Find the Beethoven's 9th with the SA youth chorus for a chilling experience.
Bill said:
Cliff said:
This is an important point in the larger scheme of things around here. Each of us have a different point where we say, "That sounds good enough to enjoy the music, at least for now."
But I think there is also an awareness among some that this may be perceived by our peers as giving up too soon, or failing the noble quest for Perfect Sound, or just being lazy or pussing out. I know I think about this from time to time (I'm pretty happy with the sound I'm getting these days).
Plenty of different squishy human-type variables at play here.
Last edited:
.
If you take the current crop of 14 bit / 2600 MSPS ADCs as a starting
point for oversampling, that should be easily possible, even if you
lose a few dB here and there.
For example: AD9208, AD9680, TI ADC32RF44 etc,
8192 times oversampling or so.
Harmonics is the problem, probably, but you can do
large scale dither in frequency ranges that are later completely
filtered out.
Don't know. Some people can hear undithered 16-bit audio though. Audibility seems to depend a lot on the particular type of sound.
Again, some people can hear undithered 16-bit audio. That's uncontroversial.
Some people can probably hear some things at even lower levels, although that does remain controversial. Nobody wants to know bad enough to pay for research to find out for certain, evidently. Some people do like to argue about it for free, though.
- what exactly “hear” means in this context?
- why would anybody want to listen to undithered 16-bit, that must sound horrible in a preference test?
- why is this “uncontroversial”, any cite/quote available?
Don't remember the cite. JCX posted it before, maybe he would be kind enough to do so again?
(EDIT: To best of my recollection it was JCX, anyway. I recall being surprised that it was described as not being controversial.)
As far as what does it sound like, I have heard it. The only ways I can think of to describe it would involve analogies or words that other people use that don't have a technical definition, such as "grainy."
Probably better to find out for oneself. Perhaps download a freeware VST bit crusher plugin and start cutting bits until you hear it at some level. Once you do, you could try cutting less bits and see if you can still hear it once you know what it sounds like.
(EDIT: To best of my recollection it was JCX, anyway. I recall being surprised that it was described as not being controversial.)
As far as what does it sound like, I have heard it. The only ways I can think of to describe it would involve analogies or words that other people use that don't have a technical definition, such as "grainy."
Probably better to find out for oneself. Perhaps download a freeware VST bit crusher plugin and start cutting bits until you hear it at some level. Once you do, you could try cutting less bits and see if you can still hear it once you know what it sounds like.
Last edited:
Not a citation, but JCX talking a little about it here: http://www.diyaudio.com/forums/analogue-source/249132-future-analogue-sources-8.html#post3798354
Again, I feel like you're trying to take me to task for point I didn't make. We've got a couple people here (names been named) which basically state that digital is crap and analog is gold. This includes the entire processing chain. I thought I made it clear that this is an utterly bonkers perspective to take. There are plenty of releases that are processed/mastered in both mediums (and hybrids thereof) which sound incredible. Same goes for terrible. Then you can talk about playback mediums which may come from any process as well.
But, sure lots of crappy remakes of originals for digital. That has nothing to do with the actual methodology/technicals but the nut behind the wheel.
Scott,
First off you should be old enough to remember the sin of unwrapping the celophane from a cough drop during a concert. That is over 100 dB dynamic range from an unwrap to a forte.
Then my listening space is most likely quieter than yours and has more dynamic range. As to masking it is well demonstràted that some signals are percèived below the noise level.
We almost came to a point of agreement. Virtually all modern recordings are compressed. Many of the records if compressed it was done either manually by tape saturation or a single (or stereo) unit. Today it is often done at every input. As most folks recognize clipped digital sounds horrible, the signal is usually recorded with 20 dB or more of headroom. It is all the compression in some recordings that I find objectionable.
Now the question you failed to ask is "Why is a dynamic range that is found in a live performance not acceptable when reproduced?" Some folks actually want to hear the now hidden details excluding of course the celophane noise.
Then again the issue is drifting. I mentioned the system specifications required to cover the entire range of music. Now what is required to match perception is a different issue and so far requires encoding the data in other than a simple linear format.
Gerhard,
That is exactly the approach I always considerd if I wanted to build a high resolution converter.
For humor I have a salesman stopping by with a 32 bit microphone preamp!
Derfy,
I think we are beginning to converge. Digital can be junk and some folks experience is based on an incomplete sample. Vinyl listening has a bias.
My experience is listening to broadcast audio (multiple methods) goes from incredible crap (Sirius) to not quite there. Direct playback of them large round things sounds far better.
Now one of the things I find improves CD player performance is recognizing the implementation limits and addressing them. JJ is annoyed in his CD player the 5 volt rail is used as the D/A reference. I find the EMI noise in the case degrades the sound and an outboard optical linked D/A is a real improvement.
For many tests I use a small personal CD player. It is better than many systems.
Then there was the time I was going into an arena with a CD player and a student seeing this commented "Wow retro."
I think we are beginning to converge. Digital can be junk and some folks experience is based on an incomplete sample. Vinyl listening has a bias.
My experience is listening to broadcast audio (multiple methods) goes from incredible crap (Sirius) to not quite there. Direct playback of them large round things sounds far better.
Now one of the things I find improves CD player performance is recognizing the implementation limits and addressing them. JJ is annoyed in his CD player the 5 volt rail is used as the D/A reference. I find the EMI noise in the case degrades the sound and an outboard optical linked D/A is a real improvement.
For many tests I use a small personal CD player. It is better than many systems.
Then there was the time I was going into an arena with a CD player and a student seeing this commented "Wow retro."
Last edited:
I'm looking on the net. I can't find it. Any other pointers you can reveal?
Last edited:
There are four in this list that seem to comply (40AC, AM, BE, DP)
Measurement microphone cartridges
The noise level (masking effect and sensory dithering)
Rotational instability and resonances (vibrato effect and varying slight inharmonicity).
Pressure variation (tremolo effect)
So true. Thus soft compression from tape saturation and some old tube gear soft overload were/are desirable.
Bill, only in case you are familiar with this musical piece, have a good listening please.
YouTube
George
- Status
- Not open for further replies.