Lexicon PCM plugin bundle. Actually both threads are by him but a bunch of other former Lexicon folk pitch in.

If you haven't read the first one, the bestiary, I highly recommend it. It's a bit of a survey course on Lexicon reverb algorithms as well as history/background.

A question I have for UAD was actually sparked by the statement on their 480 product page. This paragraph specifically:
"Now available exclusively for UAD hardware and UA Audio Interfaces, the UAD Lexicon 480L Digital Reverb and Effects plug-in is the world's only Lexicon-endorsed emulation of this classic studio reverb"

World's only Lexicon-endorsed emulation...

Does this mean UAD have licensed the source code from Lexicon/Harman?

No.

Okay. So what’s the endorsement for then?

The only plugin then that has the original lexicon algorithms that were in the 480 is the PCM bundle. Unless I’m missing something?

There is something special on this Relab 480. Love the way it sounds. Got soooo much reverb plugins right now. I bought this 480 because of THIS sound. Like a unique signature that is hard to reproduce with others reverbs. Got 7th Heaven & Cinematic Rooms: absolutely fabulous. But it sounds modern. Relab 480 for better or worst sounds 80s ! And that is great when needed

No currently commercially available plugin (or hardware for that matter) has the original algorithms that were in the Lexicon 480L.

The PCM bundle has the PCM96 algos.
The PCM96 is the successor to the PCM91.
Like its predesessor the PCM96 does not sound like the classic Griesinger line of Lexicon reverbs.
One could argue it mimics them.

No idea what this endorsement deal is about.

Yes but the PCM96 algorithms are direct descendants of the 224 and 480 algorithms according to Carnes. At least Concert Hall (224) and Random Hall (480). The entire history of Lexicon seems to revolve around just a few algorithms (less than 10) and Carnes didn't reinvent anything to do with the the classics for the PCM96. He called out what was actually new in the PCM96 vs what was ported. He did say that algorithms are being improved upon using an iterative approach (refinement over time basically where it makes sense).

Regarding the classics he said he ported them from existing source code. I assume he used "ported" because you're moving from fixed point and assembly language to floating point and high level languages like C+ etc.
Actually that's probably where all the bodies are buried in the whole "doesn't sound like a 480" debate. He mentions that most of what people associate with that are math artifacts as related to dsp. But he has access to the original source code (or rather had access; he's no longer there).

To hear him tell it, those are the algorithms that were in the classics.

- I think he said it's a direct port. Whatever is modern in them is to do with floating point and high level C vs assembly in the old machines.

- yeah it would be nice if UAD would chime in. Further reading of their page suggests they tried to reverse engineer the 480 (firmware/hardware analysis) but you'll not get the source code that way and you'll always end up with an approximation. Only Lexicon has the actual algorithms though and unless you have that you're in the dark kind of. Bit like blue meth.

It is my understanding that the old 224->480L source codes where "lost" when Lexicon was bought by Harman.

Also I am not sure a "port" of the old fixed point logic design source code to a floating point dsp would be possible as such. If this required in depth rewriting of the code, it could be argued that it would leave the realm of what can be called a port and will most likely impact the sound of the reverb. But I am out of my depth now so I'll leave this as an assumption only.

I'd be curious to know the technical reason for the difference I am hearing: When using different Lexicon reverbs my ears repeatedly and consistently tell me that the PCM 90/91/96/PCM Bundle stem from another design family than the 224/224X(L)/480L/960L.

This is true for the 224 not for the 480. This is in the "bestiary" thread actually.


- not a problem in terms of what's possible. If you have access to the source code it's merely a matter of implementing it. It's just an engineering problem really. Possible. I should say that typically algorithms and programs are necessarily the same thing. Algorithms are often high level descriptions. That's the heart of the matter anyway. You can implement an algorithm in the language of your choosing if you have access to it.

This is also in the "bestiary" thread. He reckons math artifacts in DSP when going from fixed point to floating etc.
Same family. As in they are not new/different algorithms. Saying that though, Carnes did say that the algorithms both stay the same and evolve.

The right to slap the name Lexicon on the gui.

Thanks for the link!

Can you point me to it?
I browsed the "bestiary" thread and could not find a statement claiming the 480L source code is still available.

I remember reading about it being lost a long time ago. But can't find a quote for you. Maybe it wasn't on GS.

No. It is the 480L code that was lost. I found the quote:

Also reading this it sounds like the 224/PCM70 Random Hall/Concert Hall either were if not lost, at least not ported to the PCM96/Native Plugin platform:

To me all this reads like: None of the later Lexicon creations Like PCM90/91/96/Native Plugin contain ported algorithms of the classic 224/480/960 "Griesinger et al." Lexicon processors.
The former contain algorithms written by Mr. Michael Carnes specifically for the new processing platforms.
This would also match the general sonic differences I am hearing when comparing these reverbs.
Two different families of Lexicon reverbs.

Right so Casey says it's the 480 code that's lost and Michael says it was the 224. In that same thread. Maybe it's both then. The algorithm isn't lost though and that's really the key. All these guys know what the dsp is supposed to be doing so it's a matter of implementation. You said the 480 sounds different and the reasons for that are manifold. The random hall algorithm between the units however is fairly consistent.

Michael was saying that a lot of what people ascribe to the 480, e.g. graininess, is to do with math artifacts going from fixed point environments. More so than AD/DA he speculates. But that said I'm sure the hardware is another cooperative condition that accounts for these differences. If we knew the insides of the algorithms then this would be an easier task. To recreate the 480 I mean. But we don't. Lexicon engineers do however. Or at least they did when the PCM96 was developed. The random hall in the PCM96 wasn't invented by Michael at the time of the PCM96 release. It goes back to the 480. He merely reiterated he goes on to say. We all want to know what's in the algorithm. That's what matters. If UAD doesn't have access to that then I would say this is just another in a long line of interpretations of the 480.

Which is quite an asset.

That's what Michael told me. I had some beloved presets I'd made for the 960L and when switching to plugins asked if I could somehow get a shortcut to recreating those 960L presets in the PCM verb, but there was no copy/paste-type shortcut available. We arrived at approximations by matching the 960's parameters, then tweaked a bit further by ear, but they never matched.

Thanks for verifying my assumption

Back in the 90s I bought a PCM90 and a PCM80 and always wondered why their reverbs sound so fundamentally different when the PCM90 was marketed as the serious reverb counterpart to the FX oriented PCM80.

I assume Lexicon back then wanted to become less dependent on David Griesinger's work so they started developing a second family of reverb algorithms from the ground up.
If that was the case it's pretty obvious that Lexicon wouldn't have been too keen to communicate this strategy publicly since the Griesinger algorithms back then where considered to be the core of the must have Lexicon reverb sound.

Same here regarding the 960L. Still haven't found a good soft- or hardware replacement for what it does.

It is not about subtleties or theoretical considerations...

Listen to this. There is nothing else similar I am aware of, not even remotely.

So I just did this test and matched levels and settings between both units. Relab gave me much more of that familiar yummy lexicon feeling when I had a hardware 480l.

Yup, LX480 is scary close. I will be honest, I always felt something went very wrong with the UAD 480L. Five minutes spent with a convolution reverb and IRs of a real 480L, going preset for preset with the UAD, should be required listening for anyone who feels UA nailed this one. And I was always bothered by UA’s Will Shanks’ extremely circumlocutory PR speak (on page 2 of this thread) when it came to the real provenance of the plugin’s algorithms - with no disrespect meant to the team which I’m sure worked very hard on it, I’d suggest that “approximation” is probably the kindest word you could use. Not that Relab has exactly been forthcoming either about his mysterious “black box” reverse engineering of the 480, but the proof is in his pudding.

This is not a dig at UAD by the way - I’ve had a chance to put their 224 plugin against the real thing and it fared very, very well. But the 480 shouldn’t have the Lexicon name on it.

I'm also impressed by the 224. UAD's 480 I've not used, but I have Relab's and like it quite a lot. Can't tell if it's 100% faithful but it's faithful enough for me not to care.

What utter BS

I've only compared UAD and Relab, never had a chance to do a proper comparison with an actual hardware 480L, but the 480L is a digital reverb and Relab claim their version is a sample-accurate copy, so unless they are lying, whatever difference you're hearing between the hardware and the LX480 is simply just what you want to hear.

Also, it doesn't matter which is better because the Lexicon PCM Native Suite, which is based on Lexicon's state-of-the-art PCM 96, sounds much, much better than the old 480L anyway. The 480L is great and has been important historically, but people get overly obsessed with it.