Do I need an expensive digital cable?

@kennyc 

you’ll notice I did not say wife, spouse, or anything of the sort

Maybe you mean porcupine?

Same as a Porsche, according to my then 3-yo daughter, but all the pricks are on the outside 

@richardbrand 

I'll have to work on this. I know a place in the Mojave Desert where I might have the room to lay out and stack that much graph paper. That's going to be quite a delivery from Amazon. 

This discussion brings me back to when I was selling computers at Hewlett Packard to the Navy and Air Force. I would ask a support person a question and she would go on a bit longer than I expected, or could explain to the customer.Support people wondered why sales people were so stupid and sales people wondered why we were needed to boil down explanations for the customer.

But I'm pretty much with you. And thank you for taking the time to explain how silence and error correction are achieved.

Last night I asked chatgpt (you were right about the spelling, and nobody has ever corrected me before. Thanks.) about word length and sampling rate. It said that the word length was for amplitute. Basically what you said. I asked why a 24-bit word length seems to extend a note's decay, as well as making the note sound more fleshed out. It explained how the16 million plus more possibilities for amplitude expression (between 16 bits and 24 bits) could accomplish that.

The higher sampling rates put more dots on the graph paper. Since greater word lengths always seem to accompany higher sampling rates, chatgpt said it was difficult to always know whether my subjective experience of the recorded sound came from  the greater word length or higher sampling rate. 

In language a non-techie like me can understand, can you explain timing. What exactly needs to be timed? Why do I and others often experience jitter as a smearing of the sound? Sorry, my mind works better when relating to subjective experience rather than math. 

A bit more about why I hate I2S as an interconnect 'standard'!  From Wikipedia:

The I²S connection was not intended to be used via cables, and most integrated circuits will not have the correct impedance for coaxial cables. As the impedance adaptation error associated with the different line lengths can cause differences in propagation delay between the clock line and data line, this can result in synchronization problems between the SCK, WS and data signals, mainly at high sampling frequencies and bitrates. As the I²S bus doesn't have any error detection mechanism, this can cause significant decoding errors.

@audio-b-dog 

seems to extend a note's decay, as well as making the note sound more fleshed out

This is easiest to hear with DSD.  There are simply more bits to fill in the original sound pressure graph.  More information = more detail.  It is like the difference between night and day on my system.

High resolution PCM should perform in a similar way, except for the monotonicity issue.  I'll restate that.  When the numbers go up in order, the output should also increase linearly, but switching big values in to replace a set of smaller values makes it almost impossible to actually achieve monotonicity with PCM.

More later

The idea of I2s being inferior isn't supported by what a lot of people are experiencing.  Some gear is optimized for i2s.  In my system, I2s always sounded better than coaxial until I got a better coaxial cable.  Now, they're pretty much equal to my ears.