 | | Tasks vs. program slots —Mark Ferris
|
| | | I am curious to clarify something. The NQC documentation states that an NQC program can have up to 5 tasks (I think that's the right number). In addition, the brick itself has the 5 program slots. Are they related or is it a coincidence of (...) (25 years ago, 17-Sep-99, to lugnet.robotics.rcx.nqc)
|
| | | |
| | | | | Re: Tasks vs. program slots —Matthew Miller
|
| | | | | (...) I believe that there's a maximum of ten tasks. So it would be a coincidence, but it isn't. :) (25 years ago, 17-Sep-99, to lugnet.robotics.rcx.nqc)
|
| | | | | |
| | | | | | | Re: Tasks vs. program slots —Dennis Clark
|
| | | | | | (...) Now you have ME wondering. Can we build 5 programs each with 10 tasks, each of these programs using all of the RAM, max program space, etc.? I have yet to do anything that ambitious, but it would be cool to switch behaviours by just changing (...) (25 years ago, 17-Sep-99, to lugnet.robotics.rcx.nqc)
|
| | | | | | |
| | | | | | | | Re: Tasks vs. program slots —Matthew Miller
|
| | | | | | (...) Not each using the _total_ max of course. But yeah. (...) Yep. I do this all the time. I often make a with-sound and a silent version, for example. It's also nice if you have a version of your program which works well but you want to make some (...) (25 years ago, 17-Sep-99, to lugnet.robotics.rcx.nqc)
|
| | | | | | |
| | | | | Re: Tasks vs. program slots —Dave Baum
|
| | | | (...) RCX contains 5 programs. Each program can have up to 10 tasks and 8 subroutines. Only one program may run at any given time, multitasking is between tasks in the same program. There is no known programmatic way of transferring control from one (...) (25 years ago, 17-Sep-99, to lugnet.robotics.rcx.nqc)
|
| | | | |
| | | | | | Re: Tasks vs. program slots —Matthew Miller
|
| | | | (...) This is why it might be nice to get the remote control reverse engineered -- we may discover a way. (25 years ago, 17-Sep-99, to lugnet.robotics.rcx.nqc)
|
| | | | |
| | | | | | Re: Tasks vs. program slots —Dave Baum
|
| | | | (...) The remote control always sends the same sort of packet: d2 xx yy where xxyy is a sixteen bit bit-field indicating what features should be activated: xxyy 0001 Message 1 0002 Message 2 0004 Message 3 0008 Motor A Forward 0010 Motor B Forward (...) (25 years ago, 18-Sep-99, to lugnet.robotics.rcx.nqc)
|
| | | | |
| | | | | | Re: Tasks vs. program slots —Matthew Miller
|
| | | | | (...) Too bad! So, the next thing I wonder is if there's any way to get an RCX to _send_ these packets to another RCX... (25 years ago, 18-Sep-99, to lugnet.robotics.rcx.nqc)
|
| | | | | |
| | | | | | Re: Tasks vs. program slots —Nick Taylor
|
| | | | Hello Dave, First, let me thank you for NQC ... you've made my RCX playing tremendously more rewarding. I've been wanting to use a Sharp IR receiver to demodulate the signal from a LEGO remote control, and then a PIC to decode the rx output for use (...) (25 years ago, 18-Sep-99, to lugnet.robotics.rcx.nqc)
|
| | | | |
| | | | | | Re: Tasks vs. program slots —Dave Baum
|
| | | | (...) What exactly does the Sharp IR receiver do? If it uses Sharp's ASK modulation, then you may have a problem. Ideally, what you want is something like a phototransistor to a good preamp, followed by a bandpass around 38kHz, then some sort of (...) (25 years ago, 18-Sep-99, to lugnet.robotics.rcx.nqc)
|
| | | | |
| | | | | | Re: Tasks vs. program slots —Nick Taylor
|
| | | | | (...) The Sharp GP1U581Y is a 38kHz receiver that incorporates an amp, a limiter, a bandpass filter, and a demodulator ... outputs a low while receiving an IR signal, a +5V high when not. It does NOT do any decoding. (...) From what I've been seeing (...) (25 years ago, 18-Sep-99, to lugnet.robotics.rcx.nqc)
|
| | | | | |
| | | | | | | Re: Tasks vs. program slots —Dave Baum
|
| | | | | | (...) They are one and the same. When transmitting, the IR tower just uses the RS-232 TXD line to switch on the 38kHz carrier. It also has some timing circuitry to keep it alive, and an Rx side, but the Tx is painfully simple. You'll probably see (...) (25 years ago, 18-Sep-99, to lugnet.robotics.rcx.nqc)
|
| | | | | | |
| | | | | | Re: Tasks vs. program slots —Dennis Clark
|
| | | | (...) Dave, It is basically just exactly what you describe. It usually takes about 200us or so to "see" the incoming 38KHz signal, common I suppose for an integrator to react to the signal getting through the band-pass filter, then a Schmidt trigger (...) (25 years ago, 20-Sep-99, to lugnet.robotics.rcx.nqc)
|
| | | | |
| | | | | | Re: Tasks vs. program slots —Dave Baum
|
| | | | (...) Yes, it appears to use the same packet level formatting. It does use a new bytecode (0xd2) that isn't seen elsewhere. This bytecode needs two bytes of data to follow. These three bytes of "payload" are encoded the usual way with a three byte (...) (25 years ago, 20-Sep-99, to lugnet.robotics.rcx.nqc)
|
| | | | |
