Dec 25, 2024 at 10:51 AM Post #18,541 of 19,075
The biggest issue with obsessing over different lossless codecs: sooner or later you realise that in theory your files could have been converted a bit better, using a different codec or different settings. If you are of the obsessive disposition, that means you can re-convert your whole library, again and again.

So I don't obsess about it. If I don't hear a problem, then for me there is no problem :)
The CD transport I used could have been better. It used to keep me up at night. I was only aiming for 80% and I got 90%. I'm happy. Merry Christmas ❤️
 
Dec 25, 2024 at 1:43 PM Post #18,545 of 19,075
What do we make of this. I'm watching it before bed, so I'll think it over tomorrow. But from my extremely layman's POV, I'm thinking... what is he getting at? Is this a very roundabout way of pooh-poohing measurements... objectivity? And the bit about cables... I still don't really feel like I perceive microseconds. But then what about the 'Perceiving Timbre' section. About different musical instruments being indistinguishable if you trim the first few microseconds from their recorded sound.

Wait... wouldn't that mean that if I was listening with a cheap cable, or audio gear, I wouldn't know what I was listening to? I've never had that experience either.

 
Last edited:
Dec 25, 2024 at 2:31 PM Post #18,546 of 19,075
Lossless is a dirty word. I wish I liked FLAC as much as WAV :)

Once you uncompress FLAC, you get the exact WAV. You can imagine something like the bitstream going 1000000000010000000000110110110 can be compressed to 1 0x10 x2 110x3. Not exactly that but you get the idea. Repetitive data can be described in less bits.
 
Dec 25, 2024 at 5:20 PM Post #18,547 of 19,075
What do we make of this. I'm watching it before bed, so I'll think it over tomorrow. But from my extremely layman's POV, I'm thinking... what is he getting at? Is this a very roundabout way of pooh-poohing measurements... objectivity? And the bit about cables... I still don't really feel like I perceive microseconds. But then what about the 'Perceiving Timbre' section. About different musical instruments being indistinguishable if you trim the first few microseconds from their recorded sound.
it’s a great question, and honestly, your "extremely layman’s POV " is already pretty spot-on in picking up the key points he's trying to make. Here's how I'd break it down:

The big takeaway seems to be that hearing is way more nuanced and complex than some give it credit for. Measurements like frequency response and distortion—which are the basis of most "objective" evaluations of audio gear—are obviously important, but they’re not the whole story. He’s not dismissing measurements but just pointing out that they might not always capture the subtleties that our brains and ears can pick up, especially when it comes to things like timing, transients (those ultra-short bursts of sound at the start of a note), and spatial detail.

The part about microseconds and cables might feel a bit abstract, and that’s fair— it’s hard to grasp something so tiny when you’re not consciously aware of it. (for example, the brainstem processes raw auditory data before it even reaches conscious awareness). his timbre example with instruments is a good way to think about it: even a seemingly insignificant detail, like the first few microseconds of a sound, carries a ton of information that helps your brain figure out what you’re hearing. If you strip that away, it suddenly becomes much harder to tell a trumpet from a clarinet.

So, what’s he getting at? ... I think he’s trying to remind us that audio perception is incredibly rich and layered, and there’s a lot going on that standard measurements can’t fully explain. And that’s why subjective impressions still matter, even if they don’t always align perfectly with what we measure in a lab. Whether or not cables or timing differences really make an audible difference for everyone is obviously a whole 'nother debate, but his broader point seems to be: our ears are pretty incredible, and there’s more to sound than numbers on a graph.
 
Last edited:
Dec 25, 2024 at 5:52 PM Post #18,548 of 19,075
What do we make of this.
Based on past videos I ignore and don't give more views :-)

And the bit about cables... I still don't really feel like I perceive microseconds. But then what about the 'Perceiving Timbre' section. About different musical instruments being indistinguishable if you trim the first few microseconds from their recorded sound.
Are any concrete examples given, which you can listen and hear the supposed difference? If not then I'd put it in "spreading FUD" category.
 
Dec 25, 2024 at 6:42 PM Post #18,549 of 19,075
The biggest issue with obsessing over different lossy codecs: sooner or later you realise that in theory your files could have been converted a bit better, using a different codec or different settings. If you are of the obsessive disposition, that means you can re-convert your whole library, again and again.
I realized that, so before I started ripping to build my media server, I spent the better part of two weeks ripping all sorts of music to various codecs and data rates. I determined where the line of transparency was and arrived at AAC256 VBR. Never needed to look back. It’s perfect and compact.
 
Dec 25, 2024 at 6:55 PM Post #18,550 of 19,075
He’s not dismissing measurements but just pointing out that they might not always capture the subtleties that our brains and ears can pick up, especially when it comes to things like timing, transients (those ultra-short bursts of sound at the start of a note), and spatial detail.
If those ‘subtleties’ can be recorded… why would a complete set of measurements not be able to capture them?
 
Dec 25, 2024 at 9:54 PM Post #18,551 of 19,075
I suppose that the waveform does capture the entire sound, starting with the transients that he says define timbre. But microseconds... I don't know.

If a cheap cable or audio gear smears the sound due to minute errors in the transient timing, then that's something I haven't heard. Instruments always sound like what they are.
 
Dec 26, 2024 at 2:53 AM Post #18,552 of 19,075
Digital audio is by definition a measurement of sound. An ADC takes analog sound and measures it into samples. Those measured samples are used to reconstruct the analog sound in the DAC. If there is anything in sound that can’t be measured, it can’t be recorded digitally. (I.e. square waves, super audible frequencies beyond the sampling rate, or low level sound below the noise floor)

A sample is 1/44,000th of a second, and CD quality sound can capture that. It captures frequencies perfectly up to a 20kHz wave, which is a 20,000th of a second. If it couldn’t, it wouldn’t be able to reproduce the highest frequencies. A musical transient is multiple orders of magnitude below that. The fastest drum hit transient is between a 5,000th and a 1,000th of a second (usually much longer than that). So CD quality sound can capture four times faster than the fastest musical transient. Before you decide whether digital audio can measure and reproduce a transient properly, you should probably find out how fast of a transient you want it to reproduce. Obviously, it can’t reproduce a square wave, but it can certainly reproduce anything found in music.

The math would be the same for a cable too. If it can accurately pass a 20kHz tone, it’s obviously accurate to a 20,000th of a second.
 
Last edited:
Dec 26, 2024 at 3:30 AM Post #18,553 of 19,075
Yeah, I sorta assumed that however marvellous our ears are, that technology can match or better it.
 
Dec 26, 2024 at 3:35 AM Post #18,554 of 19,075
It depends. It’s hard to measure directionality and reflection but our ears can hear that. But when it comes to commercially recorded music, we can measure everything in it.
 
Dec 26, 2024 at 3:59 AM Post #18,555 of 19,075
I meant more broadly, like sonar, but yeah.
 

Users who are viewing this thread

Back
Top