Testers wanted for new program

[Freddie]

Did you actually test that?

I wrote that poorly. Thanks for pointing it out. I just rewrote it.

It mixes up three things -- the number of PCs on the network, the number of copies of HotkeyNet that are connected to each other, and the number of copies of HotkeyNet to which a single hotkey sends commands.

The first point has been tested on a network with many millions of PCs, the Internet. The other two points have not. However there's nothing in HotkeyNet that places a limt on those numbers, and the program uses the operating system's heaviest duty communications methods -- asynch socket calls and a thread pool -- which are the same techniques used by web server software.

I suspect that if there's a limit here it's point (3), and the limt is either a hard limit on the number of socket connections in the OS or a soft performance limit (10,000 simultaneous sends are going to be, at best, very slow).

Did you actually test that? (e.g. with broadcasting movement keys where you send 10 keys per second).

HotkeyNet doesn't send movement keys that way. It sends one press and one release.

One of these days I'll probably add typematic repeats to the program, but they won't be implemented in the way you describe. They will be generated by the copies of HotkeyNet on the target PCs. There's no reason to clog up the network with them.