Messages in this thread | | | Date | Fri, 12 Nov 1999 15:11:46 -0500 | From | "Theodore Y. Ts'o" <> | Subject | Re: Need info about serial port programming -UPDATE |
| |
Date: Fri, 12 Nov 1999 18:22:40 +0100 From: Jens Benecke <jens@pinguin.conetix.de>
> The bottom line is that this is a broken design/architecture; this kind > of sampling really should be done in hardware. The problem is that this > appears to be some kind of school project, with artificial design > constraints. Given that we generally don't optimize for broken designs,
That's right.. I would have used a proper sound card but our subject is "Sound recording and analysis in hardware" and the PC was originally supposed to be only for displaying the samples. (Now we use it to display average values as well, but that's another issue.)
Exactly; and if it is recording and analysis in *hardware*, the hardware should have a buffer where it records 100-200 samples, and then you have the computer collect the sample once a second. By having the computer sample under software control, you're adding RT constraints, and it really is not a great design.
I would be glad if there was. Otherwise we would have to do it in DOS and I have no initiative to start fiddling with interrupts and IO adresses.
... or RT Linux, which still force you to use direct UART access, but at least you'll have all of the amenities of Linux
- Ted
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |