Mind blowing moves made by the Star Opal

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
Steve B
Site Admin
Posts: 10146
Joined: Sun Jul 29, 2007 10:02 am
Location: New York City USofA
Contact:

Post by Steve B »

Monsieur Plastique wrote:Hi Steve,

It really could be something simple. For example, a faulty / blown / overheated voltage regulator, burst capacitor, failed diodes, etc. I would not give up on such a precious gem.

I am now worried your last machine will expire during this stressful match. On a serious note, if you don't want to do the match, that is fine.

I would be very upset if the machine expired during this match.
MATCH IS STILL ON DUDE
:P
no point in owning these oldies if i cant use them once in a great while
if it goes down while playing ..then it goes down

Lets Rock Regards
Steve
User avatar
Steve B
Site Admin
Posts: 10146
Joined: Sun Jul 29, 2007 10:02 am
Location: New York City USofA
Contact:

Post by Steve B »

Monsieur Plastique wrote:Star Opal 24 Mhz takes all of 5 seconds to find Marshall's famous 23...Qg3 sacrifice:

[fen]5rk1/pp4pp/4p3/2R3Q1/3n4/2q4r/P1P2PPP/5RK1 b - - 0 1[/fen]


Baffling Regards
While i wont best testing this position with my one remaining Mark V(its been tested and deemed GOOD TO GO for the match and i don't want to press my luck before game time)
i will feed the position to the Epsilon
hopefully i will be showering the board with gold pieces in celebration of Peri finding the move (just as the spectators showered the board when Marshall placed his Queen smack dab into a nest of white pawns)

Steve
User avatar
Steve B
Site Admin
Posts: 10146
Joined: Sun Jul 29, 2007 10:02 am
Location: New York City USofA
Contact:

Post by Steve B »

Steve B wrote:
Monsieur Plastique wrote:Star Opal 24 Mhz takes all of 5 seconds to find Marshall's famous 23...Qg3 sacrifice:

[fen]5rk1/pp4pp/4p3/2R3Q1/3n4/2q4r/P1P2PPP/5RK1 b - - 0 1[/fen]


Baffling Regards
While i wont best testing this position with my one remaining Mark V(its been tested and deemed GOOD TO GO for the match and i don't want to press my luck before game time)
i will feed the position to the Epsilon
hopefully i will be showering the board with gold pieces in celebration of Peri finding the move (just as the spectators showered the board when Marshall placed his Queen smack dab into a nest of white pawns)

Steve
Interestingly the Epsilon takes more time then the Star Opal
it finds the move in 7 seconds
the move choices were..
1-2 seconds Ne7+
3-6 seconds Qb4
7 seconds Qg3

Given that the Epsilon took 40% more time then the lowly overclocked Star Opal
i showered the board with air freshener spray instead of Gold pieces .. in protest
Disenchanted Regards
Steve
User avatar
Harvey Williamson
Site Admin
Posts: 6079
Joined: Sun Jul 29, 2007 6:57 am
Location: Media City, UK
Contact:

Post by Harvey Williamson »

There is more hype here than in Ali v Frazer - when the hell does the action start????? :D
User avatar
Steve B
Site Admin
Posts: 10146
Joined: Sun Jul 29, 2007 10:02 am
Location: New York City USofA
Contact:

Post by Steve B »

Harvey Williamson wrote:There is more hype here than in Ali v Frazer - when the hell does the action start????? :D
Tomorrow ..11:30 Pm Server time(or is that 23:30 server time??)
Unless Jon Chickens out of course Regards
Steve
User avatar
Steve B
Site Admin
Posts: 10146
Joined: Sun Jul 29, 2007 10:02 am
Location: New York City USofA
Contact:

Post by Steve B »

Monsieur Plastique wrote:
[fen]r3r1k1/pp3pbp/1qp3p1/2B5/2BP2b1/Q1n2N2/P4PPP/3R1K1R b - - 0 1 [/fen]

Not to put a damper on it though, the stock Star Opal finds this move in 15 seconds as well,
i decided to load the Q sac position up on the Epsilon to see if it could beat your Stock Star Opal's 15 seconds

i am stunned to report it took the Epsilon 16 Minutes!(not seconds) and seven ply to see Be6
preferring rather to sac the Q with Bxf3..a far inferior move
it intended ..
..Bxf3
Bxb6 axb6
Qxc3
with an Eval of +.69 in its favor

just to be certain that there was nothing wrong with the Epsilon i pulled a Stock Par Excellence off the shelf and loaded up the position
while the Par has no display screen it can show the depth search and intended move
as played by the Epsilon it intended Bxf3 and then switched to Be6 after 7 ply and about 20 minutes
the Epsilon actually has the same hardware as the Fidelity Designer 2100 which is 6 Mhz while the Par runs at 5 Mhz ..both have the same Par program though..so that explains the slower time for the Par

Still..as your Stock Star Opal found the move in 15 seconds ..i imagine your Star Opal on Steroids would find it in 5 seconds

5 seconds vs 16 minutes
:shock:

i think i am being lead like a lamb to the slaughter here


Baaaaaaah Regards
Steve
User avatar
Monsieur Plastique
Senior Member
Posts: 1014
Joined: Thu Jul 03, 2008 9:53 am
Location: On top of a hill in eastern Australia

Post by Monsieur Plastique »

Yes, well the steroidal StarOpal does find the Queen sac in 5 seconds, but as you can see from the other posts, it helpfully throws away it's knight in a From's Gambit after 3 minutes thought, can lose it's Queen in a 2 ply blunder and lose a piece because it's Queen gets trapped - all when given decent thinking times. So at the same time it can make a blunder only a rank beginner can make, it has also defeated Mephisto Miami that is rated around 1900!

My extensive studies have shown that some of these problems are due to the silly way it can sometimes handle the variable time levels, which is why I just use this computer on the fixed time levels instead. For example, it is currently playing a handicap match against Novag Constellation where Star Opal Turbo is on level WC6 and Constellation has 30 seconds. Thus far, the score is 2 to 1 in favour of Star Opal, with no silly blunders thus far played. But the previous match on the equivalent variable time level saw Star Opal lose 2-4. So Opal only needs one draw in the next 3 games to better the previous effort.

Secondly, I reckon there might be some sort of issue with the way it handles the RAM - perhaps this explains why it often makes even worse moves the longer it thinks for.

Thirdly, I think there is something wrong with the randomisation algorythms. This program, unlike other Novags, has the randomisation permanently enabled with no way to turn it off. I have seen it play up to 4 different reponses in a given position at the exact same variable time level. It is not hard to imagine that in any given position there are going to always be 4 moves - all just as good as each other. On the other hand, with the fixed time levels, I have only seen it come up with two replies in a given position, so on these levels it seems to be "best" and "next best". This is possibly why there seems to be less blundering on the fixed levels - at least in my testing thus far.

And finally, this thread has revealed the most obvious characteristic of all. This program, written as it was by an American, finds combinations made by Americans extremely quickly.

Enjoy the match against the Star Opal Marshall Edition :)


Star Spangled Banner Regards
Chess is like painting the Mona Lisa whilst walking through a minefield.
User avatar
Steve B
Site Admin
Posts: 10146
Joined: Sun Jul 29, 2007 10:02 am
Location: New York City USofA
Contact:

Post by Steve B »

Monsieur Plastique wrote:
Enjoy the match against the Star Opal Marshall Edition :)

Star Spangled Banner Regards
exactly 24 Hrs to game time(as i write) Dude

going into News/Media Blackout Regards
Steve
User avatar
Monsieur Plastique
Senior Member
Posts: 1014
Joined: Thu Jul 03, 2008 9:53 am
Location: On top of a hill in eastern Australia

Post by Monsieur Plastique »

Yes, hide under the covers. It is the safest place. In the meantime, I am off to recruit Bruce Buffer to do the player intros tomorrow.
Chess is like painting the Mona Lisa whilst walking through a minefield.
User avatar
Harvey Williamson
Site Admin
Posts: 6079
Joined: Sun Jul 29, 2007 6:57 am
Location: Media City, UK
Contact:

Post by Harvey Williamson »

Monsieur Plastique wrote:Yes, hide under the covers. It is the safest place. In the meantime, I am off to recruit Bruce Buffer to do the player intros tomorrow.
Hmm only if he buys a serial number for playchess ;-)
User avatar
Steve B
Site Admin
Posts: 10146
Joined: Sun Jul 29, 2007 10:02 am
Location: New York City USofA
Contact:

Post by Steve B »

Match to begin soon Jon

i will be in the Engine room at exactly 11:30 pm server time(23.30)..about 1.5 Hrs from now


when you get to the engine room i will Challenge you to an UNRATED NO CLOCK game
you will have White for the first game
my nick is.. OutOfWork
the Mark VI will be set to 30 sec AVG(I can also use Fixed time 30 sec if you prefer that)
Match..Best of Three Games ..which means i will only need 2 games to blow you away
:P
2-0 Regards
Steve
User avatar
Harvey Williamson
Site Admin
Posts: 6079
Joined: Sun Jul 29, 2007 6:57 am
Location: Media City, UK
Contact:

Post by Harvey Williamson »

I look forward to seeing the game but sadly wont be online to watch it as I am doing a nightshift at work. Will check here though for the commentry.
User avatar
Monsieur Plastique
Senior Member
Posts: 1014
Joined: Thu Jul 03, 2008 9:53 am
Location: On top of a hill in eastern Australia

Post by Monsieur Plastique »

Ready to go Steve. 24 minutes and counting. I will have Star Opal at fixed 30 seconds per move, only because from my experience it is less likely to make those terrible game-ruining blunders than on the corresponding variable level.

As I mentioned previously though, I will force Star Opal to move immediately if there is only one legal move, since otherwise it will still take the full 30 seconds, despite what the user manual says.

I'd suggest that the variable level would probably be strongest on the Mark VI, especially given that Star Opal cannot think on it's opponent's time, thereby being unable to take any advantage of moves where Mark VI might take a lot longer than average.

On the other hand, it has been so many years since I used a Mark V / VI, I can't recall whether it moves much faster than average on the faster levels. But I do remember it moves pretty fast on it's tournament level - being left with a lot of spare time at the end.

So perhaps you should quickly check to see if the average 30 seconds really will use an average of 30 seconds, otherwise you could be at a disadvantage in terms of total time used.
Chess is like painting the Mona Lisa whilst walking through a minefield.
User avatar
Steve B
Site Admin
Posts: 10146
Joined: Sun Jul 29, 2007 10:02 am
Location: New York City USofA
Contact:

Post by Steve B »

I Am There now

ill go with Avg 30
if it takes more then a minute on any one move ill force it to move

Later
Steve
User avatar
Steve B
Site Admin
Posts: 10146
Joined: Sun Jul 29, 2007 10:02 am
Location: New York City USofA
Contact:

Post by Steve B »

It is now 23:30 server time and you are not in the Engine room
i claim the match on time forfeit

Have A Nice Day Regards
Steve
Post Reply