logo Welcome, Guest. Please Login or Register.
November 17, 2019, 08:23PM
Home Help Search Login Register
News: Mirror of the old site for Tasc.nl at www.chesscomputers.org

  Chess Computer Community
  General Category
  Discussions about Chess Computers
(Moderators: Endspielgott, Overtom, bobosse)
  Dedicated Machines - progress?
previous next
Pages: [1] Reply Notify of replies Send the topic Print 
   Author  Topic: Dedicated Machines - progress?  (Read 903 times)
Manjushri
Newbie
*

Offline

Posts: 12

I love Boardnation!

View Profile
Re:Dedicated Machines - progress?
« on: May 26, 2006, 09:14AM »
Reply with quote

I'm sure you're right about the space angle. I guess I am concentrating on the full size boards too much. The President for instance has acres of space under the playing surface - about 4 times more than my Vaio laptop.  The manufactureres will of course wish to use the same modules across their whole range including the tiny portables, to maximise their R&D layout.
Report to moderator Logged
Robert Weck
Hero Member
*****

Offline

Posts: 791



I love Boardnation!

View Profile
Re:Dedicated Machines - progress?
« Reply #1 on: May 26, 2006, 04:53AM »
Reply with quote


Quote from: Manjushri on May 26, 2006, 03:19AM   

Hi Robert,
I know it's not a trivial thing and one has to look at power supplies, heat disipation, bus speeds etc etc, and I am certainly not belittling Ruud's fantastic work.  But we know you're not going to get much more than 2400 elo out of a 200mhz chip don't we.
Why such a slow chip?  I had a laptop in 1996 which ran at 233mhz - which is the point I was making about 10 years behind.

I would think, the problem is, that you don't have enough space on a module, that fits in a Exclusive board, for all the peripherie that is needed with a modern PC CPU.

The StrongArm doesn't need all this and can be used on a very small board (as in the PPC's and Resurrection)

This should be the reason, why even a modern chip (like the StrongArm or e. g. the faster PXA270) are still far behind the corresponding Desktop/Notebook CPU's...

With a fast CPU, you would also need much RAM for Hash Tables, and this also needs space...


best regards,
Robert
Report to moderator Logged
Ismenio
Horsie to pointy guy 6
Administrator
*****

Offline

Posts: 1753





View Profile WWW E-Mail
Re:Dedicated Machines - progress?
« Reply #2 on: May 26, 2006, 04:53AM »
Reply with quote

Nice discussion folks!

What I would like to see is someone piecing together things like this: the Space Cube, a small, yet very powerful PC:

http://www.epicempire.com/computers/space-cube.html



(also http://linuxdevices.com/news/NS3619879482.html)

add a powerful open source chess engine and if one could adapt an interface with a regular chess board to use reed switches, you could have a very powerful "dedicated" chess computer with lots of features! Maybe even add an optional tiny LCD and a way to save off games to a standard USB thumb drive, etc.

I know cost is always a problem but I think it would be cool to have something like this.

Really, the PCs have made a big impact on chess.

And Robert, I agree with you 100% about those very annoying sound effects Excalibur added to their computers!



Regards,

Ismenio

PS: BTW, Manjushri, welcome!
Report to moderator Logged

Map of chess computer friends: http://www.frappr.com/chesscomputers
Manjushri
Newbie
*

Offline

Posts: 12

I love Boardnation!

View Profile
Re:Dedicated Machines - progress?
« Reply #3 on: May 26, 2006, 03:19AM »
Reply with quote

Hi Robert,
I know it's not a trivial thing and one has to look at power supplies, heat disipation, bus speeds etc etc, and I am certainly not belittling Ruud's fantastic work.  But we know you're not going to get much more than 2400 elo out of a 200mhz chip don't we.
Why such a slow chip?  I had a laptop in 1996 which ran at 233mhz - which is the point I was making about 10 years behind.
I am not a hardware buff but was in software for decades and I do know that porting a tiny piece of software like a chess engine to another platform can't be that difficult.
The thing about a quality chess computer is that it is an investment for life.  That is what justifies the high price hobbyists may be prepared to pay.  My President was 300 way back and if I were in the market now I would be happy to pay 600, but I'd expect something much closer to state of the art technology.
With regard to using the PC programs and a board attached - no thanks.  Part of the joy of an autosensory board made of wood is the whole aesthetic experience.  This is destroyed for me once you start attaching screens and cables.
Many thanks for your thoughts.
Report to moderator Logged
Robert Weck
Hero Member
*****

Offline

Posts: 791



I love Boardnation!

View Profile
Re:Dedicated Machines - progress?
« Reply #4 on: May 26, 2006, 12:11AM »
Reply with quote


Quote from: Manjushri on May 25, 2006, 03:04PM   

Thanks for that Terry.
I agree but surely the basic hardware is already in production... boards, sensors, keypads etc have all been made pretty perfectly by now.  Just needs an incremental addition of a few gigs worth of hash tables and a dual intel chip here and there.  Can't be that expensive. 

I'm afraid, it is not as easy as that!

You can't use Intel X386 CPU's, because the programs are not written for it.

The mainboard would have to be completely new designed.

Just look at Ruud Martin's Resurrection; it uses a simple 200 MHz StrongArm CPU, which you can find in many older PocketPC's. It took him about 2(?) years of development and i'm afraid not many computer chess enthusiasts would pay the price, that it at leasts costs.

Even the primitive DGT board (no display, no LED's) costs here in Germany >400 Euros! (ok, ok, maybe not really primitive as it has piece recognition! )

But i think, if somebody wants the most modern and strong programs (Fritz, Shredder (i don't know, if UCI engines are working?)) without having to look at a PC screen, the DGT board seems to be the best solution!

Quote:

Instead we see more and more little silver portable jobs with flashy designs but no better playing strength.

But for about 98% of all chessplayers this is enough and the rest isn't enough to make a new development profitable (at least, this is, what the companies are thinking)

Look at Excalibur: most of their computers have terrible sound FX (galloping horses, clinking swords and so on; awful!>:() which have nothing to do with chess and are playing very weak, but it seems, that this is, what at least many people (in the US?) want! 

Also here in Germany you find many, many cheap "Millenium" computers, which play also very weak, but this seems to be the way, the companies earn the most profit!

Sad, but true...


Robert
« Last Edit: May 26, 2006, 12:18AM by Robert Weck » Report to moderator Logged
Manjushri
Newbie
*

Offline

Posts: 12

I love Boardnation!

View Profile
Re:Dedicated Machines - progress?
« Reply #5 on: May 25, 2006, 03:04PM »
Reply with quote

Thanks for that Terry.
I agree but surely the basic hardware is already in production... boards, sensors, keypads etc have all been made pretty perfectly by now.  Just needs an incremental addition of a few gigs worth of hash tables and a dual intel chip here and there.  Can't be that expensive.  Instead we see more and more little silver portable jobs with flashy designs but no better playing strength.
Hey, you mean you have trouble beating your computer?? 
Report to moderator Logged
TerryG
Jr. Member
**

Offline

Posts: 63

I love Boardnation!

View Profile E-Mail
Re:Dedicated Machines - progress?
« Reply #6 on: May 25, 2006, 01:57PM »
Reply with quote

I think the reason is dedicated (ded?) machines have had there day in the glorious 70's-80's and 90's and a lot of money was made then as the PC programs available were pretty poor until the early 90's when they started to overtake on strengh when the Pentium processor came out. It costs alot to bring out a machine to play at a strength of say 2400 on a dedicated machine that perhaps plays only chess. When you can for 20/$40 produce a CD rom that plays 2600 or more. Mind you I find it MUCH more preferable to own these 'dinosaurs' of the past. I have trouble looking at a screen for too long, preferring the traditional board instead.
Anyhow how many of us can play to a strength of 2000, let alone 2400 or more!!!!
Report to moderator Logged

TerryG
Manjushri
Newbie
*

Offline

Posts: 12

I love Boardnation!

View Profile
Dedicated Machines - progress?
« Reply #7 on: May 25, 2006, 06:11AM »
Reply with quote

I have had a Kasparov President for many years now and it is still an exquisite pleasue to play on this beautiful machine. However, even when I bought it the limitations were obvious.  It simply isn't powerful enough to do serious analysis.
I can't really understand why such little progress has been made with dedicated machines since then.  The very best (Mephisto? still struggling to reach a 2400 elo.  And meanwhile I have Hiarcs on my Palm playing at over 2600 elo!!
Chips/memory are now really cheap.  Why are these machines still 5-10 years behind? Is it just commercial market constraints that are holding them back?
I really wish I had found the extra 100 or so to get on the Mephisto Senator hardware though. At least I could be gradually upgrading now.
I presume nobody has done any work modifying the President?  Such a lovely machine.
Report to moderator Logged
Pages: [1] Reply Notify of replies Send the topic Print 
previous next
Jump to:  

Login with username, password and session length

Chess Computer Community
Get your Ad Free Message Board at Boardnation!   © 2013 Boardnation  Web hosting by Nubis