| |
Hello to all,
Thanks for all those who wrote in with suggestions for how to get
my RCX bricks to work with the PowerMac 5500s that I was having trouble
with. Here's a summary of my experiences and how I eventually got enough
working machine to teach the programming part of my class. Since it's
been so long since my first post, I've included it at the end of this
message.
First, what worked for me:
I was eventually able to get the systems administrators to cough up a
spare stand-alone PowerMac 5300 that ran the Robolab software
successfully. This machine, unlike all of the other machines in the
school, did not have the access-restriction software (Mac Control from
BDW) installed on them. I had very few problems getting the RCX with
either the black mac serial cable or the hybrid cable (consisting of a Mac
modem cable, a DB25-to-DB9 pin adaptor, and the pc serial cable) to work
with this machine. Unfortunately, this was the only working machine they
gave to me.
I was able to borrow a PowerMac 6300 and a PowerMac 8500 from another
school to use in my class. The positive side was that they both worked
perfectly in any configuration that I tried them in. The negative was
that I had to haul them around with me from one school to another.
This meant that I had three working machines when I had hoped to run the
class with four. Fortunately, the class size dwindled a little bit after
the 3rd week of class and I was able to reorganize the kids into three
groups, one per machine.
What didn't work:
I was never able to get the PowerMac 5500s to work properly. I got a lot
of helpful suggestions from folks about using the NQC downloader to assist
in the debugging. Unfortunately, due to the draconian nature of the
systems staff at the school, I was unable to (a) install the software
myself to test it out and (b) actually meet with them at a time that we
both could agree upon to get additional software installed. Thus, I was
never able to try NQC on these machines.
I was able to get the RCX firmware to download successfully only _once_
from the PowerMac 6400. I had been trying all of the "black magic"
approaches to exorcising the IR demons, including various lighting
conditions, various IR shielding setups, various cables, etc..., and
suddenly it started working. However, once the firmware had been
downloaded, it stopped working and I never got anything working properly
with it again.
We tried purchasing a Belkin USB to serial port adaptor for the one iMac
that was in the classroom. However, we were never able to get it to work.
The systems administrators even came over and fought with it for a few
hours, but had no luck either. No matter what setting we tried in either
Robolab or the adaptor (hot swapping it with the option key held down
didn't work), the green LED on the IR transmitter never came on.
Since I had four RCX bricks, cables, and IR transmitters, I swapped them
around the different machines to see if any of the LEGO hardware might
have been at fault. In every case, there was nothing faulty with the
hardware when I tried it on a machine that I knew worked.
I tried installing the 1.5.1 Robolab upgrade on the machines that we were
having problems with and had absoultely to luck. I called Pitsco-Dacta
tech support and was told to upgrade to version 1.5.1. When I patiently
explained everything I had done and asked whether there was anything else
that could be tried, all I got in response was silence and then an offer
for a refund for all the LEGO hardware. I tried talking to other techs,
but had no further luck.
--------
In all honesty, the most frustrating part was not the technical side of
things, but instead the massive resistance that I received when I tried to
get the assistance of the school's systems administrators. As I was an
outsider coming into the school to teach an afterschool class, I was very
much at the mercy of the powers-that-were. The only tech person available
to help out was the one that worked out of the school district office.
However, in every encounter we had with her, she was either
uncommunicative, unresponsive or downright rude. If it were possible, I
would have commandered one of their machines, wiped the drive, installed a
clean system, and then set about systematically debugging the system.
However, it was not. No one, not even the regular teachers in the school,
was allowed to remove the machines from the lab much less reconfigure them
for custom applications. If it were possible, I would have been at the
school with the techs whenever they were available to help out. However,
my work schedule made it impossible for me to get out there until just
before the techs left for the day (around 3:00pm). Due to the nature of
their access control setup, it was impossible for me to make any debugging
headway unless they were around. The rest of the instructors in the
building were not technically minded enough to help when I was not around,
either. When I asked the teachers to try to get the tech to do
something, I would usually find out that the tech blew them off and
nothing got done.
For instance, I wanted to try robolab on the PowerMac 6400, but couldn't
since it didn't have enough RAM. I couldn't turn on virtual memory since
all the control panels were locked away due to the access control
software. I asked the teacher to get the tech to turn it on when they
came in the next day so we could try it out. When I called back the next
day, I found out that the tech had told them that virtual memory was
likely to cause too many problems and she wouldn't turn it on. When I
came back to the school for the next class, I confronted the tech and
asked why she didn't want to turn on virtual memory. She told me that
things "would just run too slowly". I told her that my PowerMac 6300 was
using virtual memory and ran robolab just fine and that using VM was the
only way we could try to get things to work. When she realized she
couldn't brush me off anymore, she finally did as I asked. The really sad
thing was that everytime I wanted to try something, I had to go through
the same runaround. Thus, the only debugging that I ever got done was in
the 30-45 minutes that I had before my class in the afternoon.
So... I'm sorry that I couldn't have been able to provide more technical
information that could help other people in my situation. Since the class
was only six weeks and is now over, I have no more access to the trouble
machines and can't work on debugging them anymore. All I can say is to be
wary of the PowerMac 5500 and 6400. However, I can report that the 5300,
6300, 8500 and G3 (beige 266Mhz desktop) worked without a problem. I
heard from Lego that thr 5400 doesn't work, either.
To Dave Baum and the other NQC folks, I'm sorry I couldn't try the NQC
downloader to see if things worked with it. If I ever run into problems
with my own setups, I'll know what to try and who to contact. It's almost
frustrating that I _haven't_ had any problems with the machines that I use
on a regular basis becuase then I could actually try some of this stuff
out.
Thanks again to everyone who replied to me.
-Paul
---
Paul E. Rybski --- http://www.cs.umn.edu/~rybski
On Sat, 20 Nov 1999, Paul E. Rybski wrote:
> Hello,
> I'm teaching an RCX-based programming class using the Robolab
> software to a group of 19 4th,5th and 6th graders. I've been trying to
> get the Robolab software installed and running on the Powermac 5500/225s
> and 6400/200s owned by the school. Every time I try to run Robolab and
> download firmware or even just test the communication between the
> transmitter and the brick, I get an error saying that the macintosh could
> not communicate to the RCX and the graphical error display shows a red X
> over the transmitter, the cable and the RCX. However, I do see the little
> green LED inside of the transmitter turning on. I have tried both the
> modem and the printer port but have had no success. I have tried plugging
> the IR transmitter into the serial port before powering up the machine.
> What's particularly infuriating is that on my G3 at home and various other
> macintoshes where I work, I have no problems with any of the LEGO
> hardware or software.
> The school's systems administration staff are running an
> access-restriction program called Mac Control from BDW software. The
> systems staff assures me that that Mac Control will not affect access to
> the serial ports, but I have no way of verifying this short of contacting
> the company (which I am currently trying to do). The other possible cause
> of the problem may be that the serial ports on these macintoshes are not
> functioning properly. Does anyone have any experience with these
> particular macintosh models and the availablility (or lack thereof) of
> signals on the DTR and CTS lines?
> I don't know what version of the MacOS they're running (I'll have
> to check that the next time I'm there), but I see from
> <http://www.ceeo.tufts.edu/troubleshoot/> that I may have problems with
> the L2 Cache Reset. I'm using the black macintosh serial cable that came
> with the RCX. I've also tried using a three-piece cable consisting of a
> macintosh serial cable (hardware handshaking--what I use on my modem), a
> DB9-DB25 adaptor and the RCX PC cable. This hybrid cable has been
> successfully tested on my G3 at least.
>
> Any advice, words of wisdom or troubleshooting suggestions would
> be greatly appreciated.
>
> Thanks!
>
> ---
> Paul E. Rybski --- http://www.cs.umn.edu/~rybski
>
|
|
Message is in Reply To:
11 Messages in This Thread:
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
Active threads in Robotics
|
|
|
|
|
|
