Sorry, Pete

Commenting on my recent computer vs computer Jerome Gambit (1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+) matches (see "I can't seem to get the hang of these things...") Pete Banks ("blackburne"), a long-time member of the Jerome Gambit Gemeinde, posted a Comment

Hi Rick,

Have you tried matching two equally-strong computer programs? It might be interesting to see what percentage White got. I'd guess about 40%.

Pete

While I liked Pete's enthusiasm for the Jerome Gambit, my Comment reply was not as optimistic as his. After all, while I have supported the use of the Jerome when giving "odds" to a weaker player; I'm not at all confident in its prospects when the players are evenly matched.

Hi Pete,

I'll ask Rybka to play a similar match against itself. I suspect it will run something like 0-20.

But, then again, what do I know?? You see, I can't seem to get the hang of these things...

Rick

The first thing that I learned is that using the Aquarium interface I wasn't able to have Rybka 3.0 play a match against itself. However, I was able to have two similarly powerful Rybka engines face off: Rybka 3 Dynamic w32 vs Rybka 3 Human w32.

I set up a 20-game Jerome Gambit match, starting from the position after 4.Bxf7+. Each computer had 5 minutes per game. I gave Dynamic the White pieces first.

The match finished 0 - 20.

I then switched engines, giving Human the White pieces.

That match finished 0 - 20.

That's zero wins for the Jerome Gambit, against forty losses!

I've really got to stop running these experiments...

(I won't post the games, but they will be available in the 4th PGN file of games presented or referred to on this blog, when it becomes available.)

Kaissiber!

I've spoken quite positively of Stefan Bücker's awesome quarterly magazine Kaissiber, which focuses on games, analysis and history (see "To Infinity... And Beyond!", "Breaking News...", "My head is spinning" and "Jerome Gambit Blog: Tidying Up") -- especially unorthodox openings.

Kaissiber 33 is due out today (unless you are a subscriber; in which case you have already received your issue, lucky person) and is well worth tracking down. (Word is that it has some killer King's Gambit analysis...)

If you are interested in collecting back issues of Kaissiber, it is important to know that they are available in bound editions, with the fourth volume due out this coming January (or late December).

Finally, for those of you who have interest in such things, Stefan also has copies for sale of my (with Riley Sheffield; Introduction by Hugh Myers) Marshall Gambit in the French and Sicilian Defenses which focuses upon 1.e4 e6 2.d5 d5 3.Nc3 c5!? and 1.e4 c5 2.Nf3 e6 3.d4 d5!? (and which is much, much sounder than the Jerome Gambit!)

One (or both) of us needs help (Part II)

Clearly the computer's revenge match had gotten off on the wrong foot – see "One (or both) of us needs help (Part I)."

For game two I set its skill level to 1400, and we played from the initial Jerome Gambit position (1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+) again.

Kennedy - WeakDelphi (1400)
blitz 2 12 (2), 2008

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+ Ke8 6.Qh5+ Ke7 7.Nxc6+

We both quickly repeat our errors from the first match game – although mine are "inaccuracies" while its are "blunders."

7...Kf6 8.Qf5 checkmate

Yikes.

For our third game, I stepped the computer's skill level up another 200 points.

Kennedy - WeakDelphi (1600)
blitz 2 12 (3), 2008

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+ Ke8 6.Qh5+ Kf8 7.Qf7 checkmate

Not much to say about that one.

Finally, I bumped the computer's skill level up to 1800.

Kennedy - WeakDelphi (1800)
blitz 2 12 (4), 2008

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+ Ke8 6.Qh5+ Ke7

The move that gives Black the advantage is, of course, 6...g6.

7.Qf7+ Kd6 8.Qd5+

Missing the mate-in-one: 8.Nc4.

8...Ke7 9.Nxc6+ Kf6

One last gift.

10.Qf5 checkmate

No more experiments for me – for a while, anyhow.

graphic by Jeff Bucchino, "The Wizard of Draws"

One (or both) of us needs help (Part I)

After my computer vs computer Jerome Gambit (1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+) match debacle (see "I can't seem to get the hang of these things") I was sure that I had done something wrong.

Eventually I discovered that the skill setting for WeakDelfi was set at 1200 – could that have been responsible for the engine's odd play?

Having injured the honor of the electronic beast, I figured that it would only be fair for me to allow it some form of revenge – across-the-board.

So I set up a blitz game (2 12) to start after move 4 of the Jerome Gambit. I set my opponent's skill level at 1200.

Kennedy - WeakDelfi (1200)
blitz 2 12 (1), 2008

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+ Ke8

Just. Plain. Bad.

6.Qh5+

I knew that this was not the best move, but I wanted to see how my opponent would handle it.

White would have a tiny advantage after 6.Nxc6 Qh4 7.d4 Qxe4+ 8.Qe2 Qxe2 9.Kxe2 Bb6 10.Nb4 Bxd4.

6...Ke7 7.Nxc6+

Missing: 7.Qf7+ Kd6 8.Nc4 checkmate.

7...Ke6

Clueless.

8.Qf5+ Kd6 9.Qd5 checkmate



Ouch!

graphic by Jeff Bucchino, "The Wizard of Draws"

Triangulation

Ward Farnsworth's web page http://www.chesstactics.org/ has a link to this Jerome Gambit (1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+) blog, so it only seems fair to point out in return that his two books Predator at the Chessboard A Field Guide to Tactics - Book I: Introduction, The Double Attack, The Discovered Attack and Book II: The Pin and the Skewer, Removing the Guard, Mating Patterns, were winners of the "Perry PawnPusher 99% Solution Award" at Chessville.

To quote from my review:

I found it to be "excellent for a ‘rusty’ player who wants to get back his tactical chops by re-thinking the process of piece interplay; or for class/ club/tournament players (like myself) who want to un-retire from the 64 squares and get back in the action, without looking like fools. It is an excellent resource for chess coaches or teachers working with middle school or even elementary school students.

The Jerome Gambit Gemeinde lives and dies by tactics, and Predator at the Chessboard is an unmatched resource, lest we, like the thunder lizards of old, disappear too soon from the face of the earth.

I can't seem to get the hang of these things...

After a few unfruitful days of trying to coax a Jerome Gambit game (1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+) out of a lucky opponent at the Free Internet Chess Server (FICS) I decided to put together another Computer vs Computer Jerome Gambit match.

Despite feeling a bit like Jar Jar Binks trying to command the Death Star, I turned to my new Deep Rybka 3.0 Aquarium, and discovered two interesting engines: Delfi (reportedly rated 2500) and WeakDelfi (supposedly rated 1900).

Perfect!
I arranged for a 20-game match (5 minutes per side per game), with the stronger Delfi playing the White pieces each time, starting from the Jerome Gambit position after four moves.

Delfi - WeakDelfi
5 minute game
1st of 20 game match

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+ Nxe5 6.d4

6...Bxd4 7.Qxd4 Nf3+


Huh???

What is this??
8.gxf3 Qf6 9.Qc4+ Ke7 10.Qc5+ Ke8 11.Nd2 Qd6 12.Qxd6 cxd6

13.Rg1 g6 14.b3 Nh6 15.Nc4 Nf7 16.Bb2 Rf8 17.0-0-0
This is not the kind of position that usually comes to mind when someone says "Jerome Gambit".

17...b5 18.Na3 b4 19.Nb5 Ba6


More strangeness.
20.Nc7+ Ke7 21.Nxa6 Ne5 22.Nxb4 Nxf3 23.Rg2 Nh4 24.Rg4 Nf3 25.Bg7 Rfb8 26.Nd5 Kf7
27.Bd4 Nxh2 28.Rh4 Nf3 29.Rf4+ Kg8 30.Ne7 checkmate



Nervously, I looked at the second game in the match.


Delfi - WeakDelfi
5 minute game
#2 of 20 game match
1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+ Black lost on time.

What???

Unreal!

For the record, the next 8 games looked like this:
Delfi - WeakDelfi
5 minute game
#3 of 20 game match

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+ Black lost on time.

Delfi - WeakDelfi
5 minute game
#4 of 20 game match

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+ Black lost on time.


Delfi - WeakDelfi
5 minute game
#5 of 20 game match
1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+ Black lost on time.


Delfi - WeakDelfi
5 minute game
#6 of 20 game match
1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+ Black lost on time.

Delfi - WeakDelfi
5 minute game
#7 of 20 game match
1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+ Black lost on time.


Delfi - WeakDelfi
5 minute game
#8 of 20 game match

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+ Black lost on time.


Delfi - WeakDelfi
5 minute game
#9 of 20 game match

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+ Black lost on time.


Delfi - WeakDelfi
5 minute game
#10 of 20 game match

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+ Black lost on time.


Nine straight losses, all on time!

I stopped the match. I didn't have to be a Jedi master to know how the remaining 10 games would turn out.

graphic by Jeff Bucchino, "The Wizard of Draws"

With a lotta help from my friend...

My "thank you" again goes out to N. Earl Roberts (see "Cha - ching!" and "Jerome Gambit: The Proper Perspective") who quickly noticed that the vast majority of entries in the third PGN file of games from this blog are computer vs computer affairs that start from a fixed position – and that such games are handled differently by ChessBase in its "Repertoire" function.

Using an editor, he therefore added the beginning Jerome Gambit (1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+) moves and the proper ECO code to each of those games.

Those who have received the earlier, unimproved version of the third file have been sent the updated PGN file.