discuss-gnuradio
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Discuss-gnuradio] MAKING A NTSC TV RECEIVER


From: Martin McCormick
Subject: Re: [Discuss-gnuradio] MAKING A NTSC TV RECEIVER
Date: Mon, 27 Aug 2018 10:58:46 -0500

Andres Campos Santana <address@hidden> writes:
> I tried to listen to the NTSC channel audio using a FM receiver I made, 
> and I got it, I could listen to it perfectly.
> 
> That should be a good way to prove that I'm receiving the channel signal, 
> for that reason, I don't understand why I just receive a diffuse mix of 
> black, white and gray image instead of perceiving a moving black and 
> white moving images. I understand I can't receive a colour image due to 
> the RTL bw (2MHz).
> 
> 
> I was talking with Martin and he told me this problem would be due to my 
> sound card isn't sampling as fast as the program needs or my proccesing 
> unit doesn't support it. I have Intel® Core™ i5-3210M CPU @ 2.50GHz × 
> 4 in a Lenovo Thinkpad t430 computer.
> 
> The program that I am guiding myself is well done since in the results 
> you can see that black and white television.
> 
> What do you think about this problem? Thanks for your help!
> 
> Andrés.

        This is martin again.  I was confused when I
discussed the sound card not being fast enough.  I had a picture
in my mind of the sound card being somehow used to turn all those
2-million samples per second in to a base band video signal.

        In this program, the sound card has nothing to do with
what kind of video you get.  The data from the RTL dongle are
processed to decode amplitude modulation so that each 16-bit
sample represents a voltage level from black to white.  The
Thinkpad appears to be fast enough but this entire process for
each pixel has no time at all to spare.  Every module in that
video chain, both hardware and software, is probably stretched to
the limit of endurance and just one process failing to complete
in time for the next pixel decode is enough to ruin the picture.

        I am not familiar with the Lenovo Thinkpad t430
specifically, but it is a good laptop used by many.

        Desktop and laptop computer design is a sort of war
between cost and capability.  With laptops, one also has energy
usage, heat and size.  When you see a desktop with a graphics
accelerator video card, it is designed to handle math and
addressing faster than does the standard video display for the
same computer.  You could have two of the same model computer
sitting side by side running the same RTL module, operating
system and two identical copies of the same NTSC program.  One
might receive the black-and-white signal and the other might just
get a scrambled mess on it's screen.  Look inside and you would
probably find that the computer successfully receiving the image
has a graphics accelerator card or some sort of math
accelerator hardware to raise the apparent processing speed of
the system.

        How many remember the math co-processor module one needed
for the early IBM PC's?

        I remember that and never got to do any tests on speed,
but if the software could use the math co-processor, you
supposedly got quite a boost in performance with the same 4.7 MHZ
clock.  Ah, those were the days.  By the way, if your software
couldn't figure out whether you had a math co-processor, the
co-processor simply  helped make your room hotter.

Martin WB5AGZ



reply via email to

[Prev in Thread] Current Thread [Next in Thread]