PlanetZ forums - scopeusers.com

hifiboom wrote:astro, especially about the CD players, its more important in that context how well the d/a is interpolating the angled digital waveform into a sinus based one....
a digital wavefile in that context is always angled, as every sample point does represent a period of 1second/44100 = 0,02 ms of time.

for some deeper informations
http://happybob.com/marc/digital_sucks/#
http://www.teamcombooks.com/mp3handbook/11.htm
...

well, those links are pretty lame versus my original source (I quoted somewhat out of context, but the principle is really close to what Medway is talking about).
You may check Skritek, 'Handbuch der Audio-Schaltungstechnik', Franzis, page 338 following next time you happen to pass a bookstore...
[edit]oops, it's from 1988, so your local library may be a better guess...
the 2nd is a good comprehension of the technical terms, tho.

it was obvious that the seller in the shop had no idea about the 'value' of the CD player, just the drive mechanics would yield at least 10 times the amount I paid - greed is man's strongest emotion (as it was once called...)

only when checking specs I found that the Philips engineers had used the same technique Medway described to maximize the result from limited resources.
the drive's humble numeric specs are in complete contrast to it's audio performance, which makes it a good example in this context.
and it's (detailed) processing is way more complex than the typical web explanations, which are in fact often misleading

cheers, Tom

hifiboom wrote:digital waveforms are no continuous functions like analog ones....

if you lower the bit depth to very low values, you get noise, because the sine waves more or less end up in square waveforms...

thats a completly other phenomena than the noise in analog world....

if you have perfect silence in a 4-bit audiofile, you could amp it to endless volumes without getting noise, in analog world you can not.

They are not stored as continous but they are reproduced as such. What you see on your screen is only part of the equation.

There is always error when quantizing audio and it is never perfectly continous (there is always some effect of the stairstep/square wave).

But that's what dither and filtering are for, to remove the error.

jeah you are right, but the lower the bit depth is, the less the dither and filter will work!


a 16 bit data is called integer and has ~65000 absolut values or 32000 bipolar values

now a sine could look like this

0 +200 +300 +400 ....... +32000 .... +400 +300 +200 +100 0 -100 -200 ........... -32000 ....

now if you try to represent this with a 3 bit file you get something like this....

0 0 0 0 .... 1 1 1 .... 2 2 2 2 ..... 3 3 3 3..... 4 4 4 4 .... 3 3 3 3.... and so on....

the sine gets into a pulse stairs style waveform.... and its pretty hard to get back the information...

if your d/a would smooth out these pulse style waves with filters into sines again, the system would not be able to replay any pulse style waves.

the interpolation an a d/a converter only work at the sample rate level ( for example 0,02 ms)
Meaning it smoothes the squares only at the smallest quantization period, meaning 0,02 at 44,1 khz or 0,01 at 96khz.

but due to the missing resolution.... the squared steps are much bigger than 0,02

this is called bit crusing and sounds similar to a distortion effect.
but it does not have to much in common with noise-level.


bits does represent the dynamic range.... how fine are the steps between the lowest possible signal and the highest possible signal and so they are called dynamic range...

in video and image area this is called contrast....

Sounds like your mixing bits and sample rate together into some problem that doesn't exist.

Distortion from lack of bit depth is solved from dither and is done perfectly when applied correctly.

Sample rate as long as it is according to Nyquist will be perfectly reconstructed with the filter. It's not there to "smooth squares" but reproduce the waveform exactly. Just like you can draw a perfect circle with only 3 points or a line with only 2.

These are the facts of digital audio. No video analogies necessary.

medway, sampling rate and bit depth both work together to replicate a waveform (which is timebased amplitude information), so you cannot divide them into two completly different things, they both work together....

and the lower the "resolution" is, the more side effects you`ll get.



the red line is the original signal the green bars are the digital values...



the blue line is the reconstruction from the digital values into analog voltage....
you see side effects on frequencies, too.
And the less bit depth you have, the more side effects you will have on frequencies too.

same when you lower the samling rate, you get side effects on amplitude.


increasing the sampling bit depth alone won`t improve quality too much....
you have to increase bothe the bit depth and sampling rate to get a better approximation of a voltage based analog waveform.

especially when it comes down to algos with many processing stages like reverbs and filters, these are the major flaws of a digital environment and thats where analog still rules, because of its nearly endless resolution regarding amplitude/voltage and frequency throughpass ...

digital environments have the problem of degrading the sound with every stage if not carefully coded....

there is already a "quality loss" when sampling an anlog waveform.

Sorry either that image is incorrect or you are thinking it says something it doesn't.

The only effect sample rate has on dynamic range is that by say doubling from 48 to 96 you put the overall noise across a wider range of frequencies. This means half of the the noise now resides around the audible range and the other half doesn't. This effectively reduces your noise floor by about 3db.

And bit rate as long as its sufficient (about 3bits) does not affect the wavefrom through eventual reconstruction of the filter as the dither completely removes any harmonic distortion.

Yes both bits and sample rate work together to form a wave but they are corrected by separate measures.

As I've stated there are no side effects of sampling as long as dither is used properly and nyquist is observed. Whatever else you are reading is incorrect.
I suggest you find some better sources for your information on how digital works. Try sonnoxplugins.com

You're also missing 40 samples per sec in your calculations there hifiboom

medway wrote:The only effect sample rate has on dynamic range is that by say doubling from 48 to 96 you put the overall noise across a wider range of frequencies. This means half of the the noise now resides around the audible range and the other half doesn't. This effectively reduces your noise floor by about 3db.

about which noise you are talking now ? dithering noise?
or the noise coming from sampling process over an a/d convreter. ?
you should not put dithering into the same bag like digital audio bit depth....
dither is an effect that CAN be apllied but don`t have to....

so as long as you don`t use dither noise on a digital waveform the noise level on a 8 bit file is as fine on a 64bit file.

hifiboom wrote:

medway wrote:The only effect sample rate has on dynamic range is that by say doubling from 48 to 96 you put the overall noise across a wider range of frequencies. This means half of the the noise now resides around the audible range and the other half doesn't. This effectively reduces your noise floor by about 3db.

about which noise you are talking now ? dithering noise?
or the noise coming from sampling process over an a/d convreter. ?
you should not put dithering into the same bag like digital audio bit depth....
dither is an effect that CAN be apllied but don`t have to....

so as long as you don`t use dither noise on a digital waveform the noise level on a 8 bit file is as fine on a 64bit file.

Ok I just happened to stumble on your source by mistake when looking up another subject, howstuffworks.com.

Unfortunatly their information is misleading at best.

The noise I was talking about was quantization noise.

More importantly though, dither is NOT an optional 'effect', its required for a digital system to work. It's like saying oil is an 'option' for running an automobile. Sure you can leave it out but it's not going to work as intended.

Please read the Sony article I already mentioned in this thread. It explains things quite clear (although I have also tried to here as well). I gather you have read some bad information which has distorted (excuse the pun) your idea of how digital works.

As for me I have said all I can on this subject as I feel I am only repeating the facts over and over at this stage.

just for those who can't be bothered to look it up themselves, this is from the Sonnox article Medway linked earlier:

Conventional Dither
In both 24bit and 16bit output word length selections, high pass TPDF (triangular probability density function) dithering is applied to the output of the plug-in. Since any signal related error caused by finite word length limitation is turned into constant random noise with no relation to the signal itself, such dithering provides complete removal of harmonic distortion due to precision limits, which are an inescapable result of any numerical signal representation. Dithering also suppresses any possibility that the programme will suffer loss of harmonic signal resolution due to word length restriction.

...dither turns a quantised numerical signal conduit into the equivalent of a naturally continuous (un-quantised) system, which exhibits a finite signal to noise ratio with no practical limit to harmonic signal resolution. In other words the inescapable presence of quantisation in numerical systems does not forcibly lead to 'discontinuity' or 'resolution loss' in the signal. Misunderstandings of this fact underpin many of the most damaging misconceptions surrounding digital audio systems. It can also be deduced from the above plots that any undithered digital representation of an audio signal is effectively illegal.

ahh, yes. the difference between theory and reality...
if these numbers were so straight-forward in their implimentation, thaen all sequencers would sound the same. scope would sound trhe same, so would all the scope mixers.

in fact, every sequencer sounds very different. my ears tell me that unfailingly, and i don't give a f#%k why. sonar sounds different than cubase and cubase sounds different than logic and samplitude, they are all different. i have legitimate copies of each on my machines and there's no denying it.

in scope, when the stm series came out with v3x, there was a very different sound from the old "big mixer", which was commented on by many users. the stm2448 sounds very different from the stm1632. many have noticed this as well.

the question wasn't "can a 16bit file sound as good as a 24bit file if you dither?". it was, "do mixes from cubase done in scope sound better than cubase alone?". the answer to that is "YES!". that's why we all use scope. it helps. this doesn't mean that one can't do a fantastic job without scope! it means that IN MY AND MANY OTHER USER'S OPINIONS, things with scope sound better. i can tell you EVERY SINGLE user that i have installed a scope card for has come back on their own with glowing comments about how much better their music sounds, which is the bottom line.

hifiboom wrote:so as long as you don`t use dither noise on a digital waveform the noise level on a 8 bit file is as fine on a 64bit file.

In my decimator test, noise was added when I took the bit rate down from 32 bit to 8 bit.

the sony link is good for what it is, but this may not be apllied to every bit depth discussion.

first, this information is given as an explanation how the sony oxford limiter is working. this is a mastering effect and has a dithering option, and that doesn`t really mean that everything in digital world is using dithering.

at the end dithering is just adding noise to smooth out a finite quantization on a signal, but this is also some sort of quality loss, as smoothing out a signal is just more or less a trick. no more or less.

lets take the picture example again. take a 24 bit colour depth real world photo with 16 million colours per pixel, now convert it into something like 4 bit colour picture.... with 16 different colours.
It won`t look natural anymore....
now if you reconvert it into a 24 bit picture it still stays a 16 colour picture that doesn`t look natural at all.
Now dithering process would mean you add noise (random +/- values on ever colour pixel in the 24 picture)
now the picture will look smoother, but still not realistic... just blurred.

So I would say
Dither is smooting process, but not a gain in resolution.


edit: its true, we fade away from the orig. topic..... sorry !

hifiboom wrote:
at the end dithering is just adding noise to smooth out a finite quantization on a signal, but this is also some sort of quality loss

well, according to the article this is simply not the case. I don't know the technical details behind the reasoning but that's what I read there.

lets take the picture example again. take a 24 bit colour depth real world photo with 16 million colours per pixel, now convert it into something like 4 bit colour picture.... with 16 different colours.
It won`t look natural anymore....
now if you reconvert it into a 24 bit picture it still stays a 16 colour picture that doesn`t look natural at all.
Now dithering process would mean you add noise (random +/- values on ever colour pixel in the 24 picture)
now the picture will look smoother, but still not realistic... just blurred.

and we're still not talking about digital images but digital audio, which is not the same thing at all.

So I would say
Dither is smooting process, but not a gain in resolution.

well, again this is in direct contrast to what the sony article explains:

...dither turns a quantised numerical signal conduit into the equivalent of a naturally continuous (un-quantised) system, which exhibits a finite signal to noise ratio with no practical limit to harmonic signal resolution.

edit: its true, we fade away from the orig. topic..... sorry !

it's definitely interesting though I don't know who is "right" or "wrong" and I certainly don't know anything much about this material, but I do find it interesting that your claims are so directly in contrast to what's being said in that article.