>>... how much bandwidth does a single TV channel take ? Let's be generous and say 10 MHz. Multiply that by 840 channels (you do the math) and we get 8.4 GHz total bandwidth. Where does the 93 GHz come in ? Perhaps that's the center frequency of the IF modulator; so the fractional IF bandwidth used is 9 %. What's so dramatic about that? Maybe they used 840 televisions all at once? <<
Several things about your observations, and about the premise that SR employed:
Typically, TV channels on cable systems take ~ 6 KHz ea. Digitized video, if it is NTSC quality would result in somewhat more than 45 Mb/s, uncompressed, maybe as much as more than 100 Mb/s depending on the modulation and encoding schemes used. For compressed video suitable for commercial delivery over many cable systems we might view this as being sent at a reduced digital rate of 45 Mb/s.
I believe that what you have done is a common mistake, and that is to confuse MHz with Mb/s.
If you take the compressed video signal at 45 Million bits/sec and multiply it by 840 channels, we wouldn't get what the company claimed, but it's up there:
840 * 45 Mb/s = ~38 Gb/s.
If this were actually uncompressed, then, or nearing HDTV transmission speeds compressed, we'd reach their claim or surpass it, of roughly 90 or 100 Gb/s, again, depending on the encoding scheme.
But why use TV signals? Video is ultra forgiving of errors in transmission due to the poor accounting of the human eye, as is audio for the ear. I would think that a more convincing argument, albeit a more boring one, would have included some means of attesting to the accuracy of the received data... such as using a bit error rate measurement arrangement with an Andersen CPA standing by taking notes, like they do with the state lottery results. Nah... nothing glitzy there.
One potentially good thing about this whole affair, is that Malone may have at last had an opportunity to witness what 500 or more channels actually looks like. ;-) |
