Inside Chess Online seems fastest. USCF has more commentary but it's on separate pages, indexed off their game page. Their diagram page offers commentary on the current move. Chess Center falls behind quickly, but has good commentary eventually.
My attempt to come to terms with USCF's multipages (moves for game 6):
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 |
10 |
11 |
12 |
13 |
14 |
15 |
16 |
17 |
18 |
19 |
20 : White
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 |
10 |
11 |
12 |
13 |
14 |
15 |
16 |
17 |
18 |
19 |
20 : Black
Chess.net offers the fastest live coverage via telnet with commentary from world champion Karpov and others. See their computer chess page, and be prepared to register (free, for password), and 'set shout off' and 'set kiblevel 2000' and 'set open 0' to get rid of the excess noise.
I need to add info here about IBM's java-applet, but I'll have to get this running for myself first!
This wwwebpage is mainly for my own use, to follow the 'live' commentary from various sites without the annoyance of frames or bookmarks, etc. But I've also included some background-links, and may also provide live updates mirroring the other sites during games 4-6. {{Doublebraces}} means a link to DejaNews archive-- these messages may start with an ignorable warning about 'Invalid Context'.
Here's my {{postmortem}} on my own attempt to provide semi-live coverage.
["!" means recommended. "?" means uncertain link.]
Chess Mentor, a downloadable analysis program with very fine, detailed commentary on selected passages from the first few games (so far)..
one |
two |
three |
four |
five |
six : full onsite commentary by Ashley, Seirawan, Valvo (long txt files)
one? |
two |
three? |
four? |
five? |
six? : the same, unedited
one |
two |
three |
four |
five |
six : GM Seirawan postgame
one |
two |
three |
four |
five |
six : GM McDonald et al
one |
two |
three |
four |
five |
six : GM Henley
one |
two |
three |
four |
five |
six : WC Karpov et al
one |
two! |
three! |
four |
five? |
six? : GM Schwartzman or King
one |
two |
three |
four |
five |
six : synopsis of others (too many images)
one |
three |
four : GM Wolff in annoying USCF multipage layout
two |
five |
six : GM Bisguier ditto
one |
two |
three |
four |
five |
six : IBM's annoying multipage version seems to entirely lack commentary?!?
Excellent prematch interview with Kasparov.
A long {{postmortem on game one}} using three computer programs.
IBM's excellent hotlinks to online press coverage and their own post-mortem on game two
An interesting {{debate}} about whether game one was lost while DB was still following its 'book'.
Debate about whether {{Kasparov resigned too soon}} in game two. Kasparov's computer expert {{on this question}}.
It's clear that Deep Blue's programmers have done a fantastic job, and that DB is now playing at grandmaster level (a goal AI has been expecting Real Soon Now since the 1950s!). It remains to be seen whether DB can win as black, though, and there appear to be some flaws, still, in its positional analysis. There's every reason to believe that these will be fixed by next year, though, so the world championship looks quite inevitable within a year or two.
One thing you notice in the commentaries is that even grand masters are out of their depth kibbitzing these games. (With most spectator sports, as with most arts, it's comparatively easy to be a good critic. With this level of chess, though, no one can be sure Deep Blue hasn't seen vastly more than them, so all bets are off...) For example, the commentators are suspicious when DB places a piece on an unexpected square, and then later moves it to the expected one. Does this imply poor judgment on DB's part, or is it the commentators' vision that's lacking???
It's interesting that the DB team is maintaining extreme secrecy about its positional algorithms, and one can only hope this will be relaxed after the match.
And realize that Kasparov's play in the first two games was designed to uncover DB's weaknesses, so he may do better in the later games.
The Barnet Chess Club site has theoretical explanations of computer chess.
A background essay on computer chess, using 2001's HAL as the example.
IBM's old Deep Blue page from before the first match, with earlier games.
Last year's match and some photos
IBM's site for match one. Overview of all six games of match one. Onsite GM commentary for game one of match one (long).
An old, speculative {{basic Q&A about Deep Blue}}.
An article about last year's match supposedly targeted for censorship by IBM.
Newsgroups: rec.games.chess.computer and rec.games.chess.analysis.
Inside Chess Online's page for match two, and USCF's rematch page.
IBM's confusing site for match two, and their useful, hidden index page. DB's little brother, DB Jr.
Long articles on Kasparov (excellent, by David Levy) and on Deep Blue (boring).
Kasparov's own page for the match. And a recent netnews article on {{Kasparov's record against humans}}.
A recent {{Newsweek article}} by Stephen Levy (scanned, with typos).
A current thread about Deep Blue's AI.
A reporter solicits {{questions for the DB programmers}}.
{{Recommendations}} for best chess program in various categories
A nice collection of computer-chess links.
[Up:AI in games] [Map:AI] [No Next] [Robot Wisdom home page] (Feedback)