Hacker News Reader

Comment by jrajav | Hacker News Reader

Most of the time people will argue that vinyl has "warmth" and this makes it subjectively better. This is true, but it's actually because of the pleasing distortion record players add to the music. Being distortion, even if it improves your listening experience (and this depends on genre, though it generally does), it is certainly objectively worse in terms of accurate reproduction.

View on HN

YES! The article says this:

> But a few specialty houses (...) have long advocated for the warmth of analog. “Not that you can’t make good records with digital, but it just isn’t as natural as when you use the original tape,” says Bernie Grundman

which is absurd. "Warmth" is a kind of distortion, and if a signal is distorted then it's less accurate or "natural"!

They can't claim "fidelity" and "warmth" at the same time. It's one or the other.

But if analog is what you knew, that wasn't "distortion". That was "natural", because it was what you grew up with. That was the way things were supposed to be.

So "fidelity" means "fidelity to the way I remember" (and like).

Not "natural", "familiar".

Just like people buying yellow lightbulbs because that's how incandescent bulbs used to always be -- actual natural sunlight isn't that yellow. Good LEDs are more "natural", but less "familiar" to people.

On a high quality record with a high quality setup, you can have both. It's because the distortion is really subtle, and hits all the right harmonics. On bass-heavy music it makes the bass sound really nice. The other parts of "warmth" aren't really in my wheelhouse, but it makes the bass in bass music better

Exactly. If you like that warmth, awesome. There's not a thing wrong with that. It's when people confuse "warm" for "better" that the problems start.

An instrument amplified by vacuum tubes does have a distinct sound, "vinyl" also has its own signature.. magnetic tape is pretty analog too. All of these signature sounds can be digitized/discretized at sampling rates that preserve the vinyl crackle or tube warmth or combinations. To me it seems like marketing and the consumers justification of their susceptibility.

> An instrument amplified by vacuum tubes does have a distinct sound

Only if it’s pushed into distortion.

Guitar players (generally) like tube Amos because of the way they distort compared to transistor amps.

If you are not pushing the amps into distortion, they will only sound different if they’re badly designed.

Not necessarily (well I guess it depends what you mean by "pushed"), but I think tube amps naturally boost the even-numbered harmonics at all volume levels. This sounds good but is distortion.

That’d show up in the Total Harmonic Distortion figure for the amp, and for any halfway hifi quality oriented amp it’ll be well below 0.1%. Perhaps “golden ears” can hear that? I’m not convinced. Most hifi audio amps typically run at well down into the single digit percentage of their rated output at ordinary listening levels in most systems, so they’re an order of magnitude or two away from clipping. If the feedback circuit and the output devices (transistors or tubes) can’t keep the response perfectly linear the thd down under 0.1% across the entire audio frequency band in that power range, it’s just badly designed.

Can't the warmth just be reproduced in a digital player anyway?

Yup. Boost the mids on your EQ and you're halfway there.

What's the other half? Magic in the amp-circuit?

Thousand dollar rocks and crystals to place around your room.

Crystals? I only use wooden resonators, personally.

https://www.analogueseduction.net/record-clamps/SHUNMOOKprod...

Well yes but that's not actually what the intent was.

Though I would expect it'd be more of an eq profile thing.

Yes, the "warmth" generated by from literally any possible analog circuit can be reproduced digitally.

Physics doesn't care about people's beliefs, but sadly many people also don't care about physics.

In theory.

In practice, if I give you a random analog box, transforming it into an accurate DSP algorithm it's an extremely difficult problem.

Which is why people instead of doing that, they recreate the "idea" (reverb, delay, ...) digitally, but it's not a copy.

Although for some reasons 2021 was a great year for distortion/saturation plugins.

Any recommendations?

The hotly debated P42 https://www.pulsarmodular.com/product/p42-climax-line-amp/

The popular Kelvin Tone Shaper https://www.toneprojects.com/kelvin-tone-shaper.html

Maybe you want tape? London Acoustics Tapei https://www.londonacoustics.com/product/taipei-studio-tape-r...

As for freeware, the comeback of Variety of Sound with https://varietyofsound.wordpress.com/2021/11/30/tesslase-mki... or the less subtle https://kitplugins.com/pages/burier-free

This is just a part of (good) saturation plugins from 2021...

Not from 2021, but in May 2020 Fabfilter’s update to Saturn 2 is great, and the original was already a fantastic Saturation plugin I’ve used a lot.

I've heard the recent Kemper Profiler uses advanced analysis to generate fairly convincing simulations of analog equipment, given its response to specially designed test signals. I haven't tested it myself though.

You can also bake analog sounding warmth into your hi-res digital master and it's done all the time. There are a wide variety of different digital mastering plug-ins which do remarkably sophisticated analog modeling.

Ultimately, this boils down to signal which vibrates speakers. As a thought experiment, if you use some kind of theoretically perfect surface-sampling laser to capture every movement of the speaker surface at sufficient frequency and fidelity to reproduce all of the information in the original signal and speaker surface vibrations (ala Shannon->Nyquist), then a digital playback of that signal which vibrates the same speaker surface identically will sound exactly the same.

Individuals can prefer different sonic characteristics encoded in an output but that's an aesthetic choice. The entire signal chain creating that output is the result of creative and technical choices. It goes from guitar string to studio acoustics to microphone to mixing board to outboard processing gear to recording medium to duplication to distribution to playback to amplifier to speakers to room acoustics to human ear. Most of those elements significantly color the sound. Yes, mistakes can be made in the downstream signal chain which diverge from the creative intent. However, those mistakes are exceptions, and not inevitable. Done correctly, there's no reason a digital step in the chain shouldn't be completely undetectable.

> Individuals can prefer different sonic characteristics encoded in an output but that's an aesthetic choice.

I find that listening to one of three different sources (earbuds, gaming headset which is EQ'd via open source software, and inexpensive-but-fancy headphones) causes me to adjust to that particular set of headphones' sound. It seems to be like the equivalent of our brain continuously auto-white-balancing our vision.

Absolutely, personal sonic character preferences can vary per content, listening context and the play back device (plus play back signal chain like DACs, pre-amps etc). And it's not just personal taste, most human's bio-based input 'hardware' varies enough to matter, especially for those of us who are out of warranty (ie >40 yrs old). There are even interactive apps which will give you a quick hearing self-test and build an EQ profile based on your individual response curves. I use one bundled into Android and it definitely improves my listening experience.

My son told me that some hip-hop recordings incorporate digitally generated tics and scratches to simulate vinyl.

That reminds me of a thread here from a few weeks ago about how somebody was training an ML model to add noise to video, so that the noise-free version of the video can be compressed better but then the noise can be added at playback to maintain the authentic feel of old video.

Sure you can, but then you can't call it 'analog' anymore. But you can reproduce the distort...I mean, the "warmth" with digital filters or whatever and it will pass any blind test.

The finding some years back was that the harmonic distortion in tube circuits was even harmonics and in solid state was odd harmonics (or maybe vice versa). This could cause tube electronics to sound better than equally good (by measure of thd %) solid state electronics.

This says nothing about the difference between analog and digital recording media unless you think that disc lovers shun solid state electronics or something like that. Do they?

There wasn't much digital sound around when McLuhan said that "The medium is the message." I think that's the substance of this whole brouhaha. What's wrong with rejecting digital media when one deplores the inundating tide of changes in society's tempo, cultural content, aesthetics, and forms of dis-function that have been carried so far and wide by digital media? Isn't change always impugned?

I still think that "warmth" is a way to describe not actually hearing the separate digital samples subconsciously, and just hearing the sounds gliss into each other. I also think it's an anachronism that comes from way back when the digital sample rate was often a lot lower (and people would insist that their live digital effects sounded as good as analog, when you could sometimes even consciously hear the samples jumping into each other.) Other than that, it seems like something that people project onto music with bassy reverb.

You cannot hear samples, or the "steps" between them. This is a silly idea that comes from people seeing pictures of the waveform as it is digitally captured and imagining they can hear it.

You cannot, it is impossible.

Even with an unrealistically low sample rate (say 500Hz instead of 44.1kH), this will not sound like 500 beats per second as the "steps" occur, but rather will just sound extraordinarily dull, as the Nyquist frequency has dropped to 1kHz, a fairly low tone, and all info above that has been lowpassed away.

When the signal is changed to analog by the DAC in order to come out of a speaker, the "steps" become smooth curves like any analog waveform. You cannot hear "digital steps" because they cannot be audible - sound is vibration, and it cannot be made into steps in the analog realm.

as the other poster noted, it’s not possible to hear each sample individually.

what is bassy reverb?

Reminds me of the tube vs transistor amplifiers discussions where tube amplifiers are supposed to have a warmer sound.

Reply on news.ycombinator.com