On 10/25/07, Andre Pouliot <[EMAIL PROTECTED]> wrote: > So a GPU would probably not be able to pull it off. A carefully design FPGA > could possibly make it but there would still be some heavy work to be done > by a cpu to organise the data as required for transmition. A special adaptor > would be required between the connector and the communication channel used.
Let's consider the most basic function we'd be using in this, which is a digital FM encoder. Using lookup tables and some other logic, it would be easy to generate a sine wave of any frequency. Then to encode an FM signal, we just modulate the freqency. To ensure there are no discontinuities, we need to ensure that the input to the modulator does not change too fast. Oh, and doing amplitude modulation is trivial. If we want to generate a TV signal, to begin with, we need the carrier wave, so we have a sine generator that runs at that frequency. Channel 2 for NTSC is like 54MHz, which we MAY be able to approximate and modulate well enough through a 350MHz DAC. (If not, we can probably add different delay lines to each of the R, G, and B channels of our DAC so that when they're added, we get like GHz temporal resolution). In this, we can just use amplitude modulation to encode the syncs and video. For color, we generate a 3.58MHz subcarrier (for NTSC I anyhow), apply amplitude and phase modulation to that to encode the color signal, and apply that to modulate the B&W signal. Finally, we need a low-pass filter on the output to filter out the spikes caused by the DAC making discrete steps. Being sadly ignorant of RF technology, what am I missing? -- Timothy Normand Miller http://www.cse.ohio-state.edu/~millerti Open Graphics Project _______________________________________________ Open-graphics mailing list [email protected] http://lists.duskglow.com/mailman/listinfo/open-graphics List service provided by Duskglow Consulting, LLC (www.duskglow.com)
