Chess King Triomphe

This forum is for general discussions and questions, including Collectors Corner and anything to do with Computer chess.

Moderators: Harvey Williamson, Steve B, Watchman

Forum rules
This textbox is used to restore diagrams posted with the fen tag before the upgrade.
User avatar
spacious_mind
Senior Member
Posts: 4000
Joined: Wed Aug 01, 2007 10:20 pm
Location: Alabama
Contact:

Post by spacious_mind »

Ok here is game 1 comparison. I included CXG Computachess III as it was in close reach and also I am showing just for everyone to get a feel how a similar strength program looks like. In this case Ron Nelson who also happens to have different styles. This also will show how play styles influence moves.

Image

Absolutely for sure Yeno 301 XL and Chess King Counter Gambit are exactly the same.

Well I am assuming that CXG Computerchess is not quite as fast at 1.79 MHz as Chess King Counter Gambit. However Computerchess does have a tendency also to on occasion move quickly so its average time calculation works differently to Counter Gambit.

However at 82.% overall similarity which is the same as what Triomphe shows at 1 Mhz speed, I am still inclined to believe that the most likely programmer is Kaare Danielsen.

For the base comparison against all the computers shown I used of course Chess King Counter Gambit.

Anyway the tests will continue.

Best regards
Nick
bataais
Member
Posts: 135
Joined: Tue Oct 01, 2013 4:55 pm
Location: Switzerland

Post by bataais »

Thanks very much.

I replayed the same test 1 on the 301XL with Level 4 (15s) to compare it with Triomphe, and indeed it plays the same except 2 moves:

White 19. Rg5 instead of 19. Qc4, and
Black 6. ... Be6 instead of 6. ... Bf5

And here is test 2, yeno 301xl, 30seconds:

'6. ... Nb4'
'7.Nc3' '7. ... Nb4'
'8.Ng5' '8. ... Qf5'
'9.Bd3' '9. ... 0-0-0'
'10.Nd4' '10. ... Qe5'
'11.Nxe6' '11. ... 0-0-0'
'12.Nxe6' '12. ... Bf7'
'13.Nxd5' '13. ... 0-0-0'
'14.Nxc6' '14. ... Nxc6'
'15.Be3' '15. ... Bc5+'
'16.Be3' '16. ... Bb6'
'17.Nbxc7' '17. ... Bxe3+'
'18.Rxe3' '18. ... Bd7'
'19.Nxd5' '19. ... Qd6'
'20.Rd3' '20. ... Rhe8'
'21.Rxe8' '21. ... Re4'
'22.Qd3' '22. ... Re4'
'23.Qd3' '23. ... Nxb4'
'24.Qd4' '24. ... Nc6'
'25.Qd3' '25. ... Kxb7'
'26.Rb3+' '26. ... Kc8'
'27.Qb6+' 'RESIGNS'

White: 1609 Black: 1143 Combined: 1376

Automaton Regards
Michael
bataais
Member
Posts: 135
Joined: Tue Oct 01, 2013 4:55 pm
Location: Switzerland

Post by bataais »

test 3, 301xl, 30s

'4.Nc3'
'5.Qh5'
'6.Bxf4'
'7.Qh5'
'8.Bxf7+'
'9.Qh3+'
'10.Kd2'
'11.c3'
'12.Nxc3'
'13.Qg4+'
'14.Qxg7'
'15.Bxg8'
'16.Bxd6+'
'17.Bxd6+'
'18.Bd5+'
'19.d4'
'20.Bxa8'
'21.Bxa8'
'22.Qc3+'

White: 2047

'3. ... Nc6'
'4. ... b5'
'5. ... Nf6'
'6. ... Qb6'
'7. ... Bxg1'
'8. ... Kd8'
'9. ... Qxa1'
'10. ... Qf6'
'11. ... Bxc3+'
'12. ... Qxh1'
'13. ... Kd8'
'14. ... Nh6'
'15. ... Ngf6'
'16. ... cxb5'
'17. ... Kd8'
'18. ... Ka6'
'19. ... Nc5'
'20. ... Kb5'
'21. ... Kxb4'

Black: 1693

Combined: 1870

enough for today, have a nice Sunday
Michael
User avatar
spacious_mind
Senior Member
Posts: 4000
Joined: Wed Aug 01, 2007 10:20 pm
Location: Alabama
Contact:

Post by spacious_mind »

bataais wrote:test 3, 301xl, 30s

Black: 1693

Combined: 1870

enough for today, have a nice Sunday
Michael

Thanks you too. Should be finished soon with both game 2's.
Nick
User avatar
spacious_mind
Senior Member
Posts: 4000
Joined: Wed Aug 01, 2007 10:20 pm
Location: Alabama
Contact:

Post by spacious_mind »

Here are the results for Test 2 - The Turk (Automaton)

Image

Well just as soon as you get excited about the similarities in Test 1, Test 2 does exactly the opposite and shows no similarities :)

Best regards
Nick
bataais
Member
Posts: 135
Joined: Tue Oct 01, 2013 4:55 pm
Location: Switzerland

Post by bataais »

Thanks, interesting.

Again I replayed test 2 on the yeno 301xl, Level 4 (15s) to compare it with Triomphe. Only 4 moves differ:

White:
8. Ng5 instead of Nc3
10. Nd4 instead of Be3
21. Rxe8 instead of Qd3
Similarity: 85.71%

Black:
7. Qe6 instead of Bc5
Similarity: 95.24%
Total: 90.48%

(score white: 1865 black:1305 total 1585)

And Nick, the test4.xlsx does not work on my mac with open office, when I enter moves, it only shows some strange numbers and #NV..
So I skipped it and went on with test 5:

test 5, yeno 301xl, 30s:

'7.Bb3'
'8.Qf3'
'9.Qf3'
'10.a4'
'11.Bxe6'
'12.0-0'
'13.a4'
'14.a4'
'15.Nf3'
'16.Rae1'
'17.Kg1'
'18.Kh1'
'19.Kh1'
'20.Kh1'
'21.Kg1'
'22.Nd4'
'23.Rae1'
'24.gxh4'
'25.exd5'
'26.Qxg3'
'27.Nxg3'
'28.Kh1'
'29.Rg1'

White: 1458

'7. ... Be7'
'8. ... Qg6'
'9. ... a4'
'10. ... 0-0'
'11. ... Qxe6'
'12. ... 0-0'
'13. ... 0-0'
'14. ... 0-0-0'
'15. ... Bd4'
'16. ... b4'
'17. ... b4'
'18. ... Ng6'
'19. ... Rad8'
'20. ... Bb6'
'21. ... Nxe5'
'22. ... Rc8'
'23. ... Bc7'
'24. ... hxg3'
'25. ... hxg3'
'26. ... Nf4+'
'27. ... Nf4+'
'28. ... Rxh3'
'29. ... Nd3'

Black: 1821

Combined: 1639

Half Speed Regards
Michael
bataais
Member
Posts: 135
Joined: Tue Oct 01, 2013 4:55 pm
Location: Switzerland

Post by bataais »

Hi Nick,

I had to change cell AQ15 of G1_DATA (test 4) from 13.e5 to 13.e_5, because open office interpreted it as 1300000 ;)

Now the yeno 301xl tests are complete.

test 4, yeno 301xl, 30s:

'13.e5'
'14.d4'
'15.Rad1'
'16.axb5'
'17.axb5'
'18.axb5'
'19.axb5'
'20.Kh1'
'21.Bxa4'
'22.Rfc1'
'23.Kh1'
'24.Kh1'
'25.Rxa5'
'26.Rxa5'
'27.Rxa5'
'28.Rc1'
'29.Rc2'
'30.N1d2'
'31.Kh1'
'32.Rxd1'
'33.Re1'
'34.Ne3'
'35.gxf5'
'36.gxf5'
'37.Bxa3'
'38.Rxc1'
'39.g5'
'40.hxg6+'
'41.Nxe1'
'42.Kxg4'
'43.Nxf5'
'44.Kg3'
'45.Nf3'

White: 1457

'13. ... Bd7'
'14. ... cxd4'
'15. ... cxb3'
'16. ... bxa4'
'17. ... bxa4'
'18. ... bxa4'
'19. ... bxa4'
'20. ... fxg4'
'21. ... Nb5'
'22. ... Bxb5'
'23. ... fxg4'
'24. ... fxg4'
'25. ... fxg4'
'26. ... fxg4'
'27. ... Qxa5'
'28. ... fxg4'
'29. ... fxg4'
'30. ... fxg4'
'31. ... fxg4'
'32. ... fxg4'
'33. ... fxg4'
'34. ... fxg4'
'35. ... fxg4'
'36. ... fxg4'
'37. ... fxg4'
'38. ... fxg4'
'39. ... fxg4'
'40. ... Rxe1'
'41. ... fxg4'
'42. ... gxh5+'
'43. ... exf5+'
'44. ... Kg7'
'45. ... Ba4'

Black: 1789

Combined: 1623

Total 5 Games Combined: 1770

Hidden Tables Regards
Michael
User avatar
spacious_mind
Senior Member
Posts: 4000
Joined: Wed Aug 01, 2007 10:20 pm
Location: Alabama
Contact:

Post by spacious_mind »

bataais wrote:Hi Nick,

I had to change cell AQ15 of G1_DATA (test 4) from 13.e5 to 13.e_5, because open office interpreted it as 1300000 ;)


Hidden Tables Regards
Michael
Thanks glad you got it fixed, I am just completing game 5 should be able to post results in about an hour.

Regards
Nick
User avatar
spacious_mind
Senior Member
Posts: 4000
Joined: Wed Aug 01, 2007 10:20 pm
Location: Alabama
Contact:

Post by spacious_mind »

Here is Test Game 3 comparison:

Image

For the heck of it I also added Mephisto Europa.

Test Game 4 finished as follows:

Image

There was a deviation of 1 white move between Chess King Counter Gambit and Yeno 301 XL.


Here is also Test Game 5:

Image

In Test Game 5 there was also a white move deviation between Chess King Counter Gambit and Yeno 301 XL

Summary of Deviations

Image

Well I guess you can throw any Programmer you want into the WHO PROGRAMMED theory and still shed no light.

I could when I get chance test Scisys Mark V to see how that looks.

Test Table

Image

This table shows where Yeno 301 XL, Counter Gambit and CXG Computachess III ranked at the end of the tests.

1) CXG Computachess III = 1790 ELO
2) Chess King Counter Gambit = 1742 ELO
3) Yeno 301 XL = 1730 ELO
4) Chess King Triomphe = 1665 ELO


Unknown programmer regards...
Nick
bataais
Member
Posts: 135
Joined: Tue Oct 01, 2013 4:55 pm
Location: Switzerland

Post by bataais »

Thanks. At least the tests show strong similarities between Counter Gambit and Yeno 301XL and some between both and Triomphe. If you would play them on half speed (L4, 15s) and compare them to Triomphe, as I did with tests 1+2, more similarities would show.

To compare with a Mark Taylor, I think the next best thing would be a Mildton Bradley.

Two other 6301 4K chess computers were programmed by David Kittinger, Constellation Junior and Accort/Alto. When I tested an Alto, it played very differently. I don't think Chess King's David Levy would ask him while he was working for Novag.

Still Puzzled Regards
Michael
bataais
Member
Posts: 135
Joined: Tue Oct 01, 2013 4:55 pm
Location: Switzerland

Post by bataais »

I still believe it's a Morsch, take my problem 1 (Rd2 - mate-in-2):

https://www.schachcomputer.info/forum/a ... 1508064092
(A.W. Galitzky, Niwa 1905, #2, Rd2!, White: Kh5, Qa1, Ra2, Nf5, d3; Black: Kf3, Bc1)

I tested some Morsch (on the Analysis Level), they play:

Mondial II, Super Mondial, GK2000: Rc2 (leads to mate in 3)
Europa, Yeno 416XL, Yeno 301XL: Rb2 (leads to mate in 3)

All other chess computers I tested play either Rd2 or some move with Q.

This move (Rb2) is a clue to a similar search procedure in Europa and 301XL.
All my other problem examples confirm this, and I have even more. Try them on some other chess computers, you'll seldom get the same exact moves.

Stubborn Regards
Michael
bataais
Member
Posts: 135
Joined: Tue Oct 01, 2013 4:55 pm
Location: Switzerland

Post by bataais »

Well, I contacted David Levy and asked him about Triomphe, Counter Gambit and Yeno 301XL. Here is what he said:

"
So far as I can remember the 8-bit Chess King products used the Richard Lang "Cyrus" program. The 4-bit products used, I believe, a Mark Taylor program which was based on an algorithm I developed a few years previously.

I cannot be 100% certain of thos efacts but if you can tell me the markings on the ICs in your Chess King products I can confirm those details.

I am almost certain that frans Morsch did program a 6301 Hitachi chip.
"

So I sent him the markings..

Waiting for Reply Regards
Michael
bataais
Member
Posts: 135
Joined: Tue Oct 01, 2013 4:55 pm
Location: Switzerland

Post by bataais »

And here is David's answer:

"In that case please ignore what I said. This is a 6301 program which I think was written by Jon Green, but I cannot be 100% certain. At my age my memory is not what it used to be."

Does anybody know of a chess computer program by Jon Green?

"He worked for me for a few years."

Well, that's it.

Lost my Bet Regards
Michael
User avatar
Steve B
Site Admin
Posts: 10140
Joined: Sun Jul 29, 2007 10:02 am
Location: New York City USofA
Contact:

Post by Steve B »

bataais wrote:And here is David's answer:

"In that case please ignore what I said. This is a 6301 program which I think was written by Jon Green, but I cannot be 100% certain. At my age my memory is not what it used to be."

Does anybody know of a chess computer program by Jon Green?

"He worked for me for a few years."

Well, that's it.

Lost my Bet Regards
Michael
Jon Green
yet another name offered for the Triumphe and Counter Gambit

Never Heard Of Him Before Regards
Steve
User avatar
spacious_mind
Senior Member
Posts: 4000
Joined: Wed Aug 01, 2007 10:20 pm
Location: Alabama
Contact:

Post by spacious_mind »

Steve B wrote:
bataais wrote:And here is David's answer:

"In that case please ignore what I said. This is a 6301 program which I think was written by Jon Green, but I cannot be 100% certain. At my age my memory is not what it used to be."

Does anybody know of a chess computer program by Jon Green?

"He worked for me for a few years."

Well, that's it.

Lost my Bet Regards
Michael
Jon Green
yet another name offered for the Triumphe and Counter Gambit

Never Heard Of Him Before Regards
Steve
Well John Green beats "Unknown". I'll take it! :)

John "Doe" Green Regards
Nick
Post Reply