• 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!

PHILIPS RED BOOK

danadam

Addicted to Fun and Learning
Joined
Jan 20, 2017
Messages
975
Likes
1,519
16 bit done well is enough, but just barely enough. Everyone in every part of the chain must do their job for it not to get messed up.
Not sure I understand. Are you talking about using 16 bit in processing? I thought nobody argues for doing that. Or do you mean that if the final result is going to be converted to 16 bit then earlier stages (in 24 bit or more) need to be careful to not negatively affect that last stage?
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,590
Likes
239,474
Location
Seattle Area
We can hear some 10-15 db into noise floors at least.
Given a choice though, we don't way to try to hear our music through that noise. :) We like it to be noise-free which is the concept behind having 120 db of dynamic range. It is only then that we can show that the noise added by the channel is below our threshold of hearing.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,690
Likes
37,414
Given a choice though, we don't way to try to hear our music through that noise. :) We like it to be noise-free which is the concept behind having 120 db of dynamic range. It is only then that we can show that the noise added by the channel is below our threshold of hearing.
I don't know. Considering the popularity of LP maybe the hot ticket is dithered 12 bit. With that EQ curve thrown in.:)
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,690
Likes
37,414
Not sure I understand. Are you talking about using 16 bit in processing? I thought nobody argues for doing that. Or do you mean that if the final result is going to be converted to 16 bit then earlier stages (in 24 bit or more) need to be careful to not negatively affect that last stage?
Everything 24 bit until the end.

I did some recordings for a group I know. Everything in 24 bit. They wanted to make a CD to sell when they play at different places. I converted what I had to 16 bit as the final step and used shaped dither. You only want to use shaped dither once at the end. I told the people making the physical CD it was already dithered 16 bit not to change it or re-dither it. They apparently are used to getting 24 bit files and doing dither to 16 bit. So they took the already 16 bit files and re-dithered it themselves. Which served no purpose other than to add a layer of additional noise.

I've read of people like Bob Katz who remasters some 24 bit hi-res material choosing everything carefully having the same problem. The resulting streaming hires files he so carefully re-mastered somewhere along the way don't end up bit for bit copies of his final masters. People in the music biz aren't carefully handling every aspect of getting recordings out there. They are handling lots of recordings as quickly and inexpensively as possible. So as few places as possible need to change bit depth.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,690
Likes
37,414
I'm in the market for those recordings. Which sections are they in on gearslutz?

Look in the Remote Possibilities sub-forum. That is where they are most common. Often you just get a snippet though sometimes more.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,690
Likes
37,414
The recent obsession with a low fidelity format such as the LP is one of the most bizarre things to happen in the annals of audiophilia.

I hear cassette tapes are starting to catch on in the way LP has been for a few years. All that will be left is for 8 track to make a serious come back.
 

sergeauckland

Major Contributor
Forum Donor
Joined
Mar 16, 2016
Messages
3,456
Likes
9,145
Location
Suffolk UK
Maybe this'll totally destroy my credibility, but I've recently got into SQ Quadraphonics, having bought a Sony SQD-2020 decoder to replace my DIY effort.
So, not even decent stereo!

S
 

Fitzcaraldo215

Major Contributor
Joined
Mar 4, 2016
Messages
1,440
Likes
633
Maybe this'll totally destroy my credibility, but I've recently got into SQ Quadraphonics, having bought a Sony SQD-2020 decoder to replace my DIY effort.
So, not even decent stereo!

S
Ugh. Yes, your credibility now hangs by a tenuous thread. SQ Quad in the analog domain? My hunch is a lot of not very expensive Mch AVR's might do a far better job via someting like DTS:Neo 6 or even Dolby Pro logic in the digital domain. Seriously, Mch, especially discretely recorded, but even synthesized from 2.0, has come a very long way since the quad days. And, quad 4.0 just does not measure up to 5.0/.1 or higher, with the critical center channel in place.

But, since I am in a holiday mood, you are forgiven. Happy listening and happy holidays.
 

fas42

Major Contributor
Joined
Mar 21, 2016
Messages
2,818
Likes
191
Location
Australia
Interesting discussion. My brain tells me to agree with the “CD quality is more than enough for real life” crowd, but my heart tells me to side with Amir on this! Reason: I’ve always felt that one of the things which separates real acoustic music from reproduced music is the effortless dynamics of acoustic music, which hifi systems very rarely are able to mimic.
That's the apparent dilemma. "Effortless dynamics" are the goal, but almost no systems achieve it - why? Because the level of audible anomalies are too severe when the SPLs are increased to an appropriate level - and the brain rejects the sound as being wrong, wrong, wrong. This is one of the obvious markers of competent playback - that the dynamics of acoustic music making are all in place, there is nothing that irritates our hearing systems and causes us to find fault with the sound. This means, for example, that the intensity of the reproduction masks all other sounds, depending on the type of music - this has nothing to do with PA crud pulverising one's eardrums.
 

sergeauckland

Major Contributor
Forum Donor
Joined
Mar 16, 2016
Messages
3,456
Likes
9,145
Location
Suffolk UK
Ugh. Yes, your credibility now hangs by a tenuous thread. SQ Quad in the analog domain? My hunch is a lot of not very expensive Mch AVR's might do a far better job via someting like DTS:Neo 6 or even Dolby Pro logic in the digital domain. Seriously, Mch, especially discretely recorded, but even synthesized from 2.0, has come a very long way since the quad days. And, quad 4.0 just does not measure up to 5.0/.1 or higher, with the critical center channel in place.

But, since I am in a holiday mood, you are forgiven. Happy listening and happy holidays.
I have a fascination with the dead ends of audio, like Quadraphonics. I've also a 1963 Reel-Reel Ferograph tape machine I'm restoring, and have just repaired a 1971 Cambridge Audio P50 amplfier, so hopelessly nostalgic about the old stuff I grew up with.
But thank you for your forebearance.
S
 

Sal1950

Grand Contributor
The Chicago Crusher
Forum Donor
Joined
Mar 1, 2016
Messages
14,155
Likes
16,835
Location
Central Fl
I have a fascination with the dead ends of audio
Hey, I was there. This is a very bad, old photo of my Marantz 2270 receiver and the 2440 quad add-on amp. They had SQ, QS, and CD4 plug in modules available for it IIRC. A planned obsolescence proof design that was killed when Quad playback itself died :(
Marantz1976_v1.jpg
 

tomelex

Addicted to Fun and Learning
Forum Donor
Joined
Feb 29, 2016
Messages
990
Likes
572
Location
So called Midwest, USA
I made some demo files.
https://drive.google.com/open?id=1kodOlklHbIAUD3k4ISalRz_Ma4BvqgPu

original.flac: original 16/44 file with very close to 0dBFS peak
[low peak float]: peak normalized original.flac from -90 to -120dB and saved as 32-bit float
[low peak 16bit]: convert files in [low peak float] to 16-bit
[+60]: add 60dB gain to files in [low peak 16bit]

So, after trying the files, find some CDDA rips or hi-res files with meaningful (music, voice, or anything you regard as "details") and unprocessed (e.g. digital fade) contents as loud as original.flac and as quiet as -110 16bit.flac within the same track.

Also, listen to original.flac and -110 16bit.flac alternately without touching the volume knob.

These are not blind tests, but I just hope people can actually listen, before digging too much into numbers.


can you summarize your findings please with above tests
 

bennetng

Major Contributor
Joined
Nov 15, 2017
Messages
1,634
Likes
1,693
can you summarize your findings please with above tests

So, after trying the files, find some CDDA rips or hi-res files with meaningful (music, voice, or anything you regard as "details") and unprocessed (e.g. digital fade) contents as loud as original.flac and as quiet as -110 16bit.flac within the same track.
Nothing in my collection meets the above criteria.

Also, listen to original.flac and -110 16bit.flac alternately without touching the volume knob.
I hear nothing in -110 16bit.flac. No noise, no music. So either or all of the followings are true:
[1] My system is not capable of 120dB peak SPL.
[2] My listening environment is not as quiet as 0dB SPL, peak or average.
[3] I like to adjust the volume knob to find my comfortable listening volume for a specific track, but I don't like to use the volume knob while listening to a track after I found my desired listening volume.
 

Fitzcaraldo215

Major Contributor
Joined
Mar 4, 2016
Messages
1,440
Likes
633
Ah, yes. Quad. Even back then, I had a keen interest in Mch after some experiments with Hafler's Dynaquad, and some other approaches. I never adopted those approaches, and I was never tempted to make any investment in SQ or QS Quad. I thought it an overhyped gimmick, based on the atrocious, compromised delivery via the LP. Even back then in the early 70's, I had serious problems with vinyl as a playback medium, even for stereo, and, even then, I speculatively envisioned and fantasized about a form of digital playback using light beams from some new type of media rather than a dust laden LP and stylus. Within the decade, I was proven right when the Philips Laserdisc appeared, followed by the CD. I was excited. But, commercially viable Mch for music still had to wait a few more decades.

I do have some discretely remastered Mch Quad-era recordings on SACD. Some are pretty good, my personal favorites being two Allman Brothers live albums done at the Fillmore East, including especially the great, classic Mountain Jam track on the Eat A Peach album. There are others in the classical genre containing some great music and performances. But, today's discretely recorded Mch via hirez digital capture and playback is just so much better in so many ways.
 

mindbomb

Active Member
Joined
Nov 11, 2017
Messages
284
Likes
176
I understand that to reliably get 120db of dynamic range, you would need a higher bit depth. I just don't get why you would need 120db dynamic range, even considering peak values. If dither noise is like -90db for 16 bit, and the content is gonna average maybe -25db, it seems like you have a decent 65 db buffer on average with 16 bit.
 

Fitzcaraldo215

Major Contributor
Joined
Mar 4, 2016
Messages
1,440
Likes
633
I understand that to reliably get 120db of dynamic range, you would need a higher bit depth. I just don't get why you would need 120db dynamic range, even considering peak values. If dither noise is like -90db for 16 bit, and the content is gonna average maybe -25db, it seems like you have a decent 65 db buffer on average with 16 bit.
Though an admitted fan of hirez and greater bit depth. I don't think you need 120dB to be delivered to your listening room. I think the advantage of greater bit depth may be primarily in recording production, considering all the processing stages a recording goes through. I think it adds useful headroom to recording production, possibly even benefitting music downrezzed for RBCD media.
 

fas42

Major Contributor
Joined
Mar 21, 2016
Messages
2,818
Likes
191
Location
Australia
I understand that to reliably get 120db of dynamic range, you would need a higher bit depth. I just don't get why you would need 120db dynamic range, even considering peak values. If dither noise is like -90db for 16 bit, and the content is gonna average maybe -25db, it seems like you have a decent 65 db buffer on average with 16 bit.
All the twaddle about the importance of hires is exactly that - competent playback of ridiculously limited quality source material delivers a powerful, overwhelming experience, where questions about dynamic range are just laughable ...

8 bits of source quality with the right dithering are enough to deliver a hit of all that one could want - I've done experiments exploring this, and have found it remarkable how much one can 'degrade' the sound and still "get away with it".
 

Frank Dernie

Master Contributor
Forum Donor
Joined
Mar 24, 2016
Messages
6,452
Likes
15,798
Location
Oxfordshire
I am only an amateur but I have never tried to make a recording encompassing the whole of a human's audible range, since nothing I experience uses all of it.
120dB of recording capability is not "needed" however theoretically valid it may be.
Firstly none of the speakers or amplifiers I know of are capable of a 120dB dynamic range.
Secondly If i am recording something loud I adjust the recorder gain to suit. In such a recording the quiet bits have never been 120dB below the peak.
I have 109dB/ watt horns at home and a 1000wpc (into 6ohms) amplifier. Theoretically it could get quite loud in here, but if I did actually have a recording where the quiet bit was 120dB lower than the loud bit and I raised the level here so I could just discern the quiet bit the loud bit would deafen me.
So the whole idea of a resolution of greater than 16 bits being important is, IMO, absurd.
Yes a 24 bit recorder makes it easy for any Tom, Dick or Harry to make a recording encompassing all the sound at their microphone it is nevertheless easier to do so with a 16-bit recorder than it is with reel-to-reel tape, which requires skill and experience to make a good recording.

I have heard one of my recordings downsampled to 16/44 then back to 24/96. This loses any data over and above CD standard which may have been in the original but means the playback system treats it identically to the original - to eliminate any difference in the way the DAC may handle different file types.
In the examples I have tried there is no difference to my ears so I am content myself that any differences one may hear are either in the mastering or the way the DAC behaves.
OTOH in the rare case that I buy a download I choose 24/96 :)
 
Top Bottom