• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Noise shaped CDs

MCH

Major Contributor
Joined
Apr 10, 2021
Messages
2,641
Likes
2,251
Question, just for my own curiosity:
Modern DAC ICs still come with de-emphasis filters that can be activated via registers, but i don't see the DACs usually discussed here have the option to activate such filters. Is it because they are activated automatically? (i see this is a possibility with spdif signals)
I also see DAC ICs that have filters for sample rates other than 44.1 kHz (i.e. 32 or 48 kHz). Was emphasis used in other commercial formats?
 

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,879
Likes
16,665
Location
Monument, CO
Question, just for my own curiosity:
Modern DAC ICs still come with de-emphasis filters that can be activated via registers, but i don't see the DACs usually discussed here have the option to activate such filters. Is it because they are activated automatically? (i see this is a possibility with spdif signals)
I also see DAC ICs that have filters for sample rates other than 44.1 kHz (i.e. 32 or 48 kHz). Was emphasis used in other commercial formats?
The emphasis bit is (or should be) on the CD itself at the start of the track. The player reads that bit (among others, see "CD Q code" or something like that) and uses that information to program the emphasis. It is part of the CD standard.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,696
Likes
37,430
BTW, lest you think I'm just picking on this recording, the most recent recording I have of this concerto is from 2019 and I'd be shocked if it was recorded at less than 24-bit. It's louder than the older BIS recording (-33dB for the first movement) and ambient noise from the venue is more clearly audible, but the hiss is also still clearly audible in the quiet cadenzas. Why is this still a problem on a CD released in the last few years?
Other than total electronic creations, I'm not sure there is a recording with noise less than about -70 dbFS. As already said by Ifrit, mics, electronics and room noise all add up.
 
OP
darmok

darmok

Member
Joined
Mar 19, 2023
Messages
18
Likes
17
Unlikely the noise you hear is quantization, most likely digitized analog noise.
Sigh. Are you telling me that I won't hear the dithering noise in this?

Screenshot 2023-11-26 at 17.24.47.png
dither16.png



Even without the dither the quantization noise is very clearly audible. This is the exact same output level I was using for my listening tests earlier. Generating this in a higher bit depth and then using Audacity to apply noise-shaped dither to 16-bit results in much less audible hiss and at a higher frequency.

noiseshapeddither.png


Now there very well might be other sources of noise in these recordings, but here's what a 15-second excerpt of the 2019 recording looks like:

concertonoise.png


Hmmm......

concertodither.png


I guess it must be analog noise in the REW generator that's creating the exact same noise floor as this CD. :facepalm:
 
OP
darmok

darmok

Member
Joined
Mar 19, 2023
Messages
18
Likes
17
Other than total electronic creations, I'm not sure there is a recording with noise less than about -70 dbFS. As already said by Ifrit, mics, electronics and room noise all add up.

Room noise doesn't sound like hiss. It sounds like room noise, and the difference in spectral character is very apparent.

concerto4.png


Recording of a different concerto from a SACD rip overlaid on the previous one. Note that by 10KHz the noise floor is down by 12dB, or 9dB A-weighted. And yes, before someone calls me on it, I applied 6dB of gain to the SACD rip before loading this excerpt into REW, which left about 0.5dB of headroom on the whole concerto.

If the point of a recording of live music is to recreate the experience of being there, then room noise is a part of that. Hiss, whatever the source, is not.
 

Ifrit

Active Member
Joined
Apr 10, 2020
Messages
152
Likes
89
What a wide a deep field of electronics development you have in front of you, but you might just be the one that creates piece of electronics that doesn't produce noise in form of a hiss.
 
OP
darmok

darmok

Member
Joined
Mar 19, 2023
Messages
18
Likes
17
What a wide a deep field of electronics development you have in front of you, but you might just be the one that creates piece of electronics that doesn't produce noise in form of a hiss.

What was the point of this comment?

On multitone? unlikely.

Did you try it yourself? Pay attention to the parameters...
 

Ifrit

Active Member
Joined
Apr 10, 2020
Messages
152
Likes
89
What was the point of this comment?
To reinforce my previous one.
Did you try it yourself?
No. I did, however, try many different noise shaping algorithms during production of the CD/HiRes files, including cello concertos. A lot of people here would argue that the type of noise shaping and presence of dither at all is not audible on acoustic instruments recordings. I have a first hand experience about noise induced by electronics and not being affected by dither types (to my imperfect ears, of course), therefore I am skeptical.
Multitone signal still is not real musical material, and is hardly representative in this matter.
 
OP
darmok

darmok

Member
Joined
Mar 19, 2023
Messages
18
Likes
17
Did you try it yourself?
No.
"Whereof one cannot speak, thereof one must be silent."

I'm showing example excerpts from a real, commercially released recording (Eckart Runge cello, "Transitions", Capriccio C5362, 2019, track 5, 15 seconds starting at 0m10s). The intercept point between the spectrum of that segment of recording and the 16-bit TPDF dithered noise floor is at 3KHz. If I'm hearing anything at all above 3KHz, that's what it's from. So what happens if I low-pass this excerpt at 3KHz (48dB/octave)?

concerto5.png


Hey, guess what? No hiss! Of course, I had to actually go through all the effort of putting a pair of headphones on my ears to try it. Wouldn't expect anyone else to go through all that trouble.

Just for fun I switched out the closed-back headphone (Elegia) for a semi-open (Emotiva GR-1) that's sensitive enough to reveal the noise floor of my amp at full gain. I'm sitting in a treated room but the heat is on and there's plenty of electronics with fans going in here. Same result.

I am skeptical.

No, you're cynical. A skeptic is willing to actually try the experiment if the procedure is documented.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,696
Likes
37,430
Well, for the record, I always use shaped-noise dithering when mastering to 16 bits for clients.
Yeah, I did one CD, which was from material I recorded at 96/24. I used noise shaped dither during resampling to 44/16, and told the guys making the CD not to redither it. Yeah, like they listened to me............not. Redithered it with some lousy dither in an old DAW. Was their "standard" practice. :mad:
 
Top Bottom