On 9 Apr 2024 00:08:02 GMT, rbowman wrote:

>
> Well, yeah... Waveform analysis doesn't mean anything if you can't hear
> the difference. Human neurophysiology is not unlike an ADC.
>

No, it is not -- at least at the total level.

The human brain processes sound at several levels, the lowest
level being the inner ear. The next levels are the "reptilian"
brain structures and then we move on up to the cerebral cortex.

It is that final level, the cerebral cortex that makes all
the difference.

But an objective test is just fine. Like JPG or other image
compression, audio compression only strips out the highest,
and supposedly humanly inaudible, frequencies.

How well this is done can be easily measured.

MP3 compression has many parameters that a user can control.
Unfortunately, most producers of MP3 files blindly use Microslop
software and know nothing of these parameters. Consequently,
their files sound like shit.

Farley Flud <ff@linux.rocks> wrote at 19:40 this Tuesday (GMT):
> On 9 Apr 2024 00:08:02 GMT, rbowman wrote:
>
>>
>> Well, yeah... Waveform analysis doesn't mean anything if you can't hear
>> the difference. Human neurophysiology is not unlike an ADC.
>>
>
> No, it is not -- at least at the total level.
>
> The human brain processes sound at several levels, the lowest
> level being the inner ear. The next levels are the "reptilian"
> brain structures and then we move on up to the cerebral cortex.
>
> It is that final level, the cerebral cortex that makes all
> the difference.
>
> But an objective test is just fine. Like JPG or other image
> compression, audio compression only strips out the highest,
> and supposedly humanly inaudible, frequencies.
>
> How well this is done can be easily measured.
>
> MP3 compression has many parameters that a user can control.
> Unfortunately, most producers of MP3 files blindly use Microslop
> software and know nothing of these parameters. Consequently,
> their files sound like shit.

You could also do lossless compression.
--
user <candycane> is generated from /dev/urandom

Physfitfreak wrote:

>To be me?... If you had to be anything closer to me than you are, you
>had hanged yourself in your mother's bedroom before puberty.

Yeah, I probably would have. 8)

--
"Windows Phone is far more successful in its market than Linux desktop
is in its market." - some dumb fsck

On Tue, 9 Apr 2024 20:20:07 -0000 (UTC), candycanearter07 wrote:

>
> You could also do lossless compression.
>

True. But the discussion is about MP3 files.

Lossless compression is usually done through FLAC, which is yet
another FLOSS success.

I play my FLAC files thusly:

flac -d -c file.flac | aplay

What could be simpler?

A lot of music, for example the chuckleheads on YouTube, will
distribute music as VORBIS or OGG.

GNU/Linux can handle all these cases as well.

rbowman wrote this copyrighted missive and expects royalties:

> On Tue, 9 Apr 2024 08:33:01 -0400, Chris Ahlstrom wrote:
>
>> I've seen photos and videos of bicyclists fleeing from a bear.
>
> There is a canyon I hike that eventually leads to a ridgeline that is
> popular with mountain bikers. I've seen bears in the lower part of the
> canyon and I've also seen Fish & Wildlife dumping problem bears they've
> trapped up on the ridge.
>
> I've been waiting for a biker to emerge from the canyon with a bear on the
> handlebars.
>
> Considering the number of black bears in this area eventful encounters are
> remarkably few. In my experience the bear either leaves rapidly, trees, or
> ignores you. I don't recommend it but I have passed between a mother bear
> and her cubs that were treed on the other side of the trail. "Cubs? What
> cubs? I haven't seen any cubs Miz Bear."
>
> Even stranger I was driving back to the campsite from a trailhead at Kings
> Canyon when I saw a bear treed next to the road. I stopped to take a photo
> like a good tourist. The bear shimmied down the tree and walked off. Then
> the cub, who I'd not noticed, climbed down and followed her. The question
> is if she wasn't bothered by a car and a person taking pictures why had
> they treed in the first place?
>
> Kings is awash in bears. Lots of gourmet food and they will disassemble a
> Prius to get to it.

:-D Taking can-opener claws to the can!

> Bear safes are provided at each campsite with
> instructions to put anything that has a scent in the safes rather than
> leaving it in the car.

The Yogi Bear cartoons never mentioned that stuff. Just "pic-a-nic baskets".

--
The difference between the right word and the almost right word is the
difference between lightning and the lightning bug.
-- Mark Twain

rbowman wrote this copyrighted missive and expects royalties:

> On Tue, 9 Apr 2024 08:35:00 -0400, Chris Ahlstrom wrote:
>
>> Who the hell are you? Gustav Fechner?
>>
>> In my grad schools days I had a Fechner t-shirt.
>
> Sort of. My degree is in psychology, the experimental sort. The department
> head was an old school behaviorist and the rest of the professors were
> hard science types. 20 years later I would have gravitated to cognitive
> science but it didn't exist. Even computer science didn't exist at RPI. I
> learned FORTRAN but it was seen as another tool like a sliderule, not a
> career opportunity.
>
> It's been interesting over the years considering neural networks started
> with an attempt to model a neuron. As you dig into the electro-chemical
> processes going on in the axons and dendrites you sort of wonder how it
> all works.

I've seen some interesting books about neural logic circuits, but circa 1980.

--
You are destined to become the commandant of the fighting men of the
department of transportation.

Farley Flud <ff@linux.rocks> wrote:

>I play my FLAC files thusly:
>
>flac -d -c file.flac | aplay
>
>What could be simpler?

Having an actual GUI, you retard.

--
Joel W. Crump

Amendment XIV
Section 1.

[...] No state shall make or enforce any law which shall
abridge the privileges or immunities of citizens of the
United States; nor shall any state deprive any person of
life, liberty, or property, without due process of law;
nor deny to any person within its jurisdiction the equal
protection of the laws.

Dobbs rewrites this, it is invalid precedent. States are
liable for denying needed abortions, e.g. TX.

On Tue, 09 Apr 2024 17:42:23 -0400, Joel wrote:

>
> Having an actual GUI, you retard.
>

Hey! Hey! How many times did you shit the bed this week?

Huh? How many times?

Answer the question. Your mother told you to be an accommodating
member of society. So answer.

What did your daddy tell you?

Ha, ha, ha, ha, ha, ha, ha, ha, ha, ha, ha, ha!

Farley Flud <ff@linux.rocks> wrote:

>> Having an actual GUI, you retard.
>
>Hey! Hey! How many times did you shit the bed this week?
>
>Huh? How many times?
>
>Answer the question. Your mother told you to be an accommodating
>member of society. So answer.
>
>What did your daddy tell you?
>
>Ha, ha, ha, ha, ha, ha, ha, ha, ha, ha, ha, ha!

<yawn>

Fuckin' pussyboy.

--
Joel W. Crump

Amendment XIV
Section 1.

[...] No state shall make or enforce any law which shall
abridge the privileges or immunities of citizens of the
United States; nor shall any state deprive any person of
life, liberty, or property, without due process of law;
nor deny to any person within its jurisdiction the equal
protection of the laws.

Dobbs rewrites this, it is invalid precedent. States are
liable for denying needed abortions, e.g. TX.

On Tue, 9 Apr 2024 17:26:12 -0400, Chris Ahlstrom wrote:

> I've seen some interesting books about neural logic circuits, but circa
> 1980.

Rumelhart and McClelland's 'Parallel Distributed Processing' was the text
used in a seminar I took that summed up the state of the art. The
perceptron went back to McColloch and Pitts in 1944. It was a start but
had problems some of which were pointed out by Minsky. Hopfield came up
with the network named after him. Rumelhart and Hinton threw in back
propagation based on gradient descent in an '86 paper.

Anyway, read the current literature and you'll see a lot of what was
discussed in the '80s. The problem was hardware. Nvidia is making big
bucks producing $40,000+ GPUs that can handle all the matrix operations
involved. That sort of power wasn't available in the '80s outside of a few
supercomputers and even they might have been breathing heavy.

The other problem was the inevitable hype that oversold NNs.

https://www.skynettoday.com/overviews/neural-net-history

It may be tl;dr but if you're interested that covers the history very
well. It's telling that 'neural network' became a little toxic amid the
stumbles over the years and became 'machine learning'. Having watched the
technology waxing and waning for close to 40 years I don't know if the
current boom will become a bust.

https://mitsloan.mit.edu/ideas-made-to-matter/why-neural-net-pioneer-
geoffrey-hinton-sounding-alarm-ai

Hinton is entering "what have we wrought?" territory. McClelland seems to
be optimistic still.

https://deepai.org/profile/james-mcclelland

Rumelhart died relatively young before AI became a headline item.

On Tue, 9 Apr 2024 17:24:21 -0400, Chris Ahlstrom wrote:

> The Yogi Bear cartoons never mentioned that stuff. Just "pic-a-nic
> baskets".

https://www.boston.com/news/local-news/2022/05/01/black-bear-breaks-into-
connecticut-car-cody-gillotti/

Well, if the pic-a-nic basket is in the car... Note to self: maybe it's
time for some spring cleaning in the Toyota.

https://yellowstoneinsider.com/2016/07/11/old-yellowstone-history-bear-
feeding/

When we went there when I was a kid the bears would line up along the road
and beg for food. As we walked back to the cabin from the campfire talk at
night the bears were nose down in the dumpsters. The last photo with the
'51 Chevy is the right era although ours was a 4-door. The last time I
went through Yellowstone there wasn't a bear in sight.

On Tue, 09 Apr 2024 19:40:35 +0000, Farley Flud wrote:

>> Well, yeah... Waveform analysis doesn't mean anything if you can't
>> hear the difference. Human neurophysiology is not unlike an ADC.
>>
>>
> No, it is not -- at least at the total level.
>
> The human brain processes sound at several levels, the lowest level
> being the inner ear. The next levels are the "reptilian"
> brain structures and then we move on up to the cerebral cortex.
>
> It is that final level, the cerebral cortex that makes all the
> difference.

At some point the analog signal results in neurons being fired and that's
a binary process.

https://www.verywellmind.com/what-is-an-action-potential-2794811

Of course it's a bit more complicated than one neuron.

https://en.wikipedia.org/wiki/Auditory_system

For further entertainment how does the brain dredge up the song title from
a couple of bars

https://www.youtube.com/watch?v=Ti38LFY7x1Y

or decide the Stray Cats listened to a lit of Eddie Cochran?

https://www.youtube.com/watch?v=cFdAV9BOyTI

Farley Flud <ff@linux.rocks> wrote at 21:20 this Tuesday (GMT):
> On Tue, 9 Apr 2024 20:20:07 -0000 (UTC), candycanearter07 wrote:
>
>>
>> You could also do lossless compression.
>>
>
> True. But the discussion is about MP3 files.
>
> Lossless compression is usually done through FLAC, which is yet
> another FLOSS success.
>
> I play my FLAC files thusly:
>
> flac -d -c file.flac | aplay
>
> What could be simpler?
>
> A lot of music, for example the chuckleheads on YouTube, will
> distribute music as VORBIS or OGG.
>
> GNU/Linux can handle all these cases as well.
>

mpv can handle those too
--
user <candycane> is generated from /dev/urandom

On 4/9/24 9:09 PM, rbowman wrote:
> On Tue, 9 Apr 2024 17:26:12 -0400, Chris Ahlstrom wrote:
>
>> I've seen some interesting books about neural logic circuits, but circa
>> 1980.
>
> Rumelhart and McClelland's 'Parallel Distributed Processing' was the text
> used in a seminar I took that summed up the state of the art. The
> perceptron went back to McColloch and Pitts in 1944. It was a start but
> had problems some of which were pointed out by Minsky. Hopfield came up
> with the network named after him. Rumelhart and Hinton threw in back
> propagation based on gradient descent in an '86 paper.
>
> Anyway, read the current literature and you'll see a lot of what was
> discussed in the '80s. The problem was hardware. Nvidia is making big
> bucks producing $40,000+ GPUs that can handle all the matrix operations
> involved. That sort of power wasn't available in the '80s outside of a few
> supercomputers and even they might have been breathing heavy.
>
> The other problem was the inevitable hype that oversold NNs.
>
> https://www.skynettoday.com/overviews/neural-net-history
>
> It may be tl;dr but if you're interested that covers the history very
> well. It's telling that 'neural network' became a little toxic amid the
> stumbles over the years and became 'machine learning'. Having watched the
> technology waxing and waning for close to 40 years I don't know if the
> current boom will become a bust.

My personal opinion on it that its currently over-hyped.

I've had some ML projects and my general conclusion is that it is
essentially statistically-based rules, it is a good enough solution most
of the time, but it depends on the nature of the task and can be quite
quickly limited by its training set..especially if there's risks of
over-training. The ramifications are that its a good short/mid term fit
for some activities, but really complex stuff (side eye at Tesla) are
probably still a decade away, which also means that in the shortsighted
vision of the Stock Market, they're going to lose interest in 2-3 years.

-hh

On 4/9/24 20:09, rbowman wrote:
> It may be tl;dr but if you're interested that covers the history very
> well. It's telling that 'neural network' became a little toxic amid the
> stumbles over the years and became 'machine learning'. Having watched the
> technology waxing and waning for close to 40 years I don't know if the
> current boom will become a bust.

Wrong people have been attempting that. The technology is still in the
wrong hands.

On Wed, 10 Apr 2024 15:40:09 -0000 (UTC), candycanearter07 wrote:

>>
>> A lot of music, for example the chuckleheads on YouTube, will
>> distribute music as VORBIS or OGG.
>>
>> GNU/Linux can handle all these cases as well.
>>
>
>
> mpv can handle those too
>

Yes, by using libvorbis and libogg, which are are not part of mpv
per se, but rather are linked in via libvorbis and libavutil (for libogg).

IOW, mpv, like most GNU/Linux media players, is just a wrapper
around these fundamental libraries.

rbowman wrote:

>Anyway, read the current literature and you'll see a lot of what was
>discussed in the '80s. The problem was hardware. Nvidia is making big
>bucks producing $40,000+ GPUs that can handle all the matrix operations
>involved. That sort of power wasn't available in the '80s outside of a few
>supercomputers and even they might have been breathing heavy

The Nvidia H200 can do 4 PetaFLOPs FP8. That's 4,000,000,000,000,000
floating point operations a second. Incredible.

--
"They use Linux because it doesn't cost anything." - DumFSck, lying
shamelessly

On 4/9/24 11:14 AM, DFS wrote:
> On 4/9/2024 7:52 AM, shitv wrote:
>
>
>> Of course distortion, noise, and frequency response can be measured,
>> although "golden eared" audiophools will assert that the measurements
>> do not suffice.
>
>
> Speaking of which:
>
> "I like movies and music and have a pretty darn good stereo (the
> speakers alone list for over $10k)." - shitv Jul 2006
>
>
> $10K just to listen to the Ghostbusters soundtrack in hi-fi... what an
> idiot.

It might have been worth spending that much ~20 years ago, but for most
of us, mere age has reduced our hearing range capability by quite a bit,
such that Billy Joel's "cheap pair o speakers" does well enough.

-hh

On Wed, 10 Apr 2024 19:02:14 -0400, -hh <recscuba_google@huntzinger.com>
wrote in <uv75pm$189rd$1@dont-email.me>:

> On 4/9/24 11:14 AM, DFS wrote:
>> On 4/9/2024 7:52 AM, shitv wrote:
>>
>>
>>> Of course distortion, noise, and frequency response can be measured,
>>> although "golden eared" audiophools will assert that the measurements
>>> do not suffice.
>>
>>
>> Speaking of which:
>>
>> "I like movies and music and have a pretty darn good stereo (the
>> speakers alone list for over $10k)." - shitv Jul 2006
>>
>>
>> $10K just to listen to the Ghostbusters soundtrack in hi-fi... what an
>> idiot.
>
> It might have been worth spending that much ~20 years ago, but for most
> of us, mere age has reduced our hearing range capability by quite a bit,
> such that Billy Joel's "cheap pair o speakers" does well enough.
>
>
> -hh

Still helps to have a distortion-free playback system that
you can crank up to compensate for hearing loss, if that is
a concern.

I don't have hearing loss -- yet? -- so I have a nice pair of
Polk speakers wired to a USB-capable amplifier. (I _was_ using
SPDIF, but that had some reliability issues.)

One thing we do have "dotted" throughout the house are smart
speakers: Echo Dots. (I guess if any TLA wants to listen
in, they've got us covered.)

ObLinux:
There used to be a snap to run Alexa on a Linux
desktop -- but I don't use snaps. Also, I think there's
a project that turns a Raspberry Pi into a smart speaker.

--
-v

On Wed, 10 Apr 2024 23:39:42 -0000 (UTC), vallor wrote:

> Still helps to have a distortion-free playback system that you can crank
> up to compensate for hearing loss, if that is a concern.

Volume doesn't help frequency response. I haven't had a reason to have an
ear exam recently but I assume I can't hear 20K Hz anymore, if I ever
could.

> some dumb fsck wrote:
>>
>> "I like movies and music and have a pretty darn good stereo (the
>> speakers alone list for over $10k)." - shitv Jul 2006

Big, high-quality, inexpensive. Pick any two, 'cause you can't have
all three.

It's hard to believe that I've had my DALI Euphonia MS5's for over 20
years, now. What a great investment they were! Unimpeachable
performance, beautiful Danish cabinetry...

https://www.dali-speakers.com/media/3290/euphonia-ms5-productpdf.pdf

Their "new and improved" equivalent goes for $25k.

https://www.dali-speakers.com/en-us/products/epicon/epicon-8/

-hh wrote:

>It might have been worth spending that much ~20 years ago, but for most
>of us, mere age has reduced our hearing range capability by quite a bit,
>such that Billy Joel's "cheap pair o speakers" does well enough.

I disagree. I've lost only one octave of the audible range. I can
still hear the other nine octaves fine.

On Wed, 10 Apr 2024 19:52:24 -0500, chrisv wrote:

> I disagree. I've lost only one octave of the audible range. I can
> still hear the other nine octaves fine.

My guitar only covers 4. I'm not sure how high a tin whistle goes in dog
whistle mode. I try not to go there.

rbowman wrote:

> chrisv wrote:
>>
>> I disagree. I've lost only one octave of the audible range. I can
>> still hear the other nine octaves fine.
>
>My guitar only covers 4.

For its fundamentals. The highest note on your guitar may be B5
(987.77 Hz), but the harmonics of those fundamental notes are, of
course, 2X, 3X, 4X, etc higher in frequency. The harmonics are what
makes a note played on a trumpet sound different from the same note
played on, say, a flute.

>I'm not sure how high a tin whistle goes in dog whistle mode. I try not to go there.

--
"Users like this stuff." - trolling fsckwit "Ezekiel", sarcastically
attacking GNU/Linux for including the "archaic" vi, as if normal users
need to use it.

On 4/10/24 8:01 PM, rbowman wrote:
> On Wed, 10 Apr 2024 23:39:42 -0000 (UTC), vallor wrote:
>
>> Still helps to have a distortion-free playback system that you can crank
>> up to compensate for hearing loss, if that is a concern.

I'd typically limit that to headphones, as other listeners would find
the cranked-up treble to be too "tinny"/

> Volume doesn't help frequency response. I haven't had a reason to have an
> ear exam recently but I assume I can't hear 20K Hz anymore, if I ever
> could.

Increased volume can help some to compensate - its the basic principle
of a hearing aid. Overall, the loss for men is that we can expect to
experience a drop in upper range after age 45, which IIRC is often
around 33% by age 55-60 (and more later).

A lot can also depend on how much we proactively protected our hearing
from high dB hazards such as lawn mowers, leaf blowers, power tools,
etc: if you weren't already wearing hearing protection in your 20's 20+
years ago, its pretty likely that your upper range is already shot.

Finally, the loss is slow over time so its easy to overlook. My last
formal hearing test at my employer was pre-CoVid (~2018) which is how I
know that I definitely have ("age normal") loss. Since then, I've
learned that I can't hear some higher pitched electronic beeps (oral
thermometer), or some automotive "water pump bearing is going".
Probably within five years or so now of needing hearing aids.

-hh