Any science to explain the “weight” or “3D depth” of hardware audio vs software that some people claim?

[mystran]

Either way, the point I'm trying to make here is that going by some vague notions of "feels good" descriptions is not terribly helpful.

Except in the fact that, hopefully, we’re all artists and ultimately how we feel is everything. I agree with what your are saying, but at some point we have to acknowledge that we’re not scientifically constructing music, even though some of us have a science background or scientific knowledge.

This is the DSP development forum and DSP development is very much about science, even when your end goal is to make it "feel good."

[BONES]

I think in an analog studio you have hundreds of different devices that each add their own noise and sonic footprint. When I say “noise,” I’m not just talking about broadband aural chaos. It’s the noise of things like an analog filter not performing exactly the same over time. The drift of an audio rate VCO. How an analog compressor behaves inconsistently with every passing moment. Even if the effects of each thing are tiny, when put together they become the grain of the paper or the weave of the canvas.

Yes, the thing you slather paint over in order to get rid of it so it doesn't ruin your piece or distract from your actual work.

[toine6]

.........
[Warning: I’m pulling the following out of my ass.]

[potentialB.S.]I suspect that noise is sort of the aural equivalent to a visual artists canvas, or film grain. If you paint digitally or use acrylics on glass, you don’t have this and you can get what visual artists sometimes call a hyper-realistic effect. Something that looks so real it starts looking fake. The noise of a heavy tooth paper can really add something to a drawing and even digital paint programs have ways of mimicking this. The look of old film stock is aped even when the “filming” is done via digital video. Maybe this noise does something physical to our actual nerve endings, maybe it’s more of a psychological thing. Something like kinetic dithering? When you walk though a room you feel the air as you pass though it, but you also don’t. Your consciousness filters that data out within a certain parameter range, but is your subconscious using it to help define your position in space? I notice my daughter loves the feeling of her baby blanket. She calls it “Softy-Softy” but it’s not really soft at all. It’s kind of a rough hewn muslin fabric. After briefly losing it, I bought some back ups, but she rejected them even though they felt softer to me (to her too). After a lot of questioning I figured out that she loved how it had become more pilled up and worn. The new ones were too perfect. The decay of her OG Softy-Softy gives her comfort. Maybe there’s something inherent to our psyche that just craves a level of decay and patina. [/potentialB.S.]

Yes, that's kind of my feeling too. Film grain often adds a cinematic, larger than life feel. Today, so many films and shows looks super clear and realistic, but it takes the magic out of the experience somehow. I like a little grit, grime, distortion, wobble, smear, and clickity clack sometimes. Maybe it allows my imagination to be active more, and not feel so bored. It's a little like walking around with rose colored glasses for a couple hours, your mind is stimulated but kind of adjusts to the effect until you take them back off again. I'm the kind of person that will be impressed by the skill of someone who can paint ultra-realistic paintings, but feel like they might as well have just taken a picture.

A year ago I was in Berlin Germany, my first time outside of the USA. I had seen the typical old, grainy, over-contrasted black & white film clips of Berlin back to the 30's and up, and it kind made it seem like that was filmed 500 years ago or something. Recently, I stumbled upon some great colorized footage from around 1900, and the frame rate and speed is also very good, so it's not like some sped up Charlie Chaplin film. Even though people dressed different, and there were horses and buggies, the first automobiles, it almost seems like I could go walk the streets with them and it wouldn't be all that strange, especially as the building's have not changed that much over the years. While that doesn't have a lot to do with noise and grain, it certainly showed how the closer to reality you get the less "otherworldly" it seems.

Even though modern technology is so much more advanced than the 70's & 80's, there was something about Superman, Tron, Blade Runner, Willy Wonka, 2001: A Space Odyssey, and the Star Wars movies that I think lent them more character and charm than today's offerings. Today's stuff is pretty incredible as far as clarity, frame rate, and fx, but it's missing that charm that the old movies had.

That said, it can be replicated pretty well. The guy who made searching for Sugarman filmed it with an iphone, and used an app to make some of the clips seem older. I was convinced and never questioned if he was using old film or not. We have the tools to replicate the old charm if we want it, but film and music maker's end up on an endless quest for clarity and sterilization which should not be the end goal. I guess I like a touch of surrealism and I think there are a lot of other people that do to.

[soundmodel]

There's, I think, strong subjective evidence to suggest that the phenomenon is true.

If it's subjective, it's not evidence, it's hokum.

Not necessarily. Without subjective with certain requirements for consistency and repeatability etc., psychology couldn't be evidential.

Thus as I wrote:

-subjective experience may point towards a phenomenon
-in order to understand whether it's a real phenomenon or a bias, one has to conduct a test that minimizes e.g. bias, lying etc.
-so e.g. ABX test

I believe the hw weight is real, but I don't certainly know what creates it.

[puffin]

Hardware weight or analog weight? Sequential Pro 3 is hardware, contains a digital oscillator and I love it. Arturia Microfreak is hardware that is ... I should not use such words not what I was expecting.

[perpetual3]

Nonlinearities, phase offsets, random discrepancies in l/r channel frequency profiles, as well as the physics behind what happens when you drive a hardware channel, changing slew rates. Lots of phenomena.

[earlevel]

The bottom line is that analog has many shortcomings. Of the different solutions designed over decades, the ones with flaws we liked survived, and the ones we didn't faded from memory. Sometimes, blatant flaws (the flaws a guitar amp adds when abused), sometimes subtle (mic pre's); sometime mental because that hefty thing looks so cool in the rack it must be doing something good, sometime because it was used on famous tracks we like a lot and dammit we're going to give ourselves the best shot. And most importantly, the studio clients want to walk in an see the good stuff.

Digital is largely lacking those flaws, so whereas the analog goal was to try to be as accurate as possible and live with the flaws (and if lucky, like them), with digital we take a shortcut and try to emulate the analog flaws we like.

[zerocrossing]

I think in an analog studio you have hundreds of different devices that each add their own noise and sonic footprint. When I say “noise,” I’m not just talking about broadband aural chaos. It’s the noise of things like an analog filter not performing exactly the same over time. The drift of an audio rate VCO. How an analog compressor behaves inconsistently with every passing moment. Even if the effects of each thing are tiny, when put together they become the grain of the paper or the weave of the canvas.

Yes, the thing you slather paint over in order to get rid of it so it doesn't ruin your piece or distract from your actual work.

You obviously don’t paint... or have ever seen a painting.

[zerocrossing]

Hardware weight or analog weight? Sequential Pro 3 is hardware, contains a digital oscillator and I love it. Arturia Microfreak is hardware that is ... I should not use such words not what I was expecting.

Hm, I have a Pro 2 and a Microfreak, and I love them both, but as soon as I read this post I thought, “nah, not what I think of as weight.” At least not as “weighty” as I consider my full analogs. The 3 may be a different animal with its VCOs. It can be used as a 100% analog, so as hybrids go, it’s less lopsided as most. I’ve also owned all digital synths and up against similar VSTs... I didn’t sense that either was more weighty. Of course, who knows what I’m blathering on about.

I’m also not really very concerned about it. I often do a bunch of a/b testing with hardware synths when I get them, and what I’m most concerned with is just basic character in any synth that I don’t get from something else. I’ve made more than one patch on Legend that would be in indiscernible from my ATC when using it’s Moog filter. But the ATC has 3 full ADSRs, oscillator sync, two dedicated LFOs, an ARP, 303, and SEM filter to accompany the Moog. Ring mod (from an external source as well). It’s got tricks that no really good analog emulation has (yet.)

Plus... it’s got some weight to the sound, ya know?

[BONES]

Not necessarily. Without subjective with certain requirements for consistency and repeatability etc., psychology couldn't be evidential.

My point, precisely. If ever there was a hoax of a profession...

I believe the hw weight is real, but I don't certainly know what creates it.

Well, you will never be able to know because it isn't actually real.

You obviously don’t paint... or have ever seen a painting.

Not everyone paints with brushes. You won't see the canvas through a painting done with spatulas, for example. The paint itself creates the texture.

[Haptix]

So-o.. if the hw thing is not real, everyone who hears it is just hearing it wrong? Or listening it wrong? Must be, because you cannot measure it, but are you sure that you are not just measuring it wrong?

[matt42]

Firstly there are measurable differences between analog and digital. Signal processing works quite differently in some ways in each domain. Hard clipping will alias in the digital domain and can often be heard audibly and will certainly be measurable. Perhaps linear phase over sampling is used to reduce the aliasing below the noise floor. Then we now have preringing which is also certainly measurable. Could most people hear the difference between a carefully designed digital clipper and its analog equivalent? The only way to know would be to perform double blind testing, or similar.

I read somewhere in the waves literature (and I'm not wanting to say anything either way about the quality of the models) that in their listening tests they had to add noise to the model otherwise the digital version would be identified. My hunch is that this has not so much to do with the quality of the modeling (assuming a not so terrible model), but rather the most identifiable difference to the listener is the presence, or not, of analog noise floor. Double blind testing would of course be needed to know for sure.

Edit - and some analog effects may be harder to model leaving more clues for someone familiar with the equipment to distinguish correctly. I wasnt trying to suggest noise floor is the only discernible difference, just that most of the time for most people and with reasonable modeling that it would be.

[MadDogE134]

those that 'think' they know everything only irritate those of us who do lol

[matt42]

those that 'think' they know everything only irritate those of us who do lol

[BONES]

So-o.. if the hw thing is not real, everyone who hears it is just hearing it wrong? Or listening it wrong? Must be, because you cannot measure it, but are you sure that you are not just measuring it wrong?

It's a well known mental deviation called cognitive bias. Here's how they explain it over at Wikipedia - 'A cognitive bias is a systematic pattern of deviation from norm or rationality in judgment. Individuals create their own "subjective reality" from their perception of the input. An individual's construction of reality, not the objective input, may dictate their behaviour in the world. Thus, cognitive biases may sometimes lead to perceptual distortion, inaccurate judgment, illogical interpretation, or what is broadly called irrationality.'

In this instance it is a mix of several kinds of cognitive bias - declinism, groupthink, sunk cost effect or subjective-expectancy effect but mostly it's just good old truthiness. It's quite interesting to go through the whole list of them and see which ones you are guilty of and how often or deeply you get sucked in.

Firstly there are measurable differences between analog and digital.

We're not talking about analogue and digital, we all know they are different. We are talking about hardware synths versus software synths, about whether something like Modal's Argon 8 will sound better than Serum, just because it's hardware.

I read somewhere in the waves literature (and I'm not wanting to say anything either way about the quality of the models) that in their listening tests they had to add noise to the model otherwise the digital version would be identified.

That is very, very common and sometimes that's all you need to do to convince someone.