Subject:
|
Re: RCX Firmware Speed
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Wed, 10 Apr 2002 12:43:23 GMT
|
Original-From:
|
rols@rols.org^spamcake^
|
Viewed:
|
617 times
|
| |
|
|
Firstly, thanks very much for that detailed answer. That
was exactly the information I was looking for.
Somewhere in the thread it was suggested that each task
gets one opcode executed every 3ms, actually, as you show,
they all get one opcode executed each time around the
main loop. Hence, to get better performance, you should
have more, not less, tasks. For instance that means that
my driving task, which controls both motors, would be
better split into 2 tasks, one for each side, and would
then run almost twice as fast. That's definitely something
I'm going to test.
I'll also take a look at the generated byte code for the
code and see how optimal it is. I could probably, with a
little tweaking, get the 'step' calculation to take a few
less opcodes, which means quite a large difference in the
number of times it runs per second.
I should break out leJos too and try that out.
Roland
On Tue, 09 April 2002, "Dick Swan" wrote:
>
> This is a follow up to many of the recent posts on the speed of Lego's
> RCX byte code interpreter. It's a long post but the salient items are
> in the first few paragraphs.
>
|
|
1 Message in This Thread:
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
