bob wrote:I'm actually not sure what they used. There was no personal computer for some of the chips they uses (ARM, etc). But you would assume they did have some sort of development system, although no doubt it was not as fast as the very fastest thing they could get their hands on...
But it seems it will never happen, even in today's world where most every PC has an ethernet connection built in...
Serial ports were ubiquitous on the early enthusiast machines and were standard equipment on most by the end of the 1970s. Ethernet took a bit longer, but by 1996 or so every Macintosh had an Ethernet port and the PC world followed only a few years later.
Although, not every one was using the same transmission media. In the early part of the 1990s I was using 10 Mbps coaxial bus topology on my home LAN, not the cheaper alternative of 10baseT twisted pair with a hub or switch. I didn't change until 1997 and now it's switched gigabit cat-6 cabling. Well, except for the WiFi-N wireless.
But even before any Ethernet, I was using a home-rigged 9,600 bps serial link and specialized software to connect pairs of machines to play chess. It was such a setup that I used in late 1994 to play my old program Spector against this young upstart named Crafty.
bob wrote:I have been doing these events since my first in 1976. The act of moving pieces and hitting the clock is *not* an attractive part of the process. It introduces random effects into the game (one side forgets to hit the clock, or has to go to the men's room and their program moves while they are gone, or a move is entered incorrectly, or the remaining clock time is entered incorrectly, or time gets low and stress goes up as you strive to make the moves as quickly and accurately as possible. I can do without that. I have, in fact, since the CCT-type events have been all done on ICC. Watching is far more fun than actually "operating". From years of experience.
I started rather later (1987) than Bob, but I too have had plenty of experience operating a chess program in otherwise human events and pretty much everything he writes on this is correct. Synchronizing the program clock with the physical clock was always a problem. The overall task was even more difficult if one had to haul around a desktop and terminal in those pre-laptop days, and having to do this without an assistant.
What is also true is that there have been a number of cases in the early years where an operator had apparently "assisted" the program a bit too much. I don't think this is a problem nowadays when the program is most likely far stronger than the operator, but having an automated move exchange process helps eliminate this misbehavior anyway.
If there is a drawback to automated move exchange, it's that the first dozen or so moves with both programs "in book" may be played nearly instantaneously or at least too fast for the spectators to follow. I'm not sure if much can be done about this.
bob wrote:I have been doing these events since my first in 1976. The act of moving pieces and hitting the clock is *not* an attractive part of the process. It introduces random effects into the game (one side forgets to hit the clock, or has to go to the men's room and their program moves while they are gone, or a move is entered incorrectly, or the remaining clock time is entered incorrectly, or time gets low and stress goes up as you strive to make the moves as quickly and accurately as possible. I can do without that. I have, in fact, since the CCT-type events have been all done on ICC. Watching is far more fun than actually "operating". From years of experience.
I started rather later (1987) than Bob, but I too have had plenty of experience operating a chess program in otherwise human events and pretty much everything he writes on this is correct. Synchronizing the program clock with the physical clock was always a problem. The overall task was even more difficult if one had to haul around a desktop and terminal in those pre-laptop days, and having to do this without an assistant.
What is also true is that there have been a number of cases in the early years where an operator had apparently "assisted" the program a bit too much. I don't think this is a problem nowadays when the program is most likely far stronger than the operator, but having an automated move exchange process helps eliminate this misbehavior anyway.
If there is a drawback to automated move exchange, it's that the first dozen or so moves with both programs "in book" may be played nearly instantaneously or at least too fast for the spectators to follow. I'm not sure if much can be done about this.
There are a couple of options. One could add a "delay" in the program that displays the game. For example, in xboard, it would be easy to add a delay so that a move is not displayed until N seconds has elapsed since the last move was displayed. And one could always back up and replay the moves (I do this on ICC all the time when I am watching a game)...
In the second round game Rybka 3 showed the difference btween the extremely good former champion and actually the best ever. There are Worlds between them. When Rybka 3 plays it's as if G-d would play chess. Last time I saw such a move when King walks away from his officer whom he normally would defend and protect - was with Monarch back in 1985. Horrible suicide. <smile>
-Popper and Lakatos are good but I'm stuck on Leibowitz
BubbaTough wrote:When I started dabbling with chess after a long absence I was delighted to see many human tournaments where you are able to follow along move by move over the internet...delightful! It really helped encourage and maintain my interest. I find it builds suspense watching games live compared to waiting to see results. Its like watching a soccer game, rather than reading about the results. I also enjoy watching Olivier's chess wars which has a very nice live broadcast with chat. I wish I had started participating years ago.
It is unfortunate that WCCC could not manage to replicate this functionality. I am still interested in the results, but am not following carefully like I would if there were a broadcast. How surprising it is is dependent on how much you know about the history of the WCCC I guess.
-Sam
I tried for years to get them to go to an automated system. Everyone could still come and play, but they could set up a local chess server and automate the games themselves to get the humans "out of the loop". Would be much cleaner, no chess clocks or boards needed, and it could be automatically broadcast.
Alas, no such luck. And they wonder why there is such a lack of interest. Someone is _not_ thinking.
The fact that this continues to be an issue with each technologically advancing year, reinforces the realization that the decision-makers involved here are determined luddites.
mhull wrote:The fact that this continues to be an issue with each technologically advancing year, reinforces the realization that the decision-makers involved here are determined luddites.
One problem here is that as far as I know, there isn't a good, easily buildable chess server source available. Also, newcomers to the field have had a poor record of communications robustness.
What we really need is a new, free and open server along with a new protocol. The old FICS/ICC standard is kludge upon kludge and it's time for it to be replaced.
