It's all about making them as inexpensive as possible. The USB
controller has just enough bandwidth for the intended purpose. On
a consumer product, even a few cents makes a difference.
Ron
On 08/25/2018 01:13 AM, Adrian Musceac
wrote:
Hi Ron,
So in theory, replacing the USB2 chip with a USB3 would allow
access to the full sample rate, or is there some other internal
limitation?
Regards,
Adrian
On August 25, 2018 8:06:20 AM UTC, Ron
Economos <address@hidden> wrote:
The maximum Transport Stream rate of DVB-T is 31.67 Mbps,
so the USB interface only needs to deliver 4 MB/s. Since you
need two 8-bit samples in IQ mode, it's 2 Msps.
Ron
On 08/25/2018 12:44 AM, Adrian
Musceac wrote:
Hi Marcus,
You're right about the RTL sample rate, but I'm curious
about why it is so small.
Is it the bus speed? The ADC is obviously fast enough for
DVB-T2.
Regards,
Adrian
On August 24, 2018 7:42:17 PM UTC,
"Müller, Marcus (CEL)" <address@hidden>
wrote:
Hi Martin,
internally, the RTL dongles are fast enough to capture full DVB-T (not
-T2) channels, and demodulate, and decode them, and deliver the video
stream to the host. However, RTL-SDR can't use that mode - it uses a
"bypass the whole Digital TV specific stuff" mode and directly passes
IQ samples through USB.
In that mode, it simply can't do more than 2 or 3 MS/s (can't
remember), which isn't enough to cover 6 MHz - so everyone's right, you
can basically receive the AM black/white info at a partial bandwidth of
the ca 5 MHz of the luma signal, but you won't get any color
information that way, or audio with the same receiver as you do video.
Cheers,
Marcus
On Fri, 2018-08-24 at 12:22 -0500, Martin McCormick wrote:
First, I will talk about the things I know for sure. The
NTSC analog system as well as Pal systems in a lot of the rest of
the world had a lot in common with eachother. Both systems
transmitted an AM video signal in Vestigial single sideband mode
such that the carrier frequency was always about 1.25 MHZ above
the start of a channel. NTSC systems in the Americas also
transmitted an audio carrier in FM which was always 4.9 MHZ above
the video carrier. Pal systems used exactly the same type of
transmissions except that the 625-line video at 25 frames per
second made a slightly wider spectrum such that the audio and
video carriers were separated by 5.x MHZ, making each Pal channel
7 or 8 MHZ wide.
As others have suggested, you could probably get a
monochrome fuzzy image if you can get your sound card to sample
fast enough. You can also decode the mono sound by setting your
RTL receiver to behave just like a FM broadcast receiver but set
the frequency to whatever the video carrier frequency is plus 4.5
MHZ. if the video carrier is 55.250 MHZ, the audio will be at
59.75 MHZ. The deviation is 75 KHZ unlike FM radio which is 150
KHZ.
That would be a good simple test to see if you are
receiving the channel at all.
I am guessing that since the RTL chips were designed for
the European television market for cable and over-the-air
broadcasts, they can be sampled extremely fast since the digital
channels still take up the same bandwidth as their analog
ancestors.
Martin McCormick WB5AGZ
Anders Hammarquist <address@hidden> writes:
In a message of Fri, 24 Aug 2018 10:27:40 +0200, "Ralph A. Schmid,
dk5ras" writes:
Hi Andres,
just had a short look: doesn't NTSC use a nearly 6 MHz
bandwidth?
Best regards,
Marcus
Yes, no way with the RTL to catch NTSC, it does in SDR mode only
2.smth
MHz bandwidth.
Actually, you should be able to get a picture. The horizontal
resolution
will be
about half of what it would be for the full bandwidth, and no
colour (as
the colour
subcarrier at 3.58 MHz is outside the pass band). You want the pass
band
of the reciever
from just below the video carrier and as high as it will go.
/Anders
|