Hacker News Reader

Comment by 52-6F-62 | Hacker News Reader

There _was_ a license for stereo audio. See: Blumlein and EMI, Fantasound, etc

It made a lot more sense back in the analog days, when figuring out how to encode and decode stereo audio on an LP was non-trivial.

But things like 5.1 audio or HDR or spatial audio aren't that much more than adding a bunch of extra channels/bits to a stream, defining relative power levels, and the signal strength follows a curve, and oh there's some positional metadata.

The heavy lifting is done by compression algorithms which deserve to be patented because they do genuinely non-obvious stuff. Just like the way Dolby got digital audio onto a filmstrip was similarly clever.

But stuff like 5.1 surround sound... it's just channels, man. In the digital world, it seems like it should be awfully easy to design an open standard.

"Seems like", and the reality are often different. That's some of the ingenious nature of these inventions—they seem like they should be obvious and easy. And yet they aren't. Not at first. It took a heavy amount of investment, organization, and talent to get to the point of stereophonic sound alone.

> It took a heavy amount of investment, organization, and talent to get to the point of stereophonic sound alone.

Sure, because that adds a ton of new complexity!

Going from 2 to 3+ in a digital format does not add complexity.

Is certainly can add complexity when you consider bandwidth and/or processing constraints.

If you figure out a particularly clever way to save on those, sure.

But the baseline of "okay, compressed audio isn't very demanding, throw 3x as much bandwidth and processing at it" does not add meaningful complexity.

Yup. Joint encoding [1] is really the main thing, but that's something that codecs already do, ever since MP3 with stereo music.

The overall point remains: multichannel open container formats exist, and open audio codecs exist. An open standard for 5.1 surround sound, for example, seems like a relatively straightforward combination of the two. I'm not saying you can do it overnight, but compared to other open-source efforts, it's tiny.

[1] https://en.wikipedia.org/wiki/Joint_encoding

Multichannel joint encoding doesn't sound trivial to me.

Hmm, it's not:

https://patents.google.com/patent/JPH1051313A/en

Ugh. That just looks like someone's attempt to optimise some aspect, rather than anything fundamentally needed for the concept to work.

That's like saying "we should all drive hydrogen cars because you can just replace all the existing petroleum infrastructure with hydrogen infrastructure." Yes, but you have to execute on that. Any practical use of digital multichannel audio must consider bandwidth and decoding power as constraints.

When we're looking at video content, the audio processing for stereo is a tiny fraction of the bandwidth and power. Tripling that effort barely makes a difference.

In terms of petroleum infrastructure, it's like slightly adjusting the ratios at the refinery.

> Going from 2 to 3+ in a digital format does not add complexity.

What adds complexity is determining how many channels to use and what to put through them.

That is the important part now.

Have you worked on multi channel encoding/decoding or mixed a multi-channel project?

It’s not just channels. If you want just channels, you can use 5.1 PCM no problem.

Ambisonics has existed since the 1970s. It is license-free, and doesn't define a speaker layout (which is one of the reasons why it is not widely used).

> The heavy lifting is done by compression algorithms which deserve to be patented because they do genuinely non-obvious stuff. Just like the way Dolby got digital audio onto a filmstrip was similarly clever.

Or... we could have governments begin funding universities like they did in the past, and the research would be available for all?

Seriously, we have to re-think patents. The amount of money all that rent-seeking crap is costing societies each year is absurd, and not just in payments to trolls, but also stifled progress - think of stuff like e-Ink that's barely affordable.

> Seriously, we have to re-think patents. The amount of money all that rent-seeking crap is costing societies each year is absurd, and not just in payments to trolls, but also stifled progress - think of stuff like e-Ink that's barely affordable.

You think patents is why e-Ink is 'barely affordable'? Could you elaborate on what data you used to form that conclusion? A simple question, lets say price of a panel versus cost of the raw materials to make it? What margin do you think they're making? Do you know?

Previous discussion of the E-ink patent situation:

https://news.ycombinator.com/item?id=26143779

That looks like a non-answer to me. That post doesn't give any verifiable facts. Repeating it just shows a lack of desire for accurate information.

AFAIK, that was the situation at least in 2016, per media articles back at the time [1]:

> And E Ink, the company, has such a patent moat that it has acted as a monopoly, which Behzadi says has kept prices perhaps too high. But E Ink lost a big patent fight in 2015, and the market could expand soon.

The fact that almost everyone wants e-Ink technology, but e-Ink prices still are still really high [2][3] leads me to believe that either there are still patents at play that prevent competition from rising up, or that the competition hasn't managed to catch up for some other reason. It might also be the case that all of this is simply due to the aftershock of COVID supply chain disruptions, in any case I haven't found a better explanation yet.

[1] https://www.wired.com/2016/05/get-ready-world-covered-electr...

[2] https://www.digikey.de/de/products/detail/e-ink-corporation/...

[3] https://goodereader.com/blog/e-paper/e-ink-has-gallery-plus-...

> The fact that almost everyone wants e-Ink technology, but e-Ink prices still are still really high [2][3] leads me to believe that either there are still patents at play that prevent competition from rising up, or that the competition hasn't managed to catch up for some other reason.

I disagree. It leads me to believe the underlying technology, electrophoretics isn't capable of achieving the volume scaling and update speeds that would make it achieve mass market pricing. Also the links you provided, do not substantiate the main thesis that's being asserted, ie "there are still patents at play that prevent competition from rising up". Further, the quotation you provided was from Behzadi who was a kickstarter guy who failed to deliver the product he promised and then proceeded to blame everyone else except himself.

There's a simple question you can ask to prove this to yourself. Ask everyone who is making this claim, which specific patent is blocking them and how exactly it blocks their idea. You'll instantly realize the people making these claims are not actual display engineers with knowledge of electrophoretics. Typcially, at best, they're bullshitters trying to sound clever or at worst like Behzadi, scammers who are trying to hide having overpromised and then misspent other people's money.

What's wrong with 5.1 FLAC?

Imagine my shock when I found out I have been paying royalties to Doug Stereo for decades.

Stupid Doug.

Cleaver Doug!

Stupid deltarholamda!

Na. Just kidding I'm sure deltarholamda is fucking smart!

Reply on news.ycombinator.com