Pavilion Chessman Pro 2

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.
Post Reply
User avatar
appleshampogal
Member
Posts: 126
Joined: Sun Jan 12, 2014 8:53 am

Pavilion Chessman Pro 2

Post by appleshampogal »

I've been trying to find information about this particular chess computer, but details are scarce. According to the box it plays up the advanced (1800 elo) level. Apparently some Pavilion computers are also related to existing Excalibur machines as well. I'd like to see if any of you guys have experience with this particular machine and know what program it is related to / playing strength - characteristics etc. Thank you!
Tr0ntreez
Member
Posts: 24
Joined: Fri Feb 17, 2017 12:32 pm

Post by Tr0ntreez »

Hi Kat,

Might be related to this one:
https://www.schach-computer.info/wiki/i ... essman_Pro

Maybe yours is also programmed by Kaare Danielsen. Manufacturers will often re-use old programs.
https://www.schach-computer.info/wiki/i ... sen,_Kaare

How's the Star Diamond going?

Regards,

You Pal,

tt
User avatar
appleshampogal
Member
Posts: 126
Joined: Sun Jan 12, 2014 8:53 am

Post by appleshampogal »

Tr0ntreez wrote:Hi Kat,

Might be related to this one:
https://www.schach-computer.info/wiki/i ... essman_Pro

Maybe yours is also programmed by Kaare Danielsen. Manufacturers will often re-use old programs.
https://www.schach-computer.info/wiki/i ... sen,_Kaare

How's the Star Diamond going?

Regards,

You Pal,

tt

Yeah, I did see that on Spacious Mind's website, but I wonder what might the difference between the one you linked to and the Chessman Pro 2. Faster processor? A bigger word size? Just stronger in general? One amazon reviewer said that he was a 1500 rated player who felt challenged by the device (the Chessman Pro not the Chessman Pro 2) despite the relatively low rating of Spacious Mind gave the first variant. So, I'm not sure what to make of these conflicting reports. I'm hoping to get a better idea soon.

The Star Diamond is doing quite well! I'm considering having a shelf built to house all these beautiful machines as I have seven in total now. :D
User avatar
appleshampogal
Member
Posts: 126
Joined: Sun Jan 12, 2014 8:53 am

Post by appleshampogal »

Take this account here...


https://www.amazon.com/Lexibook-ChessMa ... B008A2AT2O

This is an excellent easy to use chess computer. Level 1 is suitable for beginners. Higher levels will challenge even the best players. Over the last 2 days I played the first 6 levels. My rating is about a 1500 club player. I would rate this computer about 1800 elo which is very strong. When I played level 1 it was easy to beat since it makes beginner's mistakes. Level 2 is a bit better but it still makes obvious errors. However level 2 puts up more of a fight making it perfect for experienced beginners or weak players. Level 3 gets challenging. I struggled a bit to win. Level 4 surprised me. I made an error and the computer saw it and took advantage. I lost but came back in a rematch which was a very close game. I then skipped to level 6 and lost twice. The next day I played 2 more games on level 6 and won 1 and lost the other. What a great chess computer.
I know you can't take everything you read online as absolute gospel, but it's certainly an interesting take nonetheless.
SirDave
Full Member
Posts: 610
Joined: Thu Sep 09, 2010 4:59 am
Location: Southern California USA

Post by SirDave »

appleshampogal wrote:Take this account here...


https://www.amazon.com/Lexibook-ChessMa ... B008A2AT2O

This is an excellent easy to use chess computer. Level 1 is suitable for beginners. Higher levels will challenge even the best players. Over the last 2 days I played the first 6 levels. My rating is about a 1500 club player. I would rate this computer about 1800 elo which is very strong. When I played level 1 it was easy to beat since it makes beginner's mistakes. Level 2 is a bit better but it still makes obvious errors. However level 2 puts up more of a fight making it perfect for experienced beginners or weak players. Level 3 gets challenging. I struggled a bit to win. Level 4 surprised me. I made an error and the computer saw it and took advantage. I lost but came back in a rematch which was a very close game. I then skipped to level 6 and lost twice. The next day I played 2 more games on level 6 and won 1 and lost the other. What a great chess computer.
I know you can't take everything you read online as absolute gospel, but it's certainly an interesting take nonetheless.
Fwiw, the above unit uses 4 AA batteries while the Pavilion Lexibook unit uses 3 AAs so even if the firmware is the same, the CPU will be running slower.
User avatar
paulwise3
Senior Member
Posts: 1505
Joined: Tue Jan 06, 2015 10:56 am
Location: Eindhoven, Netherlands

Post by paulwise3 »

Hi, I have the Lexibook Chesslight and the Lexibook Chessman Elite.
I looked up a picture of the Pavilion Chessman Pro II, and noticed that:
- the pieces are exactly the same shape as the Lexibook's, except for the knights, they seem much better.
- it also has a choice of four playing styles, which gives an indication that it could be a Kaare Danielsen program, probably the 4 kb ROM version.
- if it is, then the USCF elo rating cannot be more then 1500-1600, and it can solve mate up to mate in 4.

A chessfriend recently gave me some mate-in-3 problems with which he tested another chesscomputer which was suspected to also have a Danielsen program. I will look them up for you including the times to solve it. That will be a good indication.

Will be back soon regards,
Paul
2024 Special thread: viewtopic.php?f=3&t=12741
2024 Special results and standings: https://schaakcomputers.nl/paul_w/Tourn ... 25_06.html
If I am mistaken, it must be caused by a horizon effect...
User avatar
appleshampogal
Member
Posts: 126
Joined: Sun Jan 12, 2014 8:53 am

Post by appleshampogal »

paulwise3 wrote:Hi, I have the Lexibook Chesslight and the Lexibook Chessman Elite.
I looked up a picture of the Pavilion Chessman Pro II, and noticed that:
- the pieces are exactly the same shape as the Lexibook's, except for the knights, they seem much better.
- it also has a choice of four playing styles, which gives an indication that it could be a Kaare Danielsen program, probably the 4 kb ROM version.
- if it is, then the USCF elo rating cannot be more then 1500-1600, and it can solve mate up to mate in 4.

A chessfriend recently gave me some mate-in-3 problems with which he tested another chesscomputer which was suspected to also have a Danielsen program. I will look them up for you including the times to solve it. That will be a good indication.

Will be back soon regards,
Paul
Oh cool! I appreciate your input on that! Yeah, my gut was telling me that it might be the same 4 k program. In all actuality, it seems pretty amazing that such a tiny program could be that strong. How nuanced could an evaluation function be with little memory?
4 K seems like nothing to me, but then again programmers can do some amazing things. I look forward to hearing from you soon about those mate solve times!
User avatar
paulwise3
Senior Member
Posts: 1505
Joined: Tue Jan 06, 2015 10:56 am
Location: Eindhoven, Netherlands

Post by paulwise3 »

Here are two mate-in-3 tests for your Pavilion, assuming it has the 4 kb Danielsen program.
[fen]6k1/4p2p/7B/3K4/8/8/8/8 w - - 0 0[/fen]
White gives mate in 3, starting with Kd5-e6.
Depending on processor model and speed solving time between 8 and 20 secs.

[fen]r4rk1/pp3pq1/2p5/5P2/1b2N1Q1/8/Pn4PP/R4R1K w - - 0 0[/fen]
White gives mate in 3, starting with Ne4-f6+.
Depending on processor model and speed solving time between 1 min 10 secs and 8 min. 40 sec.

NB: the mate level is probably B7

If it is, you can probably use this manual: http://alain.zanchetta.free.fr/docs/Mis ... met_US.pdf

And it solves up to mate in 5, my previous post mentioned 4...

Good luck with testing regards,
Paul
2024 Special thread: viewtopic.php?f=3&t=12741
2024 Special results and standings: https://schaakcomputers.nl/paul_w/Tourn ... 25_06.html
If I am mistaken, it must be caused by a horizon effect...
SirDave
Full Member
Posts: 610
Joined: Thu Sep 09, 2010 4:59 am
Location: Southern California USA

Post by SirDave »

The Lexibook Chessman Pro Manual can be found here (the English version starts at page 23 of the PDF).

http://cdn.download.ext.lexibook.com/pu ... CG1400.pdf

The Lexibook Chessman Classic (which looks more like the Pavilion unit) manual is here (English version starts at page 11):

http://cdn.download.ext.lexibook.com/pu ... CG1410.pdf
User avatar
appleshampogal
Member
Posts: 126
Joined: Sun Jan 12, 2014 8:53 am

Post by appleshampogal »

paulwise3 wrote:Here are two mate-in-3 tests for your Pavilion, assuming it has the 4 kb Danielsen program.
[fen]6k1/4p2p/7B/3K4/8/8/8/8 w - - 0 0[/fen]
White gives mate in 3, starting with Kd5-e6.
Depending on processor model and speed solving time between 8 and 20 secs.

[fen]r4rk1/pp3pq1/2p5/5P2/1b2N1Q1/8/Pn4PP/R4R1K w - - 0 0[/fen]
White gives mate in 3, starting with Ne4-f6+.
Depending on processor model and speed solving time between 1 min 10 secs and 8 min. 40 sec.

NB: the mate level is probably B7

If it is, you can probably use this manual: http://alain.zanchetta.free.fr/docs/Mis ... met_US.pdf

And it solves up to mate in 5, my previous post mentioned 4...

Good luck with testing regards,
Paul


Hey there!


Getting back to you! So I administered the tests. The Chessman Pro 2 solved mate problem #1 in 16.85 seconds, and mate problem #2 in 2 minutes and 19.61 seconds.

Does this lend any ideas of what program it may be a clone of? I have to say I'm very fascinated of determining which it might be. I'd be willing to give it more tests to narrow it down. I've never owned a Kaare Danielsen machine, so novelty is high here.

Regards,
User avatar
paulwise3
Senior Member
Posts: 1505
Joined: Tue Jan 06, 2015 10:56 am
Location: Eindhoven, Netherlands

Post by paulwise3 »

Hi Apple,

Luuk Hofman's timings for the Enterprise S were:
First problem: 15 secs
Second problem: 2 minutes and 10 seconds.

Give or take a few secs, but this means it must be a clone of the CXG Enterprise S, so you do have a Kaare Danielsen machine! :-)

Best regards,
Paul
2024 Special thread: viewtopic.php?f=3&t=12741
2024 Special results and standings: https://schaakcomputers.nl/paul_w/Tourn ... 25_06.html
If I am mistaken, it must be caused by a horizon effect...
User avatar
appleshampogal
Member
Posts: 126
Joined: Sun Jan 12, 2014 8:53 am

Post by appleshampogal »

paulwise3 wrote:Hi Apple,

Luuk Hofman's timings for the Enterprise S were:
First problem: 15 secs
Second problem: 2 minutes and 10 seconds.

Give or take a few secs, but this means it must be a clone of the CXG Enterprise S, so you do have a Kaare Danielsen machine! :-)

Best regards,
Paul


Ooh, I guess the S stands for Super. Cool! I have found the difference in playing strength between A8 and B1 (30s to 1:00 avg resp time). Also, more interestingly it seems that the computer doesn't need to remain on the B1 level for that long to reap the benefits of more sophisticated play (perhaps due to the compounding effects of the permanent brain?) Below is such a game that illustrates this point. I used a mixed time approach to have the computer take down a 1700 player. I used 1:00 avg resp time for the first 12 moves or so, then switched back to 30s for the remaining. My goal here was to use the computer to achieve the strongest position possible before riding it out using the higher level at the early game.

1. 4k is not large, is the effect of permanent brain more dramatic when afforded a large volume of information courtesy of the 1 min level in the beginning?

OR

2. Is it simply easier for the computer to continue to do well when it is afforded a better position because it did use more strength to achieve that in the start?



Here is what I am talking about:


[Date "2017.05.29"]
[Round "?"]
[White "Comp Enterprise S clone"]
[Black "Club Player"]
[Result "1-0"]
[ECO "B35"]
[TimeControl "1800+10"]

1. e4 {[%clk 0:30:04]} c5 {[%clk 0:29:54]} 2. Nf3 {[%clk 0:29:57]} g6
{[%clk 0:29:44]} 3. Nc3 {[%clk 0:29:32]} Bg7 {[%clk 0:29:32]} 4. d4
{[%clk 0:29:08]} cxd4 {[%clk 0:29:38]} 5. Nxd4 {[%clk 0:28:33]} Nc6
{[%clk 0:29:21]} 6. Be3 {[%clk 0:27:21]} Nf6 {[%clk 0:28:29]} 7. Bc4
{[%clk 0:26:13]} O-O {[%clk 0:27:56]} 8. O-O {[%clk 0:24:55]} Nxe4
{[%clk 0:22:54]} 9. Nxe4 {[%clk 0:23:37]} d5 {[%clk 0:22:41]} 10. Nxc6
{[%clk 0:22:41]} bxc6 {[%clk 0:22:47]} 11. Nd2 {[%clk 0:21:32]} dxc4
{[%clk 0:21:45]} 12. Nxc4 {[%clk 0:21:32]} Ba6 {[%clk 0:16:03]} 13. Qd3
{[%clk 0:20:54]} Bxb2 {[%clk 0:13:25]} 14. Qxd8 {[%clk 0:20:38]} Rfxd8
{[%clk 0:13:31]} 15. Nxb2 {[%clk 0:20:37]} Bxf1 {[%clk 0:13:32]} 16. Kxf1
{[%clk 0:20:37]} Rab8 {[%clk 0:13:37]} 17. Nd3 {[%clk 0:20:33]} a5
{[%clk 0:13:36]} 18. Ke2 {[%clk 0:20:15]} f6 {[%clk 0:12:42]} 19. c4
{[%clk 0:19:49]} e5 {[%clk 0:11:19]} 20. f3 {[%clk 0:19:48]} Kf7
{[%clk 0:11:07]} 21. Rd1 {[%clk 0:19:22]} Kg7 {[%clk 0:08:45]} 22. Rc1
{[%clk 0:18:54]} Rb7 {[%clk 0:07:59]} 23. Rd1 {[%clk 0:18:19]} Re8
{[%clk 0:06:23]} 24. g4 {[%clk 0:17:59]} e4 {[%clk 0:06:15]} 25. fxe4
{[%clk 0:17:53]} Rxe4 {[%clk 0:05:16]} 26. Nc5 {[%clk 0:17:53]} Rb2+
{[%clk 0:05:21]} 27. Kf3 {[%clk 0:17:52]} Re7 {[%clk 0:04:45]} 28. Rd2
{[%clk 0:17:15]} Rb1 {[%clk 0:03:30]} 29. Ne4 {[%clk 0:16:47]} Rf1+
{[%clk 0:02:34]} 30. Rf2 {[%clk 0:16:39]} Rc1 {[%clk 0:02:33]} 1-0

Points to ponder...



:roll:

Kat
User avatar
Tibono2
Full Member
Posts: 695
Joined: Mon Jan 16, 2017 7:55 pm
Location: France
Contact:

Post by Tibono2 »

appleshampogal wrote: Ooh, I guess the S stands for Super.
Hi,
Enterprise S is reported in schach-computer.info wiki to be a 4K K. Danielsen program while Super Enterprise is reported to be the 16K one.

I tested both above mate in 3 positions with my Yeno 320XT (68HC05@2Mhz, 4K K. Danielsen program) with about x1,6 consistent time responses (27 secs / 3min and 46 secs) compared to the Chessman Pro 2. Yeno brand ended taken over by Lexibook (whose early models were clones of Yeno's; and later ones seem to all use the 4K Kaare program within different new housings)

Kind regards, Tibono
User avatar
paulwise3
Senior Member
Posts: 1505
Joined: Tue Jan 06, 2015 10:56 am
Location: Eindhoven, Netherlands

Post by paulwise3 »

appleshampogal wrote: Ooh, I guess the S stands for Super. Cool! I have found the difference in playing strength between A8 and B1 (30s to 1:00 avg resp time). Also, more interestingly it seems that the computer doesn't need to remain on the B1 level for that long to reap the benefits of more sophisticated play (perhaps due to the compounding effects of the permanent brain?) Below is such a game that illustrates this point. I used a mixed time approach to have the computer take down a 1700 player. I used 1:00 avg resp time for the first 12 moves or so, then switched back to 30s for the remaining. My goal here was to use the computer to achieve the strongest position possible before riding it out using the higher level at the early game.

1. 4k is not large, is the effect of permanent brain more dramatic when afforded a large volume of information courtesy of the 1 min level in the beginning?

OR

2. Is it simply easier for the computer to continue to do well when it is afforded a better position because it did use more strength to achieve that in the start?

Points to ponder...

:roll:

Kat
Hi Kat,

Interesting game! :-)

4kb is the program size (ROM=Read Only Memory), the working memory (RAM) is only 128 bytes...
But you are right that taking more time in the beginning leads to better moves and possibly a better position. And once in a good position, that also fits the program's playing style, makes it easier to find the good moves.

In the Leiden tournament we often play our chess computers against each other using external clocks, which allows for changing thinking time and playing style during the game. A strategy often used is letting it play 1 minute/move in the beginning, hoping for a quick positional or even material advantage, and then go back to 30 secs/move or even faster, to consolidate the win within 30 minutes each.

But back to your Chessman: I once played a game with the CXG Sphinx Comet (also Kaare Danielsen) against the CXG Sphinx Concerto, which is about 300 elo stronger rated. The Concerto played a passive opening, and the Comet very active. And it even won the game convincingly! Of course that was a once in a lifetime win, but very frustrating for the Concerto...

And as a last remark: it is said that Kaare Danielsen's 4 kb program is the strongest 4 kb program that exists. And I like it's playing style. Only in the endgame it can play very weak, so if you want to beat it... ;-)

Best regards,
Paul
2024 Special thread: viewtopic.php?f=3&t=12741
2024 Special results and standings: https://schaakcomputers.nl/paul_w/Tourn ... 25_06.html
If I am mistaken, it must be caused by a horizon effect...
Post Reply