Comparing Boris versions

[bataais]

Hi,

Maybe you all know there were two wooden Boris models, the conventional and the Master.

Here is what Daniel Collin says about them on his nice Boris website:
https://www.borischesscomputer.com/homepage.htm

On the "Boris Master", you can connect the adapter on the unit’s back via a removable connector (onto the preceding "Boris", the adapter wire is welded directly onto the unit inside the compartment). Another characteristic of this model, the programmers changed the comments that "Boris Master" gives on human moves (it would have been preferable to use this intervention in the memory to create a small openings book).

Update - ATTENTION concerning information of the preceding paragraph - I realized that some units of the conventional Boris were marketed (in America as in Europe) with removable connector on the back of the unit as for the Master model. Moreover, these units rather post the Master model starting message on their activation "Boris awaits your move" than the "Boris plays black" of the conventional Boris. Its look like obvious that Applied Concepts used the electronic units of the Master model to produce Boris conventional units...

Now I compared both machines in MAME/Mess debugger and found out that they have a different board representation in memory, and also play different. (I let the machines go through all 20 opening moves and stopped)

I also compared them to the Boris Diplomat and the Novag Delta-1. Here are the memory dumps:

Boris (rev. 00) & Boris Diplomat:

Code: Select all

0C00:  04 02 03 05 06 03 02 04 00 44 40 40 00 40 44 00  .........D@@.@D.
0C10:  01 01 01 01 01 01 01 00 44 48 40 C8 C8 40 48 44  ........DH@..@HD
0C20:  00 00 00 00 00 00 00 00 8C 8C CC 8C 8C CC 4C CC  ..............L.
0C30:  00 00 00 00 00 00 00 01 00 00 00 00 00 00 00 44  ...............D
0C40:  00 00 00 00 00 00 00 00 00 00 00 00 00 00 4C 00  ..............L.
0C50:  00 00 00 00 00 00 00 00 23 23 33 23 23 33 23 23  ........##3##3##
0C60:  FF FF FF FF FF FF FF FF 11 12 10 32 32 10 12 11  ...........22...
0C70:  FC FE FD FB FA FD FE FC 00 11 10 10 00 10 11 00  ................
0C80:  38 13 23 13 33 14 24 14 34 00 00 00 00 00 00 00  8.#.3.$.4.......
0C90:  00 0F 0F 0F 18 12 13 0E 16 10 21 20 17 28 16 10  ..........! .(..
0CA0:  18 20 17 19 1F 10 10 0F 00 00 00 00 00 00 00 00  . ..............
0CB0:  00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
0CC0:  00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
0CD0:  00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
0CE0:  00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
0CF0:  00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................

Boris Diplomat has the same MEM (0C00-0C8F, except 0c80 is 18 instead of 38, the rest of the MEM in Boris is IMO used for the display).
They both play the same.

Boris (rev. 01):

Code: Select all

0C00:  00 50 40 40 00 40 50 00 50 60 40 E0 E0 40 60 50  .P@@.@P.P`@..@`P
0C10:  B0 B0 F0 B0 B0 F0 70 F0 00 00 00 00 00 00 00 50  ......p........P
0C20:  00 00 00 00 00 00 70 00 0B 0B 0F 0B 0B 0F 0B 0B  ......p.........
0C30:  05 06 04 0E 0E 04 06 05 00 05 04 04 00 04 05 00  ................
0C40:  04 02 03 05 06 03 02 04 01 01 01 01 01 01 01 00  ................
0C50:  00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 01  ................
0C60:  00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
0C70:  FF FF FF FF FF FF FF FF FC FE FD FB FA FD FE FC  ................
0C80:  F7 03 13 23 13 33 14 24 14 34 00 00 00 00 00 00  ...#.3.$.4......
0C90:  00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
0CA0:  00 00 D8 D8 00 E3 DF E0 DA E1 D8 D6 EB D6 DA DD  ................
0CB0:  E1 D8 ED DF E2 E0 D8 DC DF EA D9 D8 D8 00 00 00  ................
0CC0:  00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
0CD0:  00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
0CE0:  00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
0CF0:  00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................

CC Delta-1:

Code: Select all

2000:  00 50 40 40 00 40 50 00 50 60 40 E0 E0 40 60 50  .P@@.@P.P`@..@`P
2010:  B0 B0 F0 B0 B0 F0 70 F0 00 00 00 00 00 00 00 50  ......p........P
2020:  00 00 00 00 00 00 70 00 0B 0B 0F 0B 0B 0F 0B 0B  ......p.........
2030:  05 06 04 0E 0E 04 06 05 00 05 04 04 00 04 05 00  ................
2040:  04 02 03 05 06 03 02 04 01 01 01 01 01 01 01 00  ................
2050:  00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 01  ................
2060:  00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
2070:  FF FF FF FF FF FF FF FF FC FE FD FB FA FD FE FC  ................
2080:  F7 00 14 34 06 25 12 32 13 33 90 CA 78 6B 5C 7F  ...4.%.2.3..xk\.
2090:  8B 17 D1 13 91 40 2E 66 48 20 99 B1 6E 37 18 EA  .....@.fH ..n7..
20A0:  CC 49 C0 66 90 A6 2D BE E1 AB 33 F4 AE 86 D4 D8  .I.f..-...3.....
20B0:  C8 3F 73 A6 31 F5 48 F1 B4 66 14 C5 B6 B2 16 09  .?s.1.H..f......
20C0:  4D 61 3F 24 90 A4 A3 E1 2E D9 30 95 43 53 A3 7F  Ma?$......0.CS..
20D0:  68 59 B9 71 0A 6C A3 AE FD CF BD 14 52 42 80 7B  hY.q.l......RB.{
20E0&#58;  66 0E B6 5E 3C 48 EE B9 0E 4F 30 33 1F 98 F2 7E  f..^<H...O03...~
20F0&#58;  D4 AC 4B FC 04 6F 68 A4 8D A3 3D 24 28 AE 7E 48  ..K..oh...=$&#40;.~H

Delta-1 looks very similar to Boris rev01. (The Delta-1 MEM (208A-20FF) seems to be junk data.) A difference is in the evaluation / best moves (Boris rev1: 0C80-0C89 - Delta-1: 2080-2089).

I let them play two games against each other, both times Boris rev01 won.

Interestingly Boris rev01 solves the chess problems a bit faster than his predecessor.

https://www.borischesscomputer.com/bori ... oblems.htm

Boris is King Regards,
Michael

[bataais]

Another very detailed site about the Boris versions:
http://www.chesseval.com/Eldest/AppliedConcepts.htm

[Tibono2]

Delta-1 looks very similar to Boris rev01. (The Delta-1 MEM (208A-20FF) seems to be junk data.) A difference is in the evaluation / best moves (Boris rev1: 0C80-0C89 - Delta-1: 2080-2089).

Thanks for this interesting comparison. Delta-1 program author is currently unknown AFAIK, maybe David Lindsay so?
I am not aware of any relationship between Applied Concept/Chafitz & Novag...
Anyone able to put some light on this?
Cheers,
Eric

[bataais]

I am not aware of any relationship between Applied Concept/Chafitz & Novag...

We can only speculate.. Look at the first 3 Scisys/Novag models (assuming Delta-1 was held back and released after CC Super System Mark III), they are all either clones or (illegal) copies. I can't believe David Lindsay would make an update of his program weaker than in the Boris..

Pirate Regards,
Michael

[bataais]

Another very interesting memory comparison, this time between Boris (rev.00) and CC7 (BCC). I entered the move F2-F3 on both (in messchess), set some breakpoints and stopped at E7-E5 reply calculation:

Boris rev0:
breakpoint 52B2

Code: Select all

0C00:  04 02 03 05 06 03 02 04 00 44 40 40 00 40 44 00  .........D@@.@D.
0C10:  01 01 01 01 01 00 01 01 44 48 40 C8 C8 40 48 44  ........DH@..@HD
0C20:  00 00 00 00 00 01 00 00 9E 8C CC 8C 4C CC 4C 8C  ............L.L.
0C30:  00 00 00 00 00 00 00 00 00 12 00 13 4C 13 4C 10  ............L.L.
0C40:  00 00 00 00 FF 00 00 00 00 00 12 00 00 00 10 00  ................
0C50:  00 00 00 00 00 00 00 00 23 23 33 23 23 33 23 23  ........##3##3##
0C60:  FF FF FF FF 00 FF FF FF 11 12 10 32 32 10 12 11  ...........22...
0C70:  FC FE FD FB FA FD FE FC 00 11 10 10 00 10 11 00  ................
0C80:  33 63 43 64 54 61 41 62 52 62 42 00 00 00 00 00  3cCdTaAbRbB.....
0C90:  00 0F 0F 0F 18 23 15 10 12 1C 10 19 12 23 13 16  .....#.......#..
0CA0:  1B 10 10 0F 1F 10 10 0F 00 00 00 00 00 00 00 00  ................
0CB0:  00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
0CC0:  00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
0CD0:  00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
0CE0:  00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
0CF0:  00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................

CC7-BCC
breakpoint 0A2B

Code: Select all

3000:  04 02 03 05 06 03 02 04 00 44 40 40 40 40 44 00  .........D@@@@D.
3010:  01 01 01 01 01 00 01 01 44 48 40 C8 C8 40 48 44  ........DH@..@HD
3020:  00 00 00 00 00 01 00 00 9E 8C CC 8C 4C CC 4C 8C  ............L.L.
3030:  00 00 00 00 00 00 00 00 00 12 00 13 4C 13 4C 10  ............L.L.
3040:  00 00 00 00 FF 00 00 00 00 00 12 00 00 00 10 00  ................
3050:  00 00 00 00 00 00 00 00 23 23 33 23 23 33 23 23  ........##3##3##
3060:  FF FF FF FF 00 FF FF FF 11 12 10 32 32 10 12 11  ...........22...
3070:  FC FE FD FB FA FD FE FC 00 11 10 10 10 10 11 00  ................
3080:  01 01 00 00 00 00 25 15 16 00 C8 32 FF FC 00 00  ......%....2....
3090:  02 01 63 43 43 00 00 00 00 00 00 00 00 00 00 25  ..cCC..........%
30A0:  35 91 FF 4D 77 76 12 00 B3 01 22 1F 00 8E 64 44  5..Mwv...."...dD
30B0:  1F 00 5B 15 25 71 4F 4D 00 B9 B9 B9 B9 B9 B9 B9  ..[.%qOM........
30C0:  B9 B9 B9 B9 B9 B9 B9 B9 B9 B9 B9 B9 B9 B9 B9 B9  ................
30D0:  B9 B9 B9 B9 B9 B9 B9 B9 B9 B9 B9 B9 B9 B9 B9 B9  ................
30E0:  B9 B9 B9 B9 63 FB FF FF FF FF FF EF 00 00 00 00  ....c...........
30F0:  00 00 40 00 7D 0E 00 00 7D 0E 37 0E AD 30 E6 08  ..@.}...}.7..0..

They use the same board representation and also the same tabulating (for attack counts?) (0C00-0C7F, 3000-307F).

Regards,
Michael

[Tibono2]

So code or memory similarity on a limited scope of bytes provides little evidence...
Another possible authorship for the Delta-1 is Mike Johnson, who authored the Novag Chess Champion Pocket Chess on same CPU Fairchild F8, and the Novag Chess Champion Super System III using same level management (time setting) as Delta-1, and many similar ways to operate the device (such as FP/EP, standing for find position/enter position).
By design, the Delta-1 looks like a much downgraded Super System III, including much less functions & a much less powerful CPU (I estimate it 4 times slower). Might explain the loss in strenght...
best,
Eric

[Tibono2]

Hi,

I now entered a Novag Delta-1 in my stable, so trying to dig a bit further into this somewhat mysterious device.

Why mysterious?

- author unknown,
- launched late 80 while copyrighted in 79,
- its naming is an oddity, sharing the Chess Champion series name but apart from the 'MK' numbering

Year 79 can be read on the packaging, the user's manual and in the ROM as well:



Where WINKLER is Erich Winkler

In addition to the dates, the design is obviously an early one - including hardware & software. I am convinced the Delta-1 should have been the next in series after the MK-I, featuring a huge enhancement regarding the user interface, the weakest achievement from the former one. The Delta-1 would have deserved its stand in Novag/Scisys range of products, as a smaller & cheaper brother for the MK-III, headed to beginners mass market.

Mike Watters points out similarities with the PCB of MK-I as a possible reason for delaying the release, I do agree and, in addition, I point out both MK-II & MK-3 released after the MK-I and before the Delta-1 are 6502 based. My guess is the company did hold releasing the Delta-1 to mitigate any risk for a hardware case, as copyrights were more applyable to hardware design than to software ROM. Once danger away, the products went back to a series of Fairchild based devices: the delayed Delta-1, the Pocket Chess and the Chess Partner 2000.

To my opinion, the MK-II entered the series as an emergency substitute; steered by the requirement to not use a F8 and complete the MK-III position, in order to feed the low-end range. Another 6502, a known programmer (P. Jennings) and the lower strength of Microchess compared to Mike program would fit the expects. I also guess Microchess copyright coud be purchased at a fair cost and the company badly needed to step in legallity.

Now about code similarities: I performed Hex cross-comparisons between F8-based devices, so MK-I, Delta-1, Boris, Boris Diplomat, Pocket Chess.

Found no similarity between MK-I and Delta-1, thus MK-I software is not the reason why the Delta-1 release has been postponed. Reason lies elsewhere, many possibilities but some odds are with the hardware.

Found no similarity as well between Delta-1 and any Mike Johnson device ROM.

But 40% of the Boris Diplomat code can be found in the Delta-1 (counting exact matches at least 8 consecutive bytes). Using the very same counting rule, 43% of Boris Diplomat's code can be found in the Boris code - so the brotherhood level is quite as high between Delta-1 and BorisD, as it is between both Boris devices.

Delta-1's ROM being roughly twice as large as both others, shared code% drops down to 20% of the Delta-1 code found in both Boris.

The similar code might as well be related to the user interface including legal moves check, but TMHO better odds are with the chess engine, according to the RAM similarity Michael did spot in the evaluation area while running the device.

And back to:

I can't believe David Lindsay would make an update of his program weaker than in the Boris..

what about selling an early version (as the Delta-1 design was indeed much older than its release), maybe a simplified/on purpose weakened one... reasons can be imagined, such as price negociation, marketing range expectation from Novag, mitigation of copyright risk should the device enter direct competition with Boris devices... Another story could be pirated Boris code, revamped and disguised to mitigate risks in a better way than it has been done with the MK-I, resulting in a larger code and lowered playing skills. Add a copyright in the ROM as a cherry on the cake, and release on market when litigant failed...et voila.

A must read, as a reminder - with a big thank to Mike.

kind regards,
Eric

[Fernando]

All this stuff about Boris, interesting as it is, remember me of archaeologists digging the grave of a third rate warrior of some of the eight Saxon kingdoms in VII century AC in Britain.

We all became antiquarian regards
Fern