[Frameworks] **VL-JUNK** Re: Analog and digital

Steven Gladstone Steven at gladstonefilms.com
Tue Aug 30 17:06:50 CDT 2011


On 8/30/11 1:28 PM, Flick Harrison wrote:
> Steven,
>
> I don't quite understand your points... are you disagreeing with me, or
> supplementing what I'm saying? I'll go with the former, since debate is
> more fun than agreement on rowdy email lists.

Cool, let's have fun, because there is no point in getting all upset.
I'm neither right nor wrong, just a seeker.
>
> re: POINT 1: ones and zeroes aren't analog representations of "on" and
> "off." You could more fruitfully say that "on" and "off" are mechanical
> representations of 1 and 0, since the binary number system stands in the
> abstract, like the decimal or roman system.

The way I understand it the 1 and the zero come from video and waveform 
monitors, where 0 was black, and 1 was "White" or full exposure. This 
was of course for an analog signal. I know the video scale is measured 
as I.R.E. and the scale is 0 to 100 (or actually 7.5 to 100 in NTSC 
land) but my understanding is that 100 I.R.E. correlated to 1 volt.
Digital is strictly an ON/OFF voltage states. This can be represented by 
ots and dashes, ones and zeros, but through the wires, it is all on off 
voltage states - the numbers 1 and zero are not being sent through the 
wires.


>
> I can transmit the same information to you by reciting ones and zeroes
> over a tin-can telephone as I could by sending an email. You would have
> to translate them into ASCII text in order to read the message, which
> you could do with a printed reference card. On and off would not be
> needed during that process.

Yes, but then in reciting ones and zeroes you are speaking them, and 
thus sending an analog signal. :-)

>
> In a digital process, noise floor and strength of signal are eliminated
> with error checking. A digital signal doesn't degrade, it corrupts, it
> becomes unreadable or noisy and has to be re-transmitted.

No argument.

When sampling, the ANALOG value is converted to a number and this is 
sent digitally. Copying an Analog waveform it seems is not so 
clean/perfect as you point out. Send an Analog signal over a wire, and 
it may degrade. So what reads as a nice curve, may shrink or grow or 
change, but not that much.

But back to digital degredation - through Fiber optic "Digital" signals 
can be sent quite a distance, but through copper - not so much. Remember 
we are not sending numbers (not sending ones and zeroes) we are sending 
On off states, which when they are received are converted into numeric 
values. But if there is a problem with the wires, or the transmission 
distance is too great -USB is limited I believe to about 25 feet. Well, 
a voltage state of 1 volt could be considered on and a voltage state of 
zero as off - but what if you get a voltage state of .5? Is that from 
noise in the wire? Yes Error correction and such, but again, not sending 
a one and zero, just sending voltage states through the wire - light 
pulses through fibre optic cable.

  This is one of
> the reasons it challenges analog-lovers who also love degradation and
> imperfection. Noise is not the same as fading.
>
> To quote spider robinson, who was referencing william gibson,
>
> "I remember when sky the colour of television, tuned to a dead channel,
> didn't mean blue sky."
>
> re: POINT 2: if you're saying that all imagers are analog, that's true
> in the sense that the light waves coming in through the lens are analog,
> but in a digital process they get sampled at some point.

Yes, the imagers are analog - Electronic imagers use A to D converters 
(Analog to Digital) to get the info of of the chips, CCD's etc. With 
Film  yep considered Analog (maybe there is a different word for it.) 
Electronic - starts out as analog and converted to Digital (voltage 
states) for transmission and storage.

  In an analog
> process, there's no sampling anywhere; waves are recorded as waves with
> more or less fidelity, like film exposure or signal on tape. But
> sampling is different: however it's recorded, the sample isn't a wave.
>
> Of course things also end in an analog way, either when they become
> light and sound waves from the projector and speaker or at least when
> they hit your retina or ear drum.

Sounds and looks good. :-)


-- 
Steven Gladstone
New York Based Cinematographer
Gladstone films
Blog - http://indiekicker.reelgrok.com/
http://www.blakehousemovie.com
http://www.gladstonefilms.com
917-886-5858


More information about the FrameWorks mailing list