Still loving your Bricasti M7?

I think anybody interested in recording is into high quality reverb (probably the first DSP to blow my mind in the late 90s), and I know there is much love for the Bricasti (the example above sounds very very nice).

Here is what I don't understand though - if it's a digital implementation of reverb then it should be possible to make a 100% perfect copy/implementation of the algorithm using a computer (implemented as a plug-in or dedicated post-processing tool somehow, etc.). The samples ("ones and zeros") don't care how they are summed/multiplied, etc. in the digital domain.

The A/D and D/A specs (jitter, dynamic range, etc.) on the unit are quite decent but only impact the conversion itself, not the reverb algorithm (unless something quite unconventional is happening here) and with that consideration Lynx and other dedicated converters are comparable or better.

So what is it that the Bricasti is doing differently with dedicated digital computation hardware? Is it mainly that they are implementing the algorithm in a dedicated hardware box so they can protect their hard-earned DSP intellectual property? I'd completely understand that.

Otherwise, from an electronics standpoint, I just don't get it. That doesn't imply that the many happy users out there have been "tricked" or something, I'm sure it's earned its reputation. I've read comments like "other software 'verbs just can't touch it!" --> and while I'm not saying I don't believe them, I'm saying as an electrical engineer I don't yet understand what it might be doing that can't be implemented (from a technical standpoint) in a pure software product.

Any insights on this would be appreciated (in case I get tempted to jump on the Bricasti bandwagon) :-)

-Chris

You wonder what the cost is on one of those things…of course, I don’t want to besmirch the guy for getting whatever he can out of it…but when you can get two tricked out MacBook Pros for tha same price or something seemingly as complicated like the Axefx for at least a thousand less…it makes it a hard buy for me. And that’s fine…different strokes.
But I was reminded today comparing the Pulsar 78 with the AS 78…the Pulsar sounds really awesome…but the HW sounds a little more awesomer.

that’s not possible with many reverbs. different CPUs and dsp chips process math differently. Old reverb dsp chips ran in fixed point instead of floating point and were quite primitive. Every operation produced truncation and noise. Unless you write non standard code to Jerry rig a cpu to run those algorithms in the same way as the vintage chip, it will not sound the same. The algorithms were tweaked to sound good on those primitive chips, not on modern CPUs. All of the distortion greatly affects spatial perception and timbre. They would have to write different algorithms on modern cpus to sound similar to the original ones on the primitive chips in terms of depth, width, and timbre as the older reverb. Many of these effects are iconic and the sound of many genres.

This is why the Relab and UAD Lexicon plugins and the Eventide H3000 factory sound a little odd. They cannot physically sound and behave the same. The Bricasti m7 runs on multiple Analog Devices Blackfin chips that are 16/32-bit fixed point. The TC 6000 runs on motorola 56k chips like pro tools tdm. Porting the algorithms to an x64 or ARM chip in standard floating point and having them sound the same is probably impossible. But what I do know? Many Brainworx plugins have fixed point truncation everywhere for no reason.

Now the older units are far far more primitive than these. Many genres’ classic tones depended on those dying units, some of them very cheap like the Alesis Midiverb II and Quadraverb or the Yamaha SPX90 and FX500. The Lexchips are primitive. Many of these algorithms have no modern equivalent or only 1 or 2 that can do something similar and it’s often a plug or a pedal.

Of course other units ran on floating point sharcs like UAD cards and Weiss and their algorithms can easily be ported to native because Sharc chips were 32-bit float and 40-bit float extended precision while modern intel and arm is 32/64 bit.

The best option for these hardware reverbs native that cannot easily be ported is to just buy the hardware or just use something else made for modern CPUs. The Relab Sonsig, LiquidSonics Tai Chi, and Apulsoft Apverb sound fantastic to me. Goodhertz Megaverb replaced all Alesis hardware

Well it's funny you talk about the AS78 at the very end of your message, because the beginning of it, and the question mark about production cost of the Bricasti made me think immediately of something weird that happened to me some time ago that has to do with an AS78 unit.

---> I recently received a Dynax 3 compressor demo (good price, most savage compressor on earth) directly from Olivier Bolling, the funder of AL.SO.
He sent the Dynax 3 in a "pre-sale" AS78 box. The price for the unit was very low. 599 GBP. I wondered what this was about. Maybe this is the production cost for an AS78 ? We can't know for sure. Could be something else.

I seriously don't begrudge anyone for providing a product people want and making a profit. All for it. And manufacturers like AS are already cutting us a fair deal in my estimation. But I'm sure we're paying for the expertise and parts and not paying for a name. Their margins are probably the way all of pro audio should be priced. But again - I'm talking out of my ass. My next day running a supply & demand company will be my first.

I'm interested in using the M7 to use as "room mics" as I am working in a fairly dead project studio at home.

Has anyone explored the difference between sending the close mics to the M7, vs setting up a secondary pair of mics at a bit of a distance, and then sending those to the Bricasti 100% wet to create the room sound?

I've heard some people say that this is a better approach than sending the close mics; but I'm not sure what the reasoning for this is. The close mics have a better signal to noise ratio, than if I set up mics 6 feet back lets say on a quiet acoustic...anyone have insight into this? Wouldn't pre-delay allow me to "simulate" the mics being farther back from the source if needed?

One day when I have as much money as Elon, I want to test this theory. I want to buy everything I ever wanted…Helios console (wings included), vintage 251,47,48,49,67…etc. Then really compare to working tools we have…and I’ll let you know. The hardest part of all this is the Elon part.

here's an old sample of mine using a 2 mics set up slightly OOP with exactly zero processing in the front of my garage, you can do it bro
https%3A//soundcloud.com/tonycamp/u47c12-clones-over-shoulder-stereo-sample-no-processing-at-all%3Futm_source%3Dclipboard%26amp%3Butm_medium%3Dtext%26amp%3Butm_campaign%3Dsocial_sharing

Thanks! I own Seventh Heaven, and sold my M7 when I got it as I felt they were close enough...regretted the decision, so picking up another M7. In a dense mix, the differences aren't perceptible necxesarily, but in a sparse mix, I've found the M7 feels/sits better in the mix. Also I prefer the hardware when tracking; been getting some CPU overload behavior with the plugin. But it's a great sounding plugin!