On.. A Day In The Life Of the Hardcore Collector

[Steve B]

Many assume that collecting the old computers is all fun and games
Sadly..nothing could be further from the truth

sometimes there is hard, tireless,and thankless work to be done

an example...
the Mysterious Case of the Mephisto Miami and Mephisto Chess Explorer

it is generally believed amongst collectors worldwide that these two computers contained the same engine by Frans Morsch (of Fritz fame)
the only real difference between the two was thought to be in the housing.. with the Miami a peg sensory portable and the Chess Explorer a small pressure sensory table top
the Miami was released in 1998 with the Explorer a more recent release in 2005
estimated ratings by Mephisto for the Miami were in the 1900 elo range and so it was also assumed that the Explorer is 1900

however there was a troubling anomaly..
distributors of the Explorer advertised the computer at 1600 Elo..which is very unusual
you never see a a computer advertised as weaker then the company's own higher rating estimate

so ..we set about to begin the painstaking work comparing the two computers..
first we reviewed the two instruction manuals
with the exception of the sections dealing with the different housing the manuals were identical
same exact 64 levels,both computers have the same 6000 move book and all features and options are the same

next i played a game against both computers simultaneously with the following parameters
Pondering OFF(to eliminate the ability of one computer to be thinking while i am entering moves on the other computer)
Random Play OFF
Opening Book OFF
Selective Search ON
Time Contol..40/1Hr.45 Min

[White "Steve B"]
[Black "Miami/Explorer"]
[Time Control "40/1hr 45 min.]
1.a3 g6 2.h3 Bg7 3.Nf3 d5 4.d4 Nf6 5.e3 Nbd7 6.Bd3 O-O 7.O-O

[d] r1bq1rk1/pppnppbp/5np1/3p4/3P4/P2BPN1P/1PP2PP1/RNBQ1RK1 b - - 0 7

not more than 6 moves into the game and both computers do not play the same 7th move
in this position the Explorer plays 7..b6 and the Miami 7..c5

i also noticed that the displayed Evals were somewhat different as well with the Explorer generally about .1 different then the Miami on each move

further testing is needed but i am already fairly certain these two computers do not contain the same exact engines

XXX Hardcore Collecting Regards
Steve

[Steve B]

To confirm my suspicions that the two computers do not have the same exact programs i tested them on this position

arguably one of the most famous and beautiful moves of all time..
Game 9 Botvinnik v Capablanca..AVRO Tournament..1938
[d] 8/p3q1kp/1p2Pnp1/3pQ3/2pP4/1nP3N1/1B4PP/6K1 w - -

Botvinnik essays forth the hammer blow..30.Ba3!! and Capa.. a pawn ahead ..is brought to his knees in 10 moves..

here the two computers performed exactly the same..

after a full 8 ply search ..both took 7 min..55 sec and played Ba3 with an eval of 0.00
after a full 9 ply search ..both took 17 min..45 sec and played Ba3 with an eval of 0.00
and finally after a full 10 plys ..both took 1 hr.13 min.27 seconds and played Ba3 with an eval of 0.00


Hmmm Regards
Steve

[fern]

Give them a full day each and they will extract from you the very same yawn.

Free Advices& Co
Fern

[Steve B]

Give them a full day each and they will extract from you the very same yawn.

Free Advices& Co
Fern

Thanks for your words of encouragement
a typical example of the thanks and gratitude we caring collectors must endure

Altruistic Regards
Steve

[Steve B]

Not satisfied with the conflicting results of my two tests i set about to determine once and for all if these two computers are indeed the same

i pitted them against the CM9K Josh age 9 personality ( 1800 USCF) with pondering OFF

both computers had Pondering OFF,Books OFF , Random OFF and Selective Search ON
Time control was Game 30 Minutes..
the Mephistos had White..

[White "Mephisto Miami/Explorer"]
[Black "Josh - Age 9 #2"]
[Time Control "Game/30 Min.]

1.d4 Nf6 2.Nf3 d5 3.e3 c5 4.dxc5 e5 5.Nxe5 Bxc5 6.Nd2 Nc6 7.Bb5 O-O
8.Nxc6 bxc6 9.Bxc6 Rb8 10.Nb3 Qb6 11.Bxd5 Nxd5 12.Qxd5 Bd6 13.O-O Rd8 14.Qh5 Bb7 15.Nd4 Bf8 16.c4 Rbc8 17.b3 Qg6 18.Qxg6 hxg6 19.Rd1 Be7 20.Bb2 Bf6 21.Rd3 Rd7 22.Rad1 a5 23.h3 a4 24.Kf1 axb3 25.axb3 Ra8 26.b4 Ra2 27.R1d2 Rc7 28.Rc3 Ra4 29.b5 Rcxc4 30.Rxc4 Rxc4 31.Rd1 Rc5 32.b6 Rd5

[d] 6k1/1b3pp1/1P3bp1/3r4/3N4/4P2P/1B3PP1/3R1K2 w - - 0 33

up to move 32 both computers played exactly the same however here they diverge..
the Explorer plays 33.Bc3 and goes on to win the game
while the Miami plays 33.Rb1 and goes on to draw

Explorer game from 33.Bc3:

33.Bc3 Rd6 34.Rb1 Ba6+ 35.Kg1 Bb7 36.Rb5 Bd8 37.Ba5
Bf6 38.Bb4 Rd8 39.Nf3 Rc8 40.Rc5 Rb8 41.Ne5 Be7 42.Rc4 Bxb4 43.Rxb4 Rc8 44.Nc4 Bd5 45.Nd6 Rd8 46.b7 Bxb7 47.Nxb7 Rb8 48.Kf1 g5 49.Ke2 g6 50.Rb6 Kg7 51.Kf3 Kg8 52.Kg4 f6 53.Kf3 Kh7 54.Ke4 f5+ 55.Kd5 Kg7 56.f3 Kh6 57.e4 fxe4 58.fxe4 g4 59.hxg4 Kg5 60.e5 Kxg4 61.Rxg6+ Kh5 62.Rb6 Kg5 63.e6 Kf6 64.g4 Rg8 65.Rb4 Rb8 66.Kd6 Re8 67.Nc5 Kg5 68.Ne4+ Kg6 69.Rb7 Rd8+ 70.Ke5 Rf8 71.e7 Ra8 72.Nd6 Kg5 73.Nf7+ Kxg4 74.Nd8 Ra3 75.Rb4+ Kf376.Rf4+ Ke2 77.e8=Q Re3+ 78.Re4 Kf2 79.Ne6 Rg3 80.Qh5 Re3 81.Qh2+ Kf1 82.Qf4+ Ke1 83.Qxe3+ Kd1 84.Rd4+ Kc2 85.Rd2+ Kb1 86.Qg1# 1-0

FINAL POSITION -Explorer v Josh
[d] 8/8/4N3/4K3/8/8/3R4/1k4Q1 b - - 0 86

Miami game from 33.Rb1

33.Rb1 .Ba6+ 34.Ke1 Bb7 35.g3 Rd8 36.Ke2 Ba6+ 37.Kd2 Bb7 38.Kd3 g5 39.e4 Rd6 40.Bc3 Bd8 41.Ba5 Bf6 42.Rb4 Ba6+ 43.Ke3 Rd7 44.f4 gxf4+ 45.gxf4 g5 46.f5 Bb7 47.Ra4 Rd8 48.Nb3 Rc8 49.Bb4 Bd8 50.Bc5 Bxb6 51.Bxb6 Rc3+ 52.Kd4 Rxb3 53.Bd8 Rxh3 54.Ra7 Rh1 55.Bxg5 Rd1+ 56.Ke5 Rb1 57.Be7 Rb5+ 58.Kf4 Rb1 59.Bf6 Rb4 60.Bd4 Bc6 61.Rc7 Rxd4 62.Rxc6 Rd1 63.Rc7 Kg7
64.e5 Rf1+ 65.Kg4 Rg1+ 66.Kf3 Rf1+ 67.Ke4 Re1+ 68.Kd4 Rd1+ 69.Ke3 Re1+ 70.Kf4 Rf1+ 71.Kg5 Rg1+ 72.Kh5 Rh1+ 73.Kg4 Rg1+ 74.Kh4 Rf1 75.e6 Rxf5 76.Kg3 Rf6 77.exf7 Rxf7 1/2-1/2

FINAL POSITION Miami v Josh

[d] 8/2R2rk1/8/8/8/6K1/8/8 w - - 0 78
Drawn

In Summary..
while both programs are very similar.. the results of this exhaustive analysis can only conclude that they are

Not Exactly Regards
Steve

[fern]

I am almost sure you can say that they are also equal to at least 3 or 4 other units I know, all from Morsh programming. And they are all a kind of derivation of first Fritz for PC, Fritz 1.....

Every computer maker is living without qualms on the basis of old ideas regards
Fern

[Steve B]

Every computer maker is living without qualms on the basis of old ideas regards

i am finished with the Explorer.. i have squeezed out all there is to learn from this computer
soon i will turn my gaze to the Explorer Pro.. released just last year ..the Ex Pro is reported to have the same exact Engine as the Saitek Cougar(2050 Elo)

we shall see...

Myth Busting Regards
Steve

[Steve B]

soon i will turn my gaze to the Explorer Pro.. released just last year ..the Ex Pro is reported to have the same exact Engine as the Saitek Cougar

we shall see...
Myth Busting Regards

And now we compare the Mephisto Explorer Pro and the Saitek Cougar
the Cougar was released in 1998 and has an elo of 2021(Selective Search),the Ex Pro was released in 2007
its not so thrilling to see a computer released just last year with reportedly the same exact engine as a computer released 9 years ago.. so lets see if we can debunk this ..like we did with the Explorer and the Miami..

first a comparison of the manuals shows the same exact 64 levels ,take backs of 50 ply and almost the same features and options
however notable differences soon appear..

the Ex Pro has a rather unique "Studies" feature with consists of 20 classic games allowing the player to load up a game and then at certain positions in the game the computer halts the game and sets up a 3 minute clock
the player is awarded points if the correct move is chosen ..after the game is over the points are added up and the player is given an Elo for the game with a maximum elo of 2650 awarded
more on this feature in a future post..

another glaring difference is that in the manual for the Ex Pro there is no mention of the option to turn off pondering,switch between Selective and Brute Search or to turn off the opening book
however these options are available but only when turning on the computer
once the options are chosen they disappear and can only be activated upon the next power on/off
some what bizarre to say the least...

tossing the manual aside i decided to test both computers on this position ..which is from the last round of the 2008 "Paris" Dedicated Chess Computer Tournament
this world famous event is conducted each year in Paris France by member Alain Zanchetta
held each Spring ,it brings hardcore collectors from all over the world together to compete
more here:
http://www.zanchetta.net/echecs/Paris2008/default.aspx

i was operating the Saitek Renaissance Sparc-4mb(2187) and the opponent was the Novag Star Sapphire(2171)
RS had white and was on the move
TC was 30 sec per move..
[d] 6B1/2k4n/2p2p1P/P1P3p1/3K3P/8/5P2/1b6 w - - 0 44
1.Bxh7! should win followed by 1..Bxh7 and then 2.hxg5 fxg5 3.f3
and whites King marches over to the Queen side to capture the c6 pawn with a a well timed sac of the a5 pawn to deflect the Black King out of position..and since the Black bishop is tied down on h7 white should be able to promote the c5 pawn

Sadly Sparky played the howler h4xg5 and went on to draw the game

i let both the Ex Pro and Cougar chew on this position for a full 13 ply search which took close to 10 hours
they both chose the honker h4xg5 scoring the position as -.6

so far they appear to have the same engine..
more to follow regards
Steve..

[CRoberson]

soon i will turn my gaze to the Explorer Pro.. released just last year ..the Ex Pro is reported to have the same exact Engine as the Saitek Cougar

we shall see...
Myth Busting Regards

And now we compare the Mephisto Explorer Pro and the Saitek Cougar
the Cougar was released in 1998 and has an elo of 2021(Selective Search),the Ex Pro was released in 2007
its not so thrilling to see a computer released just last year with reportedly the same exact engine as a computer released 9 years ago.. so lets see if we can debunk this ..like we did with the Explorer and the Miami..

first a comparison of the manuals shows the same exact 64 levels ,take backs of 50 ply and almost the same features and options
however notable differences soon appear..

the Ex Pro has a rather unique "Studies" feature with consists of 20 classic games allowing the player to load up a game and then at certain positions in the game the computer halts the game and sets up a 3 minute clock
the player is awarded points if the correct move is chosen ..after the game is over the points are added up and the player is given an Elo for the game with a maximum elo of 2650 awarded
more on this feature in a future post..

another glaring difference is that in the manual for the Ex Pro there is no mention of the option to turn off pondering,switch between Selective and Brute Search or to turn off the opening book
however these options are available but only when turning on the computer
once the options are chosen they disappear and can only be activated upon the next power on/off
some what bizarre to say the least...

tossing the manual aside i decided to test both computers on this position ..which is from the last round of the 2008 "Paris" Dedicated Chess Computer Tournament
this world famous event is conducted each year in Paris France by member Alain Zanchetta
held each Spring ,it brings hardcore collectors from all over the world together to compete
more here:
http://www.zanchetta.net/echecs/Paris2008/default.aspx

i was operating the Saitek Renaissance Sparc-4mb(2187) and the opponent was the Novag Star Sapphire(2171)
RS had white and was on the move
TC was 30 sec per move..
[d] 6B1/2k4n/2p2p1P/P1P3p1/3K3P/8/5P2/1b6 w - - 0 44
1.Bxh7! should win followed by 1..Bxh7 and then 2.hxg5 fxg5 3.f3
and whites King marches over to the Queen side to capture the c6 pawn with a a well timed sac of the a5 pawn to deflect the Black King out of position..and since the Black bishop is tied down on h7 white should be able to promote the c5 pawn

Sadly Sparky played the howler h4xg5 and went on to draw the game

i let both the Ex Pro and Cougar chew on this position for a full 13 ply search which took close to 10 hours
they both chose the honker h4xg5 scoring the position as -.6

so far they appear to have the same engine..
more to follow regards
Steve..

13 ply on this position doesn't seem good enough.
I ran Rybka, Toga and an old Crafty(21.6) on it.

Rybka resolves it in 18 ply.
Toga resolves it in 20 ply. At 14 ply, Toga scores it 0.64.
Crafty 21.6 doesn't see it at 22 ply. At 13 ply, Crafty scores it at 0.62.

So, this position at 13 ply is insufficient to say the machines are/aren't the same.

[Steve B]

So, this position at 13 ply is insufficient to say the machines are/aren't the same.

agreed ..hence i said .."so far" and i of course would not declare them the same based on only position testing

am working on a game now where both computers play CM personality Josh age 9(no pondering) ..my favorite dedicated computer opponent as of late

still.. its not so much that both computers did not find the correct move.. as it is that they both pondered exactly the same during those 10 hours
moves searched were the exactly the same ..depths reached at the same exact times ..evals..etc etc

Game to follow regards
Steve