I'm trying to understand why we don't see any computer chess with the
ability to move the pieces unassisted.

After some research I've found only 4 models and it seems none of them
are being produced anymo

1.- Phantom Chess (M & B Electronics)
2.- Mirage (Excalibur)
3.- Novag Robot Adversary (had a robotic arm)
4.- Excalibur Robotic Chess 740 (another one with a robotic arm)

I've read some reviews on the Mirage model that seems to indicate the
product was poorly manufactured. For example, it used simple DC motors
that are found in cheap toys instead of a stepper motor. It that's true,
I can see why it was discontinued.

There are a couple of pages and even videos of the Phantom Chess on the
web and that seems to be a well built model from the 80s.
Personally, I think the concept is great and I'd love to have an updated
versions of those machines. So, I open this up for discussion: why don't
we see any model that can move its own pieces today?

Is it a lack of interest? Is it too expensive to manufacture? (With
today's technology I'd image that would not be the case)
Does anyone know of any 'open source' project to produce such a model or
even reverse engineer one that is still available so that we could build
one that could connect to a computer and even handle PGN files? I have
found a couple of on line stores that seem to still have the Mirage for
sale.

Let me know what your thoughts are.

Thanks,

Ismenio Sousa

Ismenio Sousa wrote:
I'm trying to understand why we don't see any computer chess with the
ability to move the pieces unassisted.

After some research I've found only 4 models and it seems none of them
are being produced anymo

1.- Phantom Chess (M & B Electronics)
2.- Mirage (Excalibur)
3.- Novag Robot Adversary (had a robotic arm)
4.- Excalibur Robotic Chess 740 (another one with a robotic arm)

I've read some reviews on the Mirage model that seems to indicate the
product was poorly manufactured. For example, it used simple DC motors
that are found in cheap toys instead of a stepper motor. It that's true,
I can see why it was discontinued.

There are a couple of pages and even videos of the Phantom Chess on the
web and that seems to be a well built model from the 80s.
Personally, I think the concept is great and I'd love to have an updated
versions of those machines. So, I open this up for discussion: why don't
we see any model that can move its own pieces today?

Lack of demand, high price...

High quality means high price. With low demand, economy of scale in
manufacturing doesn't apply, so there is no way to bring the price down.

Initial tool-up simply isn't worth it apparently...

Is it a lack of interest? Is it too expensive to manufacture? (With
today's technology I'd image that would not be the case)
Does anyone know of any 'open source' project to produce such a model or
even reverse engineer one that is still available so that we could build
one that could connect to a computer and even handle PGN files? I have
found a couple of on line stores that seem to still have the Mirage for
sale.

Let me know what your thoughts are.

Thanks,

Ismenio Sousa


--
Robert M. Hyatt, Ph.D. Computer and Information Sciences
University of Alabama at Birmingham
(205) 934-2213 136A Campbell Hall
(205) 934-5473 FAX Birmingham, AL 35294-1170

Robert Hyatt wrote:
Ismenio Sousa wrote:
Personally, I think the concept is great and I'd love to have an updated
versions of those machines. So, I open this up for discussion: why don't
we see any model that can move its own pieces today?

Lack of demand, high price...

High quality means high price. With low demand, economy of scale in
manufacturing doesn't apply, so there is no way to bring the price down.

Initial tool-up simply isn't worth it apparently...

I imagine that reliability might also be an issue.


Dave.

--
David Richerby Evil Widget (TM): it's like a thingy
www.chiark.greenend.org.uk/~davidr/ but it's genuinely evil!

Interesting question.

I think the reason no such products exist today are these - (a) they
didn't really work very well, (b) they were expensive, and (c) they are
basically just a novelty that is a niche market within a niche market.

My guess is, every product cited below lost significant money for their
manufacturer/investor. Regardless of what 'technology' you use, a
mechanical device to move chess pieces on a chessboard reliably is going
to be much more expensive than a good chess program with good 3D
graphics - your primary competitor.

You can always re-load software or download a patch. Who's gonna fix a
malfunctioning or broken mechanical chessboard and pay the shipping
costs? I think potential investors who research the idea will run fast
the other way.

What you describe is do-able, technically. Just the potential market is
very small to start with. Kinda like opening a fine restaurant in a
teeny, remote town in the middle-of-nowhere. Doesn't matter how nice it
is, they always lose their shirts.



I'm trying to understand why we don't see any computer chess with the
ability to move the pieces unassisted.

After some research I've found only 4 models and it seems none of them
are being produced anymo

1.- Phantom Chess (M & B Electronics)
2.- Mirage (Excalibur)
3.- Novag Robot Adversary (had a robotic arm)
4.- Excalibur Robotic Chess 740 (another one with a robotic arm)

I've read some reviews on the Mirage model that seems to indicate the
product was poorly manufactured. For example, it used simple DC motors
that are found in cheap toys instead of a stepper motor. It that's true,
I can see why it was discontinued.

There are a couple of pages and even videos of the Phantom Chess on the
web and that seems to be a well built model from the 80s.
Personally, I think the concept is great and I'd love to have an updated
versions of those machines. So, I open this up for discussion: why don't
we see any model that can move its own pieces today?

Is it a lack of interest? Is it too expensive to manufacture? (With
today's technology I'd image that would not be the case)
Does anyone know of any 'open source' project to produce such a model or
even reverse engineer one that is still available so that we could build
one that could connect to a computer and even handle PGN files? I have
found a couple of on line stores that seem to still have the Mirage for
sale.

David Richerby wrote:
Robert Hyatt wrote:
Ismenio Sousa wrote:
Personally, I think the concept is great and I'd love to have an updated
versions of those machines. So, I open this up for discussion: why don't
we see any model that can move its own pieces today?

Lack of demand, high price...

High quality means high price. With low demand, economy of scale in
manufacturing doesn't apply, so there is no way to bring the price down.

Initial tool-up simply isn't worth it apparently...

I imagine that reliability might also be an issue.


That could be solved. But it factors into the $$$ part of the equation
of course... Which is the ultimate reason why they are not commonplace.

Dave.

--
David Richerby Evil Widget (TM): it's like a thingy
www.chiark.greenend.org.uk/~davidr/ but it's genuinely evil!

--
Robert M. Hyatt, Ph.D. Computer and Information Sciences
University of Alabama at Birmingham
(205) 934-2213 136A Campbell Hall
(205) 934-5473 FAX Birmingham, AL 35294-1170

Ismenio Sousa wrote:

There are a couple of pages and even videos of the Phantom Chess on the
web and that seems to be a well built model from the 80s.
Personally, I think the concept is great and I'd love to have an updated
versions of those machines. So, I open this up for discussion: why don't
we see any model that can move its own pieces today?

Maybe because in a 5 or 10 years we will have "holographics" boards,
with pieces that resembles the "real thing", and are moved by a
computer... without any clumsy arms!

The other way, you can use a web cam and a "lego" robot-arm, and make
your electronic playing partner. Put in a good optical recognition of
the pieces and the board, and you don't need to tell it what piece you
have moved (it waits you end the move and then compares the board before
and after the move, and draw its conclusions). And it will not need a
step motor -- it can compensate the moves by looking where the arm is
and where it is supposed to be (a pair of cameras to give it 3D vision
will be a plus). If you use some radio or infrared technology
(bluetooth?), the computer can be anywhere in the house, trying to
discover what is the best move to crush you, you poor carbon-and-water
creature.

Today it is science fiction. Or not? The critical parts I think that are
the optical recognition of the board and the coordination hand-eye. All
of this is software based. Can you imagine this? Your Aibo(R) playing
chess? Ops.

[]s
--
..O. Cesar A. K. Grossmann ICQ UIN: 35659423
...O http://www.LinuxByGrossmann.cjb.net/
OOO Quidquid Latine dictum sit, altum viditur

Cesar A. K. Grossmann wrote:
Ismenio Sousa wrote:

There are a couple of pages and even videos of the Phantom Chess on the
web and that seems to be a well built model from the 80s.
Personally, I think the concept is great and I'd love to have an updated
versions of those machines. So, I open this up for discussion: why don't
we see any model that can move its own pieces today?

Maybe because in a 5 or 10 years we will have "holographics" boards,
with pieces that resembles the "real thing", and are moved by a
computer... without any clumsy arms!

The other way, you can use a web cam and a "lego" robot-arm, and make
your electronic playing partner. Put in a good optical recognition of
the pieces and the board, and you don't need to tell it what piece you
have moved (it waits you end the move and then compares the board before
and after the move, and draw its conclusions). And it will not need a
step motor -- it can compensate the moves by looking where the arm is
and where it is supposed to be (a pair of cameras to give it 3D vision
will be a plus). If you use some radio or infrared technology
(bluetooth?), the computer can be anywhere in the house, trying to
discover what is the best move to crush you, you poor carbon-and-water
creature.

Today it is science fiction. Or not? The critical parts I think that are
the optical recognition of the board and the coordination hand-eye. All
of this is software based. Can you imagine this? Your Aibo(R) playing
chess? Ops.

The optical recognition is not needed. If you fix the arm to some known
position, it can find the piece on any square without needing the cameras
and extra complexity.

Note that Novag made such a device years ago. And before that, in 1977,
chess 4.7 used a robot arm and magnetic reed switches in the chess board
to play some games vs David Levy. Same idea...


[]s
--
.O. Cesar A. K. Grossmann ICQ UIN: 35659423
..O http://www.LinuxByGrossmann.cjb.net/
OOO Quidquid Latine dictum sit, altum viditur

--
Robert M. Hyatt, Ph.D. Computer and Information Sciences
University of Alabama at Birmingham
(205) 934-2213 136A Campbell Hall
(205) 934-5473 FAX Birmingham, AL 35294-1170

Robert Hyatt wrote:

The optical recognition is not needed. If you fix the arm to some known
position, it can find the piece on any square without needing the cameras
and extra complexity.

But, if you have the optical recognition (hey, it will need to recognize
a dozen of different pieces, the board, and position of the board, what
can be more simpler than that? -- don't say), you can use it with a
regular board and pieces. You will not need a special one... If it is
mounted with the arm and the player engine, it can be portable. It can
replace the computer operator on man vs. machine matches (with a certain
appeal). Did someone wants to send me a computer, a Lego (R) robotic
arm, and two web cams? Just kidding...

Maybe you can make the whole thing work without the optical recognition
thing -- you only put the robotic arm near the board and run some kind
of location procedure (like you help the arm locate the rooks and the
kings). But wihtout the optical recogintion you will need a way to tell
the robot what move *you* played (ok, voice recognition system).

Note that Novag made such a device years ago. And before that, in 1977,
chess 4.7 used a robot arm and magnetic reed switches in the chess board
to play some games vs David Levy. Same idea...

Almost "The Turk" on electronic-steroids...

[]s
--
..O. Cesar A. K. Grossmann ICQ UIN: 35659423
...O http://www.LinuxByGrossmann.cjb.net/
OOO Quidquid Latine dictum sit, altum viditur