Behind
Deep Blue
A review by James E. DuBois
The two main objectives successfully achieved in the book Behind Deep Blue,
Building the Computer That Defeated the World Chess Champion by Feng-Hsiung
Hsu were to capture the true essence of the contest and to clarify his role
in the project. Personally I was surprised to see at this late date that another
book on Deep Blue had been published. The author was quick to point out in his
preface that this is not an “instant chess book” and concludes the
story that encompassed a twelve year span of his life with making a conscious
decision to write a book about the match “to set the record straight”
and that he has stories to tell.
Many of those stories illustrated the mischievous nature of Hsu. How on earth
could Fred Rogers (of the Mister Rogers’ Neighborhood, the PBS
television show) ever managed to be included in a quasi scientific work such
as this if it didn’t involve a prank? The author seems to be well steeped
in American pop culture since he chooses to make a hilarious connection to the
Dilbert comicstrip.
The book indeed delves into some fairly heavy discussion of chip design while
maintaining an efficient balance with the chess aspect. Certainly this must
have been a concern when writing the book to not dumb down either topics, but
to include material that people in the know in both fields of expertise found
engrossing. I found the chip design chapters informative and never boring, while
in all honesty I did not read anything groundbreaking concerning chess.
It
was enlightening to discover how the Deep Thought name was derived and the fact
that Hsu was involved in a car accident that seriously hampered progress. Some
of the best tournament experiences of the machine to be captured in the chapter
entitled, The Race for the First Machine Grandmaster chronicling Deep
Thought’s coming out at the 1988 US Open held in Boston. Perhaps most
intriguing was his emotions during the matches, in particular the 1997 final
game when Hsu writes, “At the moment of Kasparov’s resignation,
I suddenly felt very tired. The 12 years of work was finally over. I should
have been exulting, but I was feeling empty inside. The game felt too easy,
although, in hindsight, it wasn’t. Without our hard work the year before,
Kasparov might have won the game. A part of me also felt robbed. I am not a
chess player, but the chess player in me was definitely disappointed. Winning
or losing the game, I wanted the last game to be a real fight. I wanted the
win to be another great game like game 2, except without the final mistake.
If it had to be another loss, I wanted it to be another hard fought loss like
game 1, preferably without the bad moves from Deep Blue.”
Kasparov vs Deep Blue 1997
Hsu tactfully sidesteps the controversies surrounding the rematch even though
I had the feeling throughout the entire book that he wanted to disclose more
concerning certain personalities. The book concludes with the chapter entitled,
Life After Chess that fully answers the question, “Where are
they now?”
Certainly of interest to visitors of this site is the inclusion in the book
of how ChessBase as a company played an integral role in the Deep Blue story
(or Kasparov’s story, depending on your point of view) and the development
of computer chess as we know it today. Fritz is frequently mentioned along with
some knocks on “commercial” chess programs.
A first prototype version of the final Deep Blue chip
Regarding computer chess in general, Hsu writes about the goal to, “...
finally solve the Computer Chess Problem [sic], creating a chess machine that
could beat the World Chess Champion in a match.” The computer chess problem
as discussed throughout the book is technically incorrect, but from a developer’s
aspect this was the “holy grail.” The computer chess problem proper
involves developing a machine to solve the game regarding the absolute best
moves by brute force. In essence the game would therefore be “played out”
or solved. Had Deep Blue truly solved the computer chess problem, in game one
of the rematch after Kasparov made his first move of Nf3, Deep Blue would have
known the outcome of the game with best play by white. It’s not in the
realm of this review to even consider what would be required to actually solve
the problem of chess.
Princeton University
Press produced a physically solid book with a somewhat predictable, but
certainly attractive dust jacket. Missing is an index and glossary. Hsu’s
use of footnotes sometimes added little to the text and I found him to be redundant
on occasion. The appendixes provided a detailed account of Hsu’s life
in Taiwan and some important references for further reading.
Behind
Deep Blue deserves to be placed alongside Monty Newborn’s epic, Kasparov
Versus Deep Blue, but for a different reason; Feng-Hsiung Hsu presented
the match in the distinction in which he intended for his book, not as man versus
machine, but man as toolmaker.
Links
