Author: Robert Hyatt
Date: 06:25:07 09/29/99
Go up one level in this thread
On September 29, 1999 at 03:14:06, Ed Schröder wrote: >>Posted by José Berdiñas Bonefua on September 28, 1999 at 21:23:08: >> >>I am waiting the auto232 support for Rebel 10. >>You promise it!!. >> >>José > >This is not true. > >I promised that I would set free auto232 in case I was able to solve the >auto232 problems I have seen. That is quite something else. > >Anyway it is good that you bring up the topic as now after 8-9 months of >close investigations I can share some remarkable findings about auto232. > >I know now there is not a (big) problem other than that I have noticed the >below list of weird things which are good to know for everybody who use >auto232. > >#1. In rare cases I have seen Rebel's NPS drop with a factor of 3 during >an auto232 match. Thus playing a match on a PII-450 the machine at some >moment in the auto232 match the PC Rebel was running suddenly behaved >as a PII-150 machine. The factor 3 is remarkable because it is an indication >it smells to hardware and not to software. > >Of course the outcome of these matches are worthless and I wouldn't exclude >the possibility such things also happen on the machines of Rebel opponents. >The BIG problem is: how do you recognize it? Perhaps it happens a lot more >because it is difficult to notice it. I think that for the future I will add >some code to test the reliability of the program after each game. > >#2. I have seen Rebel and Rebel opponent crashes. Not that I consider an >auto232 crash as a big deal as after all auto232 is tricky software and you >are asking for trouble. But if after a crash the BIOS is corrupted too I >consider that as very serious. In such cases I consider such a match as >worthless as I have noticed in such cases Rebel or the Rebel opponent has >a big advantage in the match sore (silly match scores like 15-1) which is >a clear indication the other PC is totally nuts. > >#3. Quite often after "exit program" the memory of the PC is corrupted. I >have seen for example error silly error messages like, "Sector not found >abort, retry... " or other strange stuff. In such cases you sometimes >even can't run another program, you have to reboot first to get the PC >working again. Of course this raises the question, "how valid was the >match you just played?". > >#4. I have seen Rebel (and other programs) play moves that couldn't be >reproduced while Rebel is programmed to do so. I have not much of such >cases but the fact it happens isn't a funny thought as you don't have >the time to check every auto232 move. > >#5. I have seen Mchess give away a rook to Rebel without any reason as the >position was drawish. It was impossible to reproduce it manually. I saw the >Mchess case by accident because I was watching the game. > >#6. I remember a case (which is similar to #3) that after an auto232 session >the PC in question couldn't recognize the modem anymore when I tried to >log-in to the Internet on that machine. The most worse thing on this case was >that I was able to reproduce it time and time again. > >So far the list (which is not complete) but these are the cases I remember. >It made me conclude that auto232 is more fragile than I already thought. An >important second conclusion for me was that the problem isn't related to >Rebel but that all programs sometimes go nuts because of auto232 for >totally unknown reasons. > >I therefore see no reason any longer that Rebel10 should not support >auto232. Auto232 in Rebel10 becomes active if you copy a file. To be >more precisely do: > >COPY USERINFO.CFG to AUTO232.CFG > >Then check the menu OPTIONS (F3) and the "auto232" option should >be present. > >Ed Schroder > >Posted on CCC and Rebel-Board I have said this before, I will say it again. Auto232 is a piece of trash. It has always been a piece of trash. It still is a piece of trash. It will always be a piece of trash. Tony Hedlund and I spent many hours trying to figure out how Crafty would hang. Worked fine until it started probing the EGTBs rapidly, at which point the thing would hang, or the machine would crash every time. I gave up and any software that is so sensitive to how quickly or slowly a program produces a move (I have a delay function that is adjustable to help with this) or how quickly you access a disk drive, is, quite simply a piece of trash. Sorry for being so abrupt, but this is totally hopeless. The windows version seems to be more reasonable as it doesn't stuff an unknown piece of software into memory and hook on to the timer interrupt to check for moves by scanning video memory, etc. If you load the original auto232 software onto a windows system, you will notice that every time you look for it it will be gone. It will self-migrate to the trashbin. Exactly where it belongs. :) Bob
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.