English Chess Forum

This is prompted by a post on another thread (http://www.ecforum.org.uk/viewtopic.php?p=40641#p40641), in which Kevin Williamson mentioned a player using tablebases in a correspondence game.

I played correspondence chess (CC) in the early 1980s, when getting to a chess club proved difficult. I found it extremely helpful - I started as a 120 grade player who hadn't played seriously for a couple of years, and when I returned to OTB play found that I'd risen to over 140, essentially as a result of the study that I'd done for CC. I stopped playing CC in the mid-80s.

Back then, you were allowed to consult books, but not to obtain other sources of help (in other words, you couldn't ask other players for assistance). No computer was strong enough to be of any use beyond, perhaps, checking for obvious tactical errors.

I made a brief return a few years ago, playing one game per season for my county. Naturally, I refrained from using computer assistance, believing (wrongly, as it turns out) that it was banned. I gave up when I realised that my final opponent (whose grade was over 20 points below mine) was not doing the same. In a highly complex middlegame, full of deep tactics that I was taking days to work out, my opponent was replying practically by return of email. After the game (which of course I lost), I set an engine on the game and found that yes, every single one of his moves after we left published theory matched the computer's choice.

I'm not complaining about this - it was within the rules. But really, what is the point of CC when you let a computer do all the heavy lifting? I know that there remains a role for human judgement and that a GM using an engine is stronger than an engine alone, but for ordinary players there really aren't many opportunities to improve on an engine's play. So what is the attraction of CC today?

Well, I'm a pretty ordinary player (172ECF) and I find there are plenty of opportunities to improve on computer suggestions. The machines remain weak at long-term planning (this is particularly noticeable in openings - I have the engines' books turned off - and, more usefully, in endings: for instance, they still don't get that "all" rook-and-pawn endings are drawn!). I play on LSS where computers are allowed and are widely used, but it's no draw-fest - my drawing rate is under 40 percent. My reason for playing is simply to improve my OTB play (dubious opening lines get weeded out fairly quickly!) but I find corr play by server is fun in itself.

Tim Spanton wrote:Well, I'm a pretty ordinary player (172ECF) and I find there are plenty of opportunities to improve on computer suggestions .... I play on LSS where computers are allowed and are widely used, but it's no draw-fest ....

There are also websites where engine use is not allowed - although they seem to vary as to how much effort they put into enforcing this rule.

Not too many opinions expressed! Perhaps CC is very much a minority interest these days.

I agree with Tim that engines have their problems in openings and endings. Having used them for opening preparation myself, I know how difficult it can be to get to anywhere close to the truth, even with silicon assistance. As for endings - well, Stockfish told me very quickly where I failed to find the best lines in a four-rook ending last night, but whether the +1.2 I had at one stage would have resolved to a win or a draw is anyone's guess without detailed analysis, which I think was Tim's point.

I find it interesting that in the online world (as discussed in the Computer use thread) engine use is generally considered cheating, but that the CC world appears to accept it. Perhaps the rate of play is the difference - the slow pace of CC should mean that major errors are rare and that good players will almost always find moves that come near the top of computer choices, making engine use hard to detect or prove.

The top players love it when their opponents just believe what the software tells them. Email and servers are replacing traditional postal chess, hence removing the long wait to get a reply from someone on a different continent, and also stopping people inaccurately recording the date of receipt.