Chess

[Bomberman94]

6 minutes ago, Andrew Davie said:

Also... PAL60 is much better than PAL50.  Not sure which you're viewing.

 

... always PAL60 if i could - I used the last file of the game you uploaded here. If I should test another/special version I’m ready to test it for you. I‘m playing on a PAL CRT which also can handle other formats. 

Edited March 5, 2021 by Bomberman94

[+Andrew Davie]

I've managed to resurrect a non-working Harmony Encore cart that was sent to me.

This allowed me to view the result on my TV. I'm a tad disappointed, actually - I don't think this is a viable display for a static chess game, as the shimmer is way too noticeable to me. Athough I know it's at 20Hz, it looks to my eye much more like 4 or 5 Hz. Strange.

 

[Thomas Jentzsch]

47 minutes ago, Andrew Davie said:

This allowed me to view the result on my TV. I'm a tad disappointed,

You always have to be very careful to judge flicker by using Stella only. Stella can only be a coarse approximation, because other factors e.g. the the computer display used plays a major role too.

[Bomberman94]

1 hour ago, Andrew Davie said:

I've managed to resurrect a non-working Harmony Encore cart that was sent to me.

This allowed me to view the result on my TV. I'm a tad disappointed, actually - I don't think this is a viable display for a static chess game, as the shimmer is way too noticeable to me. Athough I know it's at 20Hz, it looks to my eye much more like 4 or 5 Hz. Strange.

 

Always ready to test with my Harmony cart and PAL Atari 2600 (light 6switch) on my TV if you have a new version of your chess (really like to play it on my Atari - therefore I would love to test it). ?

[+Andrew Davie]

35 minutes ago, Thomas Jentzsch said:

You always have to be very careful to judge flicker by using Stella only. Stella can only be a coarse approximation, because other factors e.g. the the computer display used plays a major role too.

Adjusting the brightness and contrast DOWN on the TV significantly improves the resultant image.

 

[+Andrew Davie]

I've had a good play with the colours/intensities, and reduced apparent flicker considerably.

This one looks OK to me on my TV, even with the brightness/contrast at "normal" setting (50%)

Certainly way better than my initial version. I'm feeling a bit better about it now.

 

CDFJChess20210331.bin

[Bomberman94]

On 3/31/2021 at 12:14 PM, Andrew Davie said:

I've had a good play with the colours/intensities, and reduced apparent flicker considerably.

This one looks OK to me on my TV, even with the brightness/contrast at "normal" setting (50%)

Certainly way better than my initial version. I'm feeling a bit better about it now.

 

CDFJChess20210331.bin 32 kB · 12 downloads

Less flickering - yes I can confirm. Maybe you might do a PAL60 version as colors are „a bit off“ on my TV? ?

[Bomberman94]

Do you think about a physical release when your chess is ready?

[+Andrew Davie]

4 hours ago, Bomberman94 said:

Do you think about a physical release when your chess is ready?

Still experimental at this stage. Getting to a release version would be a lot of work that I'm not really ready for at this stage.

First I'd like to see what happens in the area of a bankswitch scheme that would let me (easily) use the ARM for the chess engine. If that happens, then I think an engine might be worthwhile releasing on cartridge. In particular, I'd like to use the PlusCart design, with modified bankswitching, because then I could think about using online engines as move-generators as well.

 

 

[AW127]

On 4/23/2021 at 4:12 AM, Andrew Davie said:

Still experimental at this stage.

 

Hy Andrew. After a longer time, i wanted to ask, if this error in BitChess could be fixed in the meantime, which occurred at that time, if one of the pieces was exchanged? Do you remember, I had reported about it once here in this thread, when I let BitChess play against another chessprogram on my C64 and there was an error in this game, when the C64 program managed, to bring a piece through to the line and i then wanted to select the swap-piece (a queen).

 

Could that be fixed meanwhile, or have you not worked on the program again so far? I ask, because I would like to play a few more games on the PC, against BitChess (by myself and against other chess-programs), to be able to better judge the playing-strength of BitChess. But for that, a game should of course run through completely without an error.

 

[+Andrew Davie]

1 hour ago, AW127 said:

 

Hy Andrew. After a longer time, i wanted to ask, if this error in BitChess could be fixed in the meantime, which occurred at that time, if one of the pieces was exchanged? Do you remember, I had reported about it once here in this thread, when I let BitChess play against another chessprogram on my C64 and there was an error in this game, when the C64 program managed, to bring a piece through to the line and i then wanted to select the swap-piece (a queen).

 

Could that be fixed meanwhile, or have you not worked on the program again so far? I ask, because I would like to play a few more games on the PC, against BitChess (by myself and against other chess-programs), to be able to better judge the playing-strength of BitChess. But for that, a game should of course run through completely without an error.

 

Sorry, no I will not be returning to this project.

In the long run, a chess engine running on an ARM-based cartridge with (perhaps) dual core processor would be the way to proceed.

BitChess isn't/wasn't good enough - many months of expert development goes into 6502 chess engines, particularly the evaluation algorithms. This one isn't nearly strong enough at this stage, and I'm not inclined to spend more time as I have other projects that I expect will take all available time for the forseeable future.

Sorry... 

[AW127]

9 hours ago, Andrew Davie said:

Sorry, no I will not be returning to this project.

In the long run, a chess engine running on an ARM-based cartridge with (perhaps) dual core processor would be the way to proceed.

BitChess isn't/wasn't good enough - many months of expert development goes into 6502 chess engines, particularly the evaluation algorithms. This one isn't nearly strong enough at this stage, and I'm not inclined to spend more time as I have other projects that I expect will take all available time for the forseeable future.

Sorry... 

 

Okay, understand. Too bad, looked really good, but it's also clear to me, that ARM-based chess offers more possibilities and also higher playing strength.

[Bomberman94]

😔yes - it took a Long time while „thinking“ but I enjoyed playing your chess game as I’m a big fan of chess ♟️.