G5 : 64 bits or 32 bits ? - Page 4

LaserDiscs are not analog. This is a big myth. LD video may not be like DVD video, but it is definitely not analog.

LD video and sometimes sound is stored as an analog signal converted to a digital format using a form of pulse FM. Basically it is something like you pick a base frequency then you take your sound and clip around it to get the 0's and 1's to store on the disc. During playback you use the knowledge of the base frequency and the 0's and 1's to reform the signal. Of course some info is lost since there is no way you can store something continuous (analog signal) in a discrete format (digital) without losing something. Just like you can't make a mapping of R to Z that is a bijection.

So as you can see, LD data is discrete an thus is not analog. The method takes a video signal (NTSC for US) and encodes it in binary. It is no more analog than CD audio is.

[ 04-13-2002: Message edited by: mmaster ]

[ 04-13-2002: Message edited by: mmaster ]</p>

To put it in other words:

"The video signal on a laserdisc is not recorded digitally but is an analog, composite video signal translated into digital language (binary) and then decoded back into an analog format for display. The video quality of laserdiscs exceeds that of every consumer format besides DVD. DVD offers 540 horizontal lines of resolution with an overall better picture quality due to its pure digital format."

mmaster & JLL ,

well....wrong.

On a laserdisc, the video signal is never quantized and that's what makes the difference between digital and analog (what you use ADC's for) .

While you're right that 0/1s are written to the disc, their lengths are variable (values in a non-discrete, continuous range), that's how the amplitude is stored. The signal is still considered analog, there's no way you could convert it to a digital bit stream (for example store it in a file) without quantization.

As for " It is no more analog than CD audio": CD audio is quantized (16 bit) and therefore 100% digital (sidenote: today's laserdisc audio is also digital (but could be analog)).

[ 04-13-2002: Message edited by: 123 ]</p>

This is great talk about laserdisc! I've never been able to figure it out- is it digital, is it analog, if it's analog, how does a laser read it... So judging from the latest posts, I'm understanding it works like a duty cycle of 1's and 0's- sort of like a pulse-frequency modulation? Is that right? Well that makes sense. Pretty amazing they came up with a system like that just to take advantage of the (then new-fangled) laser pickup technology.

The most important thing to note about LD (unless I misunderstand this, as well) is that it was good for about 400 lines of horizontal resolution which may not sound as good as DVD, but this video feed was uncompressed. However clear the master was, that is exactly what you get from the LD. Yes, DVD sports 740 lines of horizontal resolution, but it is inherently a compressed format where the actual resolution you see is traded off directly as soon as you get any motion on the screen. The only time you really get 740 is on a still-screen (and you happen to have a TV that is capable of resolving 740). Anything else, and it is blurred-down to meet a data rate limit. Yes, I'm making more fuss about it than it really is. Obviously, people watch their DVD's everyday and certainly enjoy the experience. I just wanted to point out (from my personal observations watching particularly image-demanding DVD and compressed HD video sources on good display units), it became clear to me that DVD was in many ways a step forward for video, but at the same time it was just a step sideways and sometimes a step backwards in certain areas. DVD can look great, but if you look deep enough you find that it also has some messy edges that are cleverly swept under a rug where it is hard to see, but it is certainly there.

[quote]Originally posted by RazzFazz:
<strong>Could you provide a link or something? I find this a little hard to believe. How would you store analog data on optical media?

EDIT: Never mind, looked it up myself. They indeed seem to have analog tracks. My apologies. Still wondering how this is stored optically, though... :confused: </strong><hr></blockquote>

I was sure I knew. but after reading some of those later posts. I'm not so sure myself.

[quote]Originally posted by RazzFazz:
<strong>And 40 bits of overhead give you a little more than 1 trillion operations on a single sample before you have rounding errors. Can anyone say overkill?</strong><hr></blockquote>

I suppose you're possibly right. so maybe only 32 or 36-bit would be necessary.

[quote]Originally posted by RazzFazz:
<strong>And your point was?</strong><hr></blockquote>

My point was that palettized color data is stored in the file(And in memory) and transmitted through the system in exactly the same binary format as nonpalettized color. and that the only difference between the two is the use of a header string(Such as the CLUT) in the file that instructs the synthesizers on which colors are to be assigned to various binary values.

[quote]Originally posted by RazzFazz:
<strong>It seems to escape you that the the resolution grows exponentially with bit depth.
With 24 bit audio, you can distinguish 2^24=16.8mio levels of amplitude ("sonic pressure"?). Adding another 8 bits [i]doesn't just add 5.6mio (one third of 16.8mio) to that number, but rather multiplies it by 2^8=256. 64 bit audio would multiply the number by more than one trillion.</strong><hr></blockquote>

I suppose you're possibly right. so maybe only 32 or 36-bit would be necessary.

[quote]Originally posted by RazzFazz:
<strong>48bit digitizers? As in 48 bit ADCs? I don't think something like that even exists in the audio field.

Also, it would be nice if you could provide a link or anything that hints towards any existing piece of audio gear that has 48 or 64 bits of internal resolution.</strong><hr></blockquote>

Hmmm... after looking for quite a while. _I_ don't think it exists either(At least outside of laboratories and high-end installations). oh well. but you'd still need it to do good 48-64-bit editing.

[quote]Originally posted by RazzFazz:
<strong>Did you actually read my previous posts at all?
I hate to repeat myself, but as I already said at some point earlier in this thread, AltiVec can do exactly that, and can even do it for two such 64-bit-pixels at a time, in a single cycle, right now, with the current G4. This is exactly what it excels at.</strong><hr></blockquote>

It sure does. but what if you want to do an integer calculation on it? OOPS! back to square one.

Eric,

[ 04-15-2002: Message edited by: Eric D.V.H ]</p>

[quote]Originally posted by powerdoc:
<strong>as i have say here, the analogic circuits of the powermac are too bad to make a difference between 24 and 16 bits. If i want to ear good sound i do not listen to my mac even with my i stiks and subwoofer, i listen to my HI FI system, both are 16 bits but the HI FI is much, much better.
If i change the CD player of my HI FI system with a DVD audio player it will make an improvement, but an audio 24 bit on my Mac will make no difference the audio analogic circuits are too basic. 5 $ of analogical audio circuits can not match 1000 $ one.</strong><hr></blockquote>

You'd be surprised what $5(In this case. $3 and ¢8 ) <a href="http://focus.ti.com/docs/prod/folders/print/pcm1602.html" target="_blank">will buy you</a> these days. 48 or 64 bit ones might cost a bit(Heh heh) more though.

Eric,

[ 04-15-2002: Message edited by: Eric D.V.H ]</p>

123:

I did not realize that the processing of the laser pickup was continuous on LD video.
There is no reason that a pulse-FM signal can't be stored digitally though. You just end up losing accuracy at the end-points of transition from low to high and high to low. That was what I though the LD circuit did with the pulse-FM encoding.