funny thing happened compressing some of my albums to 512kbps OPUS for convenience

[Roy_D]

So an album I know back to front of which I have the very nicely dynamically mastered regular CD I came across a hi-res 24bit vinyl-sourced, captured using SOTA stylus, turntable and ADC.
I have so many records on vinyl, including the one this 24bit was done from but the convenience of digital has won me over mostly. Also curious as to how the stylus used sounded in comparison to mine on the Dual 721. Kind of heart warming to hear analog in lossless audio at the push of a button albeit a bit dishonest lol.
But then I thought to try a little experiment. Made a flac of my CD, used 512kbs OPUS transcoding on that, and did same but with added 24 bit to 16 bit and samplerate conversion (dithering set to highest quality) on the HiRes Vinyl flacs and guess what, The vinyl set came out like 20% smaller in total size in the resulting 512kbps OPUS sets I made...

 

[EJ3]

So an album I know back to front of which I have the very nicely dynamically mastered regular CD I came across a hi-res 24bit vinyl-sourced, captured using SOTA stylus, turntable and ADC.
I have so many records on vinyl, including the one this 24bit was done from but the convenience of digital has won me over mostly. Also curious as to how the stylus used sounded in comparison to mine on the Dual 721. Kind of heart warming to hear analog in lossless audio at the push of a button albeit a bit dishonest lol.
But then I thought to try a little experiment. Made a flac of my CD, used 512kbs OPUS transcoding on that, and did same but with added 24 bit to 16 bit and samplerate conversion (dithering set to highest quality) on the HiRes Vinyl flacs and guess what, The vinyl set came out like 20% smaller in total size in the resulting 512kbps OPUS sets I made...

So, it happens this way time & again?
This is intriguing.

 

[DVDdoug]

The vinyl set came out like 20% smaller in total size in the resulting 512kbps OPUS sets I made...

I don't know how Opus works but with a fixed bitrate it wouldn't be different. Bitrate is kilobits per second. There are 8 bits in a byte so 512kbps is 64kilobytes per second or 3.8MB per minute. (That's ignoring any embedded artwork.)

With variable bitrate "easier to compress" audio uses a lower bitrate. i.e. Silence and "simple sounds" will use a lower bitrate. But I don't know how that applies to Opus.

dithering set to highest quality

There should be no reason to dither. Dither is added noise that's supposed to sound better than quantization noise. There's no need to add noise to analog vinyl.

If you want to know what "low resolution" sounds like, make an 8-bit file (and don't dither). ...I've heard 8-bit quantization noise but I've never tried dithering it. I'm just not that interested in 8-bit audio. Somebody on another forum said they preferred it without dither.

 

[beeface]

Is Opus ok for music? I've only ever seen it used for speech applications like VoIP

 

[Roy_D]

Is Opus ok for music? I've only ever seen it used for speech applications like VoIP

Opus has two modes, one optimised for speech and the other for music.
Opus yields the highest quality among the 'lossy' codecs for music.

 

[Roy_D]

I don't know how Opus works but with a fixed bitrate it wouldn't be different. Bitrate is kilobits per second. There are 8 bits in a byte so 512kbps is 64kilobytes per second or 3.8MB per minute. (That's ignoring any embedded artwork.)

With variable bitrate "easier to compress" audio uses a lower bitrate. i.e. Silence and "simple sounds" will use a lower bitrate. But I don't know how that applies to Opus.


There should be no reason to dither. Dither is added noise that's supposed to sound better than quantization noise. There's no need to add noise to analog vinyl.

If you want to know what "low resolution" sounds like, make an 8-bit file (and don't dither). ...I've heard 8-bit quantization noise but I've never tried dithering it. I'm just not that interested in 8-bit audio. Somebody on another forum said they preferred it without dithe

Downsampling from 24b96kHz to 16b44.1/48 involves dithering and one can opt for quick/low or high/slow.
I mentioned I opted for high/slow to be clear ^^

 

[Roy_D]

So, it happens this way time & again?
This is intriguing.

I will see what else I have in 24/96 vinyl sourced and if I have a matching issue cd to test another.

But I suppose the simple fact mastering for the technically differing cd and vinyl format means there will be differences, period.