Re: [colorforth] ForthBox and FPGA

["Samuel A. Falvo II" <kc5tja@xxxxxxxx>]

On Monday 01 March 2004 05:37 pm, Adam Marquis wrote:
> Now it gets interesting, thats the kind of info I love to see popping
> through!
> PWM can do color on a TV? hmmm, never thought of it, perhaps because

Pulse Width Modulation is a method of encoding analog voltages using only 
digital signals.  It relies on the exploitation of a low-pass filter's 
(usually linear) skirt response to arrive at any arbitrary output 
voltage.

However, it comes at a major price.  To get the same effect as an 8-bit 
DAC, you need to drive the output at 256x oversampling.  To get the 
effect of a 16-bit DAC, you need to get 65536x oversampling.  As you can 
see, it gets expensive pretty quickly.

Let's assume that 10-bit resolution is all that is necessary.  To drive 
the NTSC signal, 6.0MHz of bandwidth is necessary to properly encode the 
lumanence, color burst, and if any, the audio subcarrier (itself both 
frequency and phase modulated to get stereo sound).  To minimize 
quantization noise, you should sample at 4x the desired bandwidth, or in 
this case, 24MHz.  With 10-bit resolution, that's another 1024x 
oversampling, or 2.4GHz.  Is it doable?  Certainly.  But in an FPGA?  
Not in any that I'm familiar with.

> Same for my car, I never want to use a petroleum powered car, never.
> And I wont allow that SHIT to continue. Air powered engines,

Yes, I would *love* to see compressed air vehicles.  However, remember 
that doing so is like riding on a bomb (literally) even more volatile 
than a tank full of gas fumes.  If you get into an accident and the air 
tank ruptures, you'll have only a few microseconds to make peace with 
the deity of your choice.

> for common vehicules, with a opensource controller (ala MegaSquirt),
> running Forth, my favorite computer interfacing language,
> on perhaps your ForthBox hardware is entirely possible to do.
> There's even a patent!! But I would love it even more running
> on custom hardware!!!

Yes, ForthBox would work very nicely for this.  There is no need to waste 
time updating video while driving, since you're too busy looking at 
where you're going.  When doing automotive diagnostics, however, video 
display can be trivial and highly useful.  Certainly MUCH more useful 
than having to count the number of times the dash board lights blink to 
get a diagnosis of the car.

> from the same controller to your sound system, for complete digital
> restitution and freedom (circuit cellar article on digital amp) of
> building active noise cancellation, anything. Also, I can interface

I'm also a fan of active noise cancellation mufflers.  They have the very 
unique advantage of being able to adjust to the engine's RPMs, which 
ensures proper cancellation at all RPMs, while allowing the exhaust to 
flow quite freely.

> I have my own little theory, that radiant energy alone (no light)
> can make plants grow, I have to try it (see keelynet, chlorophyll
> energy) and build something that enables me to prove that is has
> value.

Chlorophyl requires light to function.  Interestingly, it works by 
converting light to electricity (literally; chlorophyl releases an 
electron for every photon it receives.  It is the sole chemical 
substance, in fact, that is responsible for the research into organic 
LEDs and other organic semiconductors!  However, it's not direct 
electricity -- after the electron is released, the chlorophyl converts 
it to an ion, which is used in later chemical processes to create food 
for itself).

> Ankhenaton, radiant energy, sungazing.com, the page at keelynet,
> they're all related somehow. Even heard of an experiment using the
> sun to make a tin sheet radioactive, by using a 10 inch fresnel lens.
> (cant recall much more, sorry, but will search if someone is
> interested)

Keelynet is loaded with documents that are demonstrably (yes, 
demonstrably) false and inaccurate with the real-world.  I really would 
NOT consider Keelynet a valuable source of scientific knowledge.

--
Samuel A. Falvo II


---------------------------------------------------------------------
To unsubscribe, e-mail: colorforth-unsubscribe@xxxxxxxxxxxxxxxxxx
For additional commands, e-mail: colorforth-help@xxxxxxxxxxxxxxxxxx
Main web page - http://www.colorforth.com