new chess computer: CT800

[fern]

well, if all that is like you say it is, then which is the sense of your device?
I mean, OR you want something to run any engine and for that you already have Revelation or simply your current PC, OR you want a full piece of old fashioned hardware running an old engine.
That's the reason collector buy old stuff.
A simple tasteless box running one engine which is not old enough or strong enough, for what to have it?
The sense of your endeavor should be, IMHO, to create a housing more or less as we expect them to be according to old stuff and running an old, famous engine. It is more expensive, but it has sense. It is what Millenium did with success.
You are bot going to interest a market that want to buy an old Ford T OR something that looks as one offering them, instead, a bike...
Sorry if I looks like an unpleasant guy, but truth is truth.
Fern

[Ras]

well, if all that is like you say it is, then which is the sense of your device?

Several things:

- the Revelation cannot run on standard AA batteries because it draws too much power.

- it does not allow the user to have full access to the sources, modify them and reprogram the device at will.

- there are actually fans of dedicated units who are fed up by the umpteenth Lang or Schroeder program. In fact, this is one of the main problems that Millenium will be facing. People will NOT continue to buy the next Lang program in a slightly updated machine because they already have enough Langs. That market is pretty much exhausted.

- the project "AVR-Max" (running H.G. Mueller's Micromax) was quite successful. The old 8bit-CPU just is lacking any further potential.

- I have already been asked whether I'd want to sell the prototype because the "2016 briquet" is unique in being both something new, not just something rehashed, and still offering a retro briquet style.

- the Millenniums have an awful, cheap plastic board. Better have no board at all and using a nice external chess set, which every chess player has. The targeted price can also be lower, 30-40 EUR are possible.

- ELO 2100 is quite enough for most hobby chess players. The rating could be higher if it were using only the best opening book moves, but at this level of human play, you have to deal with non-optimal openings, too.

- the unusual point with NG-Play is that is has no concept of space, only of mobility. The results are games like I have not seen with other chess programs. It is like the difference between regarding chess as the equivalent of a field battle or as urban warfare. It helped me a lot to better understand hypermodern chess strategy.

- last but not least, I gained valuable experience with the ARM Cortex-M platform, which in itself is a practical way of keeping me updated to current hardware in the market.

[fern]

All that you already said to us and It is good to see you have faith in your endeavor. I only desire you success and any and every one of my comments have been delivered in order to help, if possible, for a more amusing device.

If ever you have this thing massively made, you can count I will purchase one.

[Ras]

thank you Fernando; what I agree with is that the design of the prototype isn't actually appealing. It looks more like a medical lab device where you insert some probe at the backside.

I had to use what I could get in terms of available electronics housings. Plus that I am not good at visual design since I'm really no artist. So of course, that will have to look a lot more attractive in a series product. My dream would be throwing in some timeless Braun inspired design. Going by your avatar, I think you remember these things.

[fern]

Not only I do not remember, but even I do not know what you refer to. ...
Designs not necessarily must be pretty, but suggest an old chess computer. Or an old brick style chess contraption as some made by Mephisto.
Fern

[Guenther]

So here is a new dedicated chess computer, the CT800:

http://www.ct800.net

...

George Georgopoulos, author of NG-Play, didn't put a dedicated licence to the software. He just stated that people are free to do anything as long as it stays open source. I felt that this is in the spirit of the GPL, so I put the project under GPLv3+. Another contribution is the KPK endgame table module by Marcel van Kervinck. Thanks to you both.

Rasmus, searching for the license of the NG-Play source I found this post here.
Today there was a 'new' program mentioned in the 'Tournaments...' forum, but it is practically NG-Play just with an uci interface added and two changed piece values.
(plus a book in bin format which strangely isn't binary in reality, but obfuscated ascii by different encoding...)

It mentions the name of NG-Plays' author in the readme:

Code: Select all

This chess engine is based on the work of George Georgopoulos and Tom Kerrigan (thanks to them).

But it is closed source and somehow I don't like this and therefor searched for a kind of license for NG-Play and landed here.

What do you think in this case? 'Based on the work' can mean a lot,
but I didn't expect that it was meant literally ;-)

Guenther

[Volker Pittlik]

...

What do you think in this case? 'Based on the work' can mean a lot,
but I didn't expect that it was meant literally

Guenther

When I saw this:

http://www.ct800.net

my first thought was: take a RaspberryPi, put Stockfish in it, do some research on realated hardware (keys, display, casing and so on, and build the best "briquet" ever made.

Nice winter project.

Volker

P.S. I really enjoyed playing with that things: https://schach-computer.info/wiki/index.php/Mephisto_I

[Ras]

Today there was a 'new' program mentioned in the 'Tournaments...' forum, but it is practically NG-Play just with an uci interface added and two changed piece values.

Yeah, I replied in the other thread. "Based" isn't the right word here, and it's in violation of NG-Plays usage terms because "Zorgus" is closed source.

[Ras]

my first thought was: take a RaspberryPi, put Stockfish in it

Mostly exists already as DGT Pi. But just doing a re-compile is too easy. The Cortex-M4 has 1 MB of flash-ROM, which is huge, but only 192 kB of RAM, and try to get a decent PC engine running with that.

The biggest reason for me was that I wanted to gain practical experience with a Cortex-M microcontroller and thought what I might do with it. That's also why it's running bare-metal and why I've coded the low-level stuff as per the MCU's reference manual.

It's also nice that standard AA batteries allow for a decent playing time. A Raspi needs much more energy.

[Guenther]

Today there was a 'new' program mentioned in the 'Tournaments...' forum, but it is practically NG-Play just with an uci interface added and two changed piece values.

Yeah, I replied in the other thread. "Based" isn't the right word here, and it's in violation of NG-Plays usage terms because "Zorgus" is closed source.

Thanks, I saw your response now in the 'Zorgus' thread.