> There has been an interesting discussion on the Yahoo group LinuxHam,
Just had a look - tnx. > The consensus is that even an RT Linux kernel has limitations due to > timing and that QRQ operation isn't feasable. > How well this is being > accomplished on the Win programs, I don't know. I don't have a scope handy, but I guess at a typical contest speed of 32 WPM, dots may be 10 ms or so. Allowing for + or - 10% jitter, it could be a bit too demanding for the kernel, and no doubt it is the same for any of the Windows strains Hence the hardware implementation (W1EL winkeyer) which may also be the best approach for Linux. It echos back to the PC any characters sent and that would allow real-time editing of the CW buffer. I see Rein has a linux driver on his todo list. It would be fun to knock out an open source version of the keyer hardware, which can't be that difficult. I have an AVR prototyping board kicking around, and Debian comes with AVR assemblers and loaders. Wilbert, ZL2BSJ _______________________________________________ Tlf-devel mailing list Tlf-devel@nongnu.org http://lists.nongnu.org/mailman/listinfo/tlf-devel
