KVR Audio

aciddose · Post by **aciddose** » Thu Sep 14, 2017 9:15 pm

Architeuthis wrote:
aciddose wrote:This is why the vast majority of claimed "analog models" are in fact nowhere near what could be considered anything even remotely like any particular circuit. They are however reasonable (albeit incredibly inefficient and impractical) software synthesizers.
I have found that the simplest equations and signal routings produce the most analog/warm/natural/acoustic/interesting sounds and behaviors. That is why I say "analog modelling" could be done very cheaply. I have little knowledge on how analog synths work, but I think I know the fundamental underlying behavior, which drives all my ideas.

Well, that's laughable.

If it were true either:

You'd be able to impress customers (become one of the top software synthesizers.)
You'd be able to impress connoisseurs (not so successful, but highly endorsed.)

To impress customers you need to be a "better" product for the common consumer. So appealing to people with a lot of knowledge of analog synthesizers is not a good aim. You want to aim to appeal to the common notion of "analog synthesizer" while impressing those who hold such misguided beliefs. This means great artwork, ease of use (great presets, leading trends in modern music genres) and over-all "great sounding" without regard to accuracy.

To impress connoisseurs you'll need to produce a more accurate simulation of electronic circuits than leading edge simulators are capable of and become indistinguishable and indeed "better" than actual analog synthesizer circuits. You'll need to be so good that people will be selling their mini-moog or jupiter-8 because those devices seem damn near useless compared to the software.

If you succeed at either of these I'll be more surprised than if you were to bend over and shoot lightning bolts straight out of your ass.

nonnaci · Post by **nonnaci** » Thu Sep 14, 2017 10:52 pm

I can see how analog would appeal to old money but the younger generation grew up with soft-synths and a steady devaluation of software towards nearly free or free + advertising. The disparity is plain obvious in the lop-sided product pricing for say PC/Mac v.s. iOS/Android platforms. Time to hang my hopes on AR/VR

stratum · Post by **stratum** » Thu Sep 14, 2017 11:02 pm

I can see how analog would appeal to old money but the younger generation grew up with soft-synths and a steady devaluation of software towards nearly free or free + advertising.

Once the generation who knows "analog modelling" actually means "we have fixed the bugs in our dsp software" passes away, nobody will be able to figure out what the debate was about.

Architeuthis · Post by **Architeuthis** » Fri Sep 15, 2017 4:17 pm

aciddose wrote:Well, that's laughable.

What is, my experience? That would be a strange statement.

fmr · Post by **fmr** » Fri Sep 15, 2017 4:26 pm

stratum wrote:
I can see how analog would appeal to old money but the younger generation grew up with soft-synths and a steady devaluation of software towards nearly free or free + advertising.
Once the generation who knows "analog modelling" actually means "we have fixed the bugs in our dsp software" passes away, nobody will be able to figure out what the debate was about.

nonnaci · Post by **nonnaci** » Fri Sep 15, 2017 4:34 pm

stratum wrote:
I can see how analog would appeal to old money but the younger generation grew up with soft-synths and a steady devaluation of software towards nearly free or free + advertising.
Once the generation who knows "analog modelling" actually means "we have fixed the bugs in our dsp software" passes away, nobody will be able to figure out what the debate was about.

But analog bugs can sound good

unlike a certain blue screen.

stratum · Post by **stratum** » Fri Sep 15, 2017 4:56 pm

But analog bugs can sound good unlike a certain blue screen.

But dsp bugs sound strange, that's why "analog modeling" means fixing bugs in the code (or unwanted consequences of the design)

nonnaci · Post by **nonnaci** » Fri Sep 15, 2017 4:59 pm

stratum wrote:
But analog bugs can sound good unlike a certain blue screen.
But dsp bugs sound strange, that's why "analog modeling" means fixing bugs in the code (or unwanted consequences of the design)

Ah the final frontier. DSP sounding more analog than analog.

Architeuthis · Post by **Architeuthis** » Fri Sep 15, 2017 5:17 pm

nonnaci wrote:Ah the final frontier. DSP sounding more analog than analog.

That's not a joke to me. I believe that's possible and I hope to prove it.

stratum · Post by **stratum** » Fri Sep 15, 2017 5:19 pm

That's not a joke to me. I believe that's possible and I hope to prove it.

Your point that noise can be musical is valid and well known, it's just that it doesn't happen naturally with dsp code. In fact it doesn't even happen naturally with many analog circuits. Many solid state amplifiers sound bad when overdriven, for example. There is good sounding noise and bad sounding noise.

Architeuthis · Post by **Architeuthis** » Fri Sep 15, 2017 5:25 pm

Stratum I agree that there's good and bad noise with analog or digital, I know that most people don't get that. But I don't agree that it doesn't happen naturally with DSP code. I argue that developers are not using the right algorithms. If you want natural sound, use "natural" algorithms. Devs don't get that. They just do what the next guy did.

I should define "more analog than analog". To me, that means the digital synth sounds just as good as the analog one (maybe not a replica, but equally as warm, quirky, whatever") AAAND no disadvantages of analog AAAND the synth is x10 easier to use because it's *difficult* to make it sound bad (by smart control scheme and signal routing by me the developer, essentially, creating macro controls that do a lot more than the label, under the hood, controls that assume what the user REALLY wants, like for example, an envelope decay knob that changes the shape of the entire envelope to compensate/keep the envelope shape natural... I mean, long tangent... when a user changes the decay time, the user really wants to make it more plucky, not just change a stupid timing factor, that's one of my major frustrations with commercial synths, no thought put into controls to keep the sound even/natural).

cron · Post by **cron** » Fri Sep 15, 2017 5:25 pm

nonnaci wrote:
stratum wrote:
But analog bugs can sound good unlike a certain blue screen.
But dsp bugs sound strange, that's why "analog modeling" means fixing bugs in the code (or unwanted consequences of the design)
Ah the final frontier. DSP sounding more analog than analog.

I'd argue it already does! A lot of plug-ins offer settings that go way shittier than analog ever sounded. For instance, tape sims offering a 200% wow and flutter setting plus absurdly high noise settings and the like. Hyperreal analog sound that taps into our memories, exaggerating the defects we actually heard back then to a comical degree.

Architeuthis · Post by **Architeuthis** » Fri Sep 15, 2017 5:30 pm

cron wrote:
nonnaci wrote:Ah the final frontier. DSP sounding more analog than analog.
I'd argue it already does!

Finally, someone of sound mind.

aciddose · Post by **aciddose** » Sat Sep 16, 2017 1:41 pm

Yes and I believe it's more nameste than nameste.

You can start to believe anything if you redefine what words mean and make them meaningless.

IncarnateX · Post by **IncarnateX** » Sat Sep 16, 2017 2:12 pm

The kind of detail that goes into analog modeling these days seems like overkill when we consider that people fail even the simplest of blind tests whenever they are posted.

And if the sound should really be as magical as detailed analog modeling presuppose, then you can achieve it very cheaply these days where analog synths in all sizes thrown at the market.

Incredible that the hype goes on despite these facts.

Analog modelling