To LUGNET HomepageTo LUGNET News HomepageTo LUGNET Guide Homepage
 Help on Searching
 
Post new message to lugnet.roboticsOpen lugnet.robotics in your NNTP NewsreaderTo LUGNET News Traffic PageSign In (Members)
 Robotics / 2218
2217  |  2219
Subject: 
Re: Automatic Gain Control
Newsgroups: 
lugnet.robotics
Date: 
Sun, 3 Jan 1999 23:01:27 GMT
Original-From: 
Laurentino Martins <(lau@mail)SayNoToSpam(.telepac.pt)>
Viewed: 
3839 times
  
> > Could it be that the AGC value is a measure for the distance between the
> > tower and the Cybermaster? Would the data you have gathered support this?
> 
> No, I would say definitely not. To me it seems like the tower increases the
> power when it has to retransmit, and tries to lower the power when
> everything works. So the reading is only connected to 'connection quality' I
> think. We had the highest reading with a short distance (3m) but the toilet
> (pipes and water) was between the tower and the MU. Going upstairs to the
> other end of the house lowered the reading again (and moving the MU just
> near the tower showed the lowest reading).


I remembered once something Dave told me: AGC is unsafe if used inside NQCC.

His words:
------------------------------------------------------------------------------

The AGC() macro is the only one to use a 5 bit selector code (0x10).  For many instructions (such as tests), the selector code must be fit into 4 bits.

I would avoid using it within conditions.  If you want to make a decision
based on the AGC value, assign it to a variable first, then test...

int foo;

task main
{
  foo = AGC();
  if (foo == 0)
  {
   ... do something...
  }
}

If you use AGC() directly in a test ("if (AGC() == 0)"), then there's a
chance that it will be assigned to the 1st argument of a test opcode, which
only supports 4 bit source types.
------------------------------------------------------------------------------


Laurentino Martins   *<||:-)

[mailto:lau@mail.telepac.pt]
[http://www.terravista.pt/Enseada/2808/]

--
Did you check the web site first?: http://www.crynwr.com/lego-robotics



Message is in Reply To:
  Re: Automatic Gain Control
 
Here's Anders Isaksson's anwer to the automatic gain control question. He emailed it to me directly but gave permission to post it, since I don't think many people know this: (...) No, I would say definitely not. To me it seems like the tower (...) (28 years ago, 3-Jan-99, to lugnet.robotics)

11 Messages in This Thread:





Entire Thread on One Page:
Nested:  All | Brief | Compact | Dots
Linear:  All | Brief | Compact
    

©2005 LUGNET. All rights reserved. - hosted by steinbruch.info GbR