Collector's Corner..2007 Annual Physicals ..Continue..

[Robert Weck]

if i was to be born again i would INSIST my parents taught me German for my future life

at least for chess computers, this would be an advantage (the same for Dutch)

i will test the position on my Dominator(2.04) and Modena

i realize this is not exactly the same as playing a full game and then waiting to see what move is selected but still...i can watch the displays to see if the move lists are similar regardless if the move is actually played

if i remember correctly, there was another move displayed for a very long time in Modena's display; just at the last seconds Bf3 was the favorite.

I did not notice, how it was for the Commander...


Robert

[Steve B]

if i was to be born again i would INSIST my parents taught me German for my future life

at least for chess computers, this would be an advantage (the same for Dutch)

i will test the position on my Dominator(2.04) and Modena

i realize this is not exactly the same as playing a full game and then waiting to see what move is selected but still...i can watch the displays to see if the move lists are similar regardless if the move is actually played

if i remember correctly, there was another move displayed for a very long time in Modena's display; just at the last seconds Bf3 was the favorite.

I did not notice, how it was for the Commander...


Robert

Hmmm Regards
Steve

[Steve B]

hi Robert

i tested the position on the Modena

the move it chooses first is a2-a3 and then after 48 seconds it switches to Bf3..the move it expects black to make next is Qd3 and not Qxf3
after Bf3 i then allowed the modena to make the move for Black and after first choosing Qd3 it switches to Qxf3 after 92 seconds

so here it understands the blunder on its own and makes the correct move in reply


i happened to have a few other Morsch computers handy so i also tested this position on the Milano Pro, Marco Polo and Atlanta


Sure enough The little Marco Polo on Level 3(start up level) chooses and plays Bf3 as its move
and even the Milano Pro on level 1 will also play Bf3 but only for 6 seconds then it chooses Bb7

while for the Atlanta it immediately chooses Bb7(the first move we can see)but perhaps due to the much faster processor we cannot see if it considered Bf3 during its first search iterations


so i think it is clear that this is a blunder of the Morsch program(search routine/evaluation function) in several Mephisto computers(but i am sure not all) and probably the same for Morsch programs in other brands of chess computers and nothing to do with just the Dominator and Modena being almost clones


Have A Nice Day Regards
Steve

[Robert Weck]

so i think it is clear that this is a blunder of the Morsch program(search routine/evaluation function) in several Mephisto computers(but i am sure not all) and probably the same for Morsch programs in other brands of chess computers and nothing to do with just the Dominator and Modena being almost clones

ok, but there are too many other issues to make the relationship clear. They play relatively identical games (what the computers, that you tested, surely do not; even when they might be far relatives, what could explain Bf3)

maybe Frans forgot to remove some critical code?

Or perhaps all his programs are based on the same code and have been just modified?

Also the handling (e.g. initialising the self-test by holding knight- and pawn button when switching on is the same on Dominator and Modena), the bugs, the BT-results, the playing strength; it is all too close together...

And why shouldn't H&G recycle the old Dominator program? They would have been stupid, not to do it, would they?

As i said, it is no clone, but a close relative (improved(?) version)


best regards,
Robert

P.S.: the CPU in the Atlanta is not much faster than Milano pro's (just 20 vs. 16 MHz); that makes the behaviour, that you experienced, a little mysterious to me.

I don't know, if Atlanta's hash-tables could explain this?...

[Steve B]

Or perhaps all his programs are based on the same code and have been just modified?
As i said, it is no clone, but a close relative (improved(?) version)

yes i think that the same Morsch code is used over and over with slight improvements along the way even between different models from different companies
and i think is not such a good idea to show positions as proof that two computers are almost clones

Fritz Sends his regards
Steve

[Robert Weck]

yes i think that the same Morsch code is used over and over with slight improvements along the way even between different models from different companies

yes, there seems to be at least a little Dominator in every Morsch-program!

Btw.: It seems, that even some single versions of the Dominator are not absolutely identical!

My Commander 2.00 shows slightly different bugs than the version 2.00, that has been tested by the Austrian magazine "Modul" (the article "Die Akte Sphinx" was readable on the internet for some time)

Maybe Frans Morsch was continously working on his program and some changes were taken into production without giving it a new version-number...


mysterious Sphinx regards,
Robert

[Steve B]

My Commander 2.00 shows slightly different bugs than the version 2.00, that has been tested by the Austrian magazine "Modul" (the article "Die Akte Sphinx" was readable on the internet for some time)

mysterious Sphinx
Robert

ohhhh!
again you must mention your Commander!!



Bad Way To Start Off My Day Regards
Steve
PS..

[Robert Weck]

ohhhh!
again you must mention your Commander!!



Bad Way To Start Off My Day Regards
Steve
PS..

you must confront yourself with your threats and fears as often as possible, to overcome them!

I only try to help you with this!...


helpful regards,
Robert

P.S.: You have so many unique items in your collection, which i will never even see in real life; this should easily help you to stand this...

[Steve B]

you must confront yourself with your threats and fears as often as possible, to overcome them!

I only try to help you with this!...

its good advise Robert
better advise is if i am confronted each morning with a nice Commander on my desk
then for certain my fears are comforted

P.S.: You have so many unique items in your collection, which i will never even see in real life; this should easily help you to stand this...

you are a young man still
you have MANY MANY years to find rare/unique chess computers
as for me...sigh..the sands of the hour glass run thin
soon they will be empty altogether
and i am to go to my grave never having played against or matched the Commander against other computers

A Very Sad End To A Collectors Career Regards
Steve
PS..Sigh..

[Steve B]

Having tested all of the OSA modules(A4 through D10++, Brute Force,and the Sprac 4MB)..save one, i loaded the Ren board with the Sparc module which "only " has 1MB

Well known of course to contain the last chess Engine programming effort by the Spracklens and rated 2218(SS)

http://new.photos.yahoo.com/cpaa2@sbcgl ... 6881048/71


On the Eve of the historic clash between the Fidelity Excel Mach 68000(also by the Spracklens) rated 1845(SS) and Hiarcs .60 running on the original Amiga computer that competed in the famous 1989 Computer Olympiad (where the then worlds top programmers competed) and operated by Mark Uniacke himself (then and now), i wanted to insure the Excel was fit for battle
Mark is of course not only the world renown programmer of Hiarcs but an avid collector of the old computers himself

http://new.photos.yahoo.com/cpaa2@sbcgl ... 787570/135

to make up for the nearly 400 point discrepancy i set the S1MB for Game/15 Minutes and the Excel Game/30 min

In the Game the Excel had White and typically searched 4-5 ply and on average 1300 nodes per second
While the Sparc reached 5-6 ply .
Sparky was out of book first at move 9


[White "Fidelity Excel 68000"]
[Black "Saitek SPARC 1MB"]
[Result "0-1"]

1.e4 c5 2.Nf3 Nc6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 d6 6.Bg5 e6 7.Qd2 Be7
8.O-O-O O-O 9.f4 Nxd4 10.Qxd4 Qa5 11.Bc4 e5 12.Qe3 exf4 13.Bxf4 Bg4
14.Be2 Qc5 15.Bf3 Rfe8 16.Qd3 Bxf3 17.gxf3 Nh5 18.Bd2 Rac8 19.Rhg1 Bf6
20.Nb5 Be5 21.h4 Nf4 22.Bxf4 Bxf4+ 23.Kb1 a6 24.Nd4 Be5 25.Nf5 g6 26.c3
Qf2 27.h5 Kh8 28.Rgf1 Qb6 29.Nxd6 Rxc3 30.Qd2 Rd3 31.Nxf7+ Kg7 32.Qh6+


[d] 4r3/1p3Nkp/pq4pQ/4b2P/4P3/3r1P2/PP6/1K1R1R2 b - - 0 32
white is up a pawn and has what appears to be a strong attack
Black however is threatening mate on the move
the Excel fails to find the" Knock out punch" it needs to win the game and resorts to passive play to defend against the mate
and soon had a lost game

Kxf7 33.Qxh7+ Kf8 34.Qh6+ Bg7 35.Qf4+ Kg8 36.Qh2 Rxd1+ 37.Rxd1 gxh5
38.Rd5 Qe3 39.Qg2 Rf8 40.Rxh5 Rxf3 41.Rh1 Rf2 42.Qxg7+ Kxg7 43.Rd1 Qe2


[d] 8/1p4k1/p7/8/4P3/8/PP2qr2/1K1R4 w - - 0 44

here Sparky announces a mate in 4

44.Rd7+ Kf8 45.Rd8+ Ke7 46.Re8+ Kxe8 47.b4 Qb2# 0-1

FINAL POSITION
[d] 4k3/1p6/p7/8/1P2P3/8/Pq3r2/1K6 w - - 0 48


i guess twice the time allotment is not enough to make up for 400 elo

with the Excel found to be physically fit and ready to do battle .. ..

Let The Battle Cry Go Forth..
From This Time And Place..

Bring On Hiarcs Regards!
Steve