Excalibur King Arthur Revisited

This forum is for general discussions and questions, including Collectors Corner and anything to do with Computer chess.

Moderators: Harvey Williamson, Steve B, Watchman

Forum rules
This textbox is used to restore diagrams posted with the fen tag before the upgrade.
User avatar
spacious_mind
Senior Member
Posts: 4000
Joined: Wed Aug 01, 2007 10:20 pm
Location: Alabama
Contact:

Post by spacious_mind »

The King Arthur Wiki Elo's can easily be explained. The KA Deluxe 16 games come from my U1400 Tournament and is the result of playing 2 games against 8 opponents. The KA shows higher because someone else posted 28 games with it. Often ratings depend on which opponents were played:

229 King Arthur Deluxe : 1303 16 (+ 9,= 5,- 2), 71.9 %

4 in 1 : 2 (+ 2,= 0,- 0), 100.0 %
CXG Enterprise S /, Merlin 4K : 2 (+ 0,= 1,- 1), 25.0 %
Junior (Sensor) / Marauder : 2 (+ 1,= 1,- 0), 75.0 %
Alto : 2 (+ 1,= 1,- 0), 75.0 %
Yeno Travel Chess : 2 (+ 1,= 1,- 0), 75.0 %
Orion 6 in 1 : 2 (+ 2,= 0,- 0), 100.0 %
Executive Chess : 2 (+ 2,= 0,- 0), 100.0 %
LCD C & C 375-2-CS : 2 (+ 0,= 1,- 1), 25.0 %

224 King Arthur : 1333 28 (+ 4,= 2,- 22), 17.9 %

Sargon 4.0 : 4 (+ 0,= 0,- 4), 0.0 %
Advanced Starchess : 10 (+ 1,= 0,- 9), 10.0 %
Cavalier Chess Companion III : 3 (+ 1,= 1,- 1), 50.0 %
Sensory 9 : 10 (+ 1,= 1,- 8), 15.0 %
Chess Station : 1 (+ 1,= 0,- 0), 100.0 %

As you can see from the above King Arthur played opponents out of it's league and lost almost every game but still recorded a better rating. Playing much higher rated opponents does not really work well with ELO ratings, especially when only few games were played.

PS. I don't believe it was Barnes either. Because of the Krypton heritage I lean towards Levy on the early weak computers, Danielsen on the improved computers ie KM3 & KA upwards as well as the LCD's and then of course Horvath on the Igor & Ivan's.

ps.. I read in a reputable monthly magazine of that time, that Mirage was bought from Eric White, if that is so then the Robot technology in Phantom Force also leads back to Danielsen or Horvath.

Those are the theories that I favor.

Best regards,

Nick
Nick
User avatar
Mentat
Member
Posts: 68
Joined: Sun Nov 02, 2014 1:11 am
Location: Nis, Serbia

Post by Mentat »

Monsieur Plastique wrote:The King Arthur is an early Excalibur program written by Ron Nelson. Craig Barnes has no association with it whatsoever but it is correct that he did have a lot to do with the 8K programs written for Saitek, such as Sensor XL, etc.

I am not sure where the 8K association with King Arthur comes from. All the Ron Nelson Excalibur programs are 32K ROM and the programs started off with 6502 processors at varying clock speeds back in the 73 level days (from memory 5 to 6 Mhz). By the time the program had been "refined" towards the end of the Excalibur days, it was still a 32K program but significantly stronger owing to the use of RISC processors and faster clock speeds, much more level variety, etc.

So basically you are starting off with early programs along the evolutionary line (73 level King Arthur running on a 6502 at slow clock speeds) through to the last (Alexandra / Deluxe Talking Touch Chess) with much more expansive levels, H8 processor and around the 10 Mhz mark.
Thank you so much for shedding more light on this! I was only trying to make a bold deduction from the scant information on the King Arthur that I'd been able to find in different places. And I was not sure that the author was really Craig Barnes. Now, with the information you offered, everything seem stop click. Still, if the program in the KA is 32k I find it very perplexing that it is so stupendously incompetent in certain endgames (where are some basic heuristics?). For example, if you set up the following position: WKh1, WhiteP a5, b5, c5; BKh8, BlackP a7, b7, c7 then the King Arthur simply can't find the winning 1.b6! for the life of it. It reaches depth 10 and gets stuck there mulling only over 1.Kg2??. This makes me still wonder about the code. And I the brute force mode made no difference either, as the KA plays right off 1.Kg2 even though on the infinite level.

Another thing: the Novag Opal Plus appears to be a cute little machine and I will consider buying one. Thanks a lot for your suggestion.
User avatar
Mentat
Member
Posts: 68
Joined: Sun Nov 02, 2014 1:11 am
Location: Nis, Serbia

Post by Mentat »

Dave, thank you again for your reply and the links you provided. BTW, I don't think that it's the ego that demands victories against your computer as you grow older, it's just the always present "infantile" wish to play on more or less equal terms against an opponent. Because, as I see it, current computer programs on super fast PCs have become so strong that humans consider them (rightly so) unbeatable. So we crave for opponents that will be beatable in principle, that will be at human level so that they can play against them (and thus satisfy that infantile need for playing games). I may be wrong, but when I think about my motives for playing chess against a program I feel that this explanation sounds right. Especially in the light of my latest attempt to take on a strong program on my PC when I lost 4-5 games in a row at game in 5 plus 5 (Bronstein increment) without having a fighting chance and feeling utterly helpless.

I want to have a chance of beating the machine regards,

Djordje
User avatar
Mentat
Member
Posts: 68
Joined: Sun Nov 02, 2014 1:11 am
Location: Nis, Serbia

Post by Mentat »

spacious_mind wrote:The King Arthur Wiki Elo's can easily be explained. The KA Deluxe 16 games come from my U1400 Tournament and is the result of playing 2 games against 8 opponents. The KA shows higher because someone else posted 28 games with it. Often ratings depend on which opponents were played:

229 King Arthur Deluxe : 1303 16 (+ 9,= 5,- 2), 71.9 %

4 in 1 : 2 (+ 2,= 0,- 0), 100.0 %
CXG Enterprise S /, Merlin 4K : 2 (+ 0,= 1,- 1), 25.0 %
Junior (Sensor) / Marauder : 2 (+ 1,= 1,- 0), 75.0 %
Alto : 2 (+ 1,= 1,- 0), 75.0 %
Yeno Travel Chess : 2 (+ 1,= 1,- 0), 75.0 %
Orion 6 in 1 : 2 (+ 2,= 0,- 0), 100.0 %
Executive Chess : 2 (+ 2,= 0,- 0), 100.0 %
LCD C & C 375-2-CS : 2 (+ 0,= 1,- 1), 25.0 %

224 King Arthur : 1333 28 (+ 4,= 2,- 22), 17.9 %

Sargon 4.0 : 4 (+ 0,= 0,- 4), 0.0 %
Advanced Starchess : 10 (+ 1,= 0,- 9), 10.0 %
Cavalier Chess Companion III : 3 (+ 1,= 1,- 1), 50.0 %
Sensory 9 : 10 (+ 1,= 1,- 8), 15.0 %
Chess Station : 1 (+ 1,= 0,- 0), 100.0 %

As you can see from the above King Arthur played opponents out of it's league and lost almost every game but still recorded a better rating. Playing much higher rated opponents does not really work well with ELO ratings, especially when only few games were played.

PS. I don't believe it was Barnes either. Because of the Krypton heritage I lean towards Levy on the early weak computers, Danielsen on the improved computers ie KM3 & KA upwards as well as the LCD's and then of course Horvath on the Igor & Ivan's.

ps.. I read in a reputable monthly magazine of that time, that Mirage was bought from Eric White, if that is so then the Robot technology in Phantom Force also leads back to Danielsen or Horvath.

Those are the theories that I favor.

Best regards,

Nick
Nick,

Danielsen's programs are more tactical and even the 4k Danielsen is able to find quick tactical solutions, which is not the case with the KA. So I think that maybe Levy is the author and that the program is quite small ands simple. It plays reasonable chess but is prone to tactical oversights, which is not the case with Kaare Danielsen's programs (his 16k Super Enterprise is a tactical wizard).

Just an idea. But, contrary to M. Plastique, I believe that the KA can't be more than a handful of kb's. 32k as surmised by him is out of the question in this case unless the slow hardware makes it so "impotent" as it were (that single chip runs at snail's pace).

Cheers,
Djordje
User avatar
Monsieur Plastique
Senior Member
Posts: 1014
Joined: Thu Jul 03, 2008 9:53 am
Location: On top of a hill in eastern Australia

Post by Monsieur Plastique »

Mentat wrote:Just an idea. But, contrary to M. Plastique, I believe that the KA can't be more than a handful of kb's. 32k as surmised by him is out of the question in this case unless the slow hardware makes it so "impotent" as it were (that single chip runs at snail's pace).
This information came straight from the man himself - Ron Nelson. I'm not sure how much more factual you are hoping to get. I have literally seen the emails typed by him before my very own eyes citing the program ROM size, chip types and clock speeds of the varying models. It's been some years now and the precise speeds are hazy but I distinctly remember they were all 32K programs with 6502 being used from King Arthur through to at least the LCD Talking and I think clock speeds (at best) around the 6 MHz mark for the more "premium" models. My recollection is that those earlier 32K models had 3MHz, 5 MHz and 6 MHz 6502 hardware depending on the model. The strength (or lack thereof) of the programs is attributable to the program being refined over the years, the changes in hardware (ultimately to RISC processors at higher clock speeds around 10 MHz) plus the inescapable fact that there were other professional dedicated programmers out there who were capable of achieving significantly better results with more modest or comparable hardware.

For the same hardware, you might find a program written by Ron Nelson may be up to 400 plus points weaker than a program written by the "big" names on the equivalent hardware. For example, MMV by Ed Schroder on the same equivalent hardware as the LCD Talking Chess was around 400 points stronger. And something like the Morsch 32K program on Saitek machines running an H8 at 10 MHz was also roughly 400 points stronger than, say, Alexandra or DTTC - both of which were amongst Nelson's final and strongest efforts.

Remember too that just because a dedicated program is 32K does not mean it really is a 32K engine. It means the ROM chip on the machine is 32K. Plus you need space for the opening book and other features not directly related to the engine itself. Back in the pure dedicated days they did not necessarily "fill up" the chips to the brim. Especially when they went from 16K to 32K chips.
Chess is like painting the Mona Lisa whilst walking through a minefield.
SirDave
Full Member
Posts: 610
Joined: Thu Sep 09, 2010 4:59 am
Location: Southern California USA

Post by SirDave »

Mentat wrote:Dave, thank you again for your reply and the links you provided. BTW, I don't think that it's the ego that demands victories against your computer as you grow older, it's just the always present "infantile" wish to play on more or less equal terms against an opponent. Because, as I see it, current computer programs on super fast PCs have become so strong that humans consider them (rightly so) unbeatable. So we crave for opponents that will be beatable in principle, that will be at human level so that they can play against them (and thus satisfy that infantile need for playing games). I may be wrong, but when I think about my motives for playing chess against a program I feel that this explanation sounds right. Especially in the light of my latest attempt to take on a strong program on my PC when I lost 4-5 games in a row at game in 5 plus 5 (Bronstein increment) without having a fighting chance and feeling utterly helpless.

I want to have a chance of beating the machine regards,

Djordje
I agree. What happened with me is that when I started back at chess a few years ago, I had victories over weaker boards (such as the portable Saitek Computer Plus Coach which I beat at all levels) and then, getting a little full of myself, moved on to stronger units setting them at stronger levels. However, then I started losing far more than winning and found myself not wanting to play as much.

It finally dawned on me that I was missing the point of playing which I think you elude to above. I started playing the boards at weaker levels and the victories and the joy returned. :) For instance, the other day, the board I was playing walked into a very nice fork which put it in a losing position. I hadn't seen a fork for a long time. The stronger boards at stronger levels don't offer forks up much at all unless you can force them into the fork and I don't have the skill to do that all that much unless I give the board less time to think.

BTW: It's the Novag Star Opal you might want to consider- the Novag Opal Plus is a weaker predecessor.
User avatar
Mentat
Member
Posts: 68
Joined: Sun Nov 02, 2014 1:11 am
Location: Nis, Serbia

Post by Mentat »

Monsieur Plastique wrote:
Mentat wrote:Just an idea. But, contrary to M. Plastique, I believe that the KA can't be more than a handful of kb's. 32k as surmised by him is out of the question in this case unless the slow hardware makes it so "impotent" as it were (that single chip runs at snail's pace).
This information came straight from the man himself - Ron Nelson. I'm not sure how much more factual you are hoping to get. I have literally seen the emails typed by him before my very own eyes citing the program ROM size, chip types and clock speeds of the varying models. It's been some years now and the precise speeds are hazy but I distinctly remember they were all 32K programs with 6502 being used from King Arthur through to at least the LCD Talking and I think clock speeds (at best) around the 6 MHz mark for the more "premium" models. My recollection is that those earlier 32K models had 3MHz, 5 MHz and 6 MHz 6502 hardware depending on the model. The strength (or lack thereof) of the programs is attributable to the program being refined over the years, the changes in hardware (ultimately to RISC processors at higher clock speeds around 10 MHz) plus the inescapable fact that there were other professional dedicated programmers out there who were capable of achieving significantly better results with more modest or comparable hardware.

For the same hardware, you might find a program written by Ron Nelson may be up to 400 plus points weaker than a program written by the "big" names on the equivalent hardware. For example, MMV by Ed Schroder on the same equivalent hardware as the LCD Talking Chess was around 400 points stronger. And something like the Morsch 32K program on Saitek machines running an H8 at 10 MHz was also roughly 400 points stronger than, say, Alexandra or DTTC - both of which were amongst Nelson's final and strongest efforts.

Remember too that just because a dedicated program is 32K does not mean it really is a 32K engine. It means the ROM chip on the machine is 32K. Plus you need space for the opening book and other features not directly related to the engine itself. Back in the pure dedicated days they did not necessarily "fill up" the chips to the brim. Especially when they went from 16K to 32K chips.
Monsieur Plastique, I really appreciate your explanation. One really can't be more factual than this! And your comparisons between the 32k in early Nelson and the same amount of memory in Morsch and Schroeder are spot on. What I found puzzling was the lack of finesse in Nelson's KA's endgame knowledge that I couldn't figure out... Anyway, now I am sure about my KA's origin and 'place of birth'. The chip itself offered lots of room for the openings, yes -- but the KA has only a 400 ply book or so and not much of the I/O needs (only the move beeps and the LCD board to display move transfer; the latter may have taken up some kb's, though...).

Thanks a lot --- I am now almost finished with exploring the little thing that I still enjoy, despite its chessic frailties. Its excellent form factor and sensible preservation of the pawn structure make it fun to play, setting aside the occasional oversight or two. It even "sees" mate in five sequences and delivers them ruthlessly. I guess I was just hard-headed at first, not ready to connect Ron Nelson of my CC7 fame with the little board.

best

Djordje
Mike Watters
Member
Posts: 429
Joined: Fri Sep 26, 2008 12:31 pm
Location: Milton Keynes
Contact:

Post by Mike Watters »

Hi Jorge

I think it is fair to say that the chess computer community is split down the middle on the question of who programmed those Excaliburs i.e the ones not obviously cloned and therefore known to be programmed by Kittinger, Taylor, Horvath or Morsch. So this debate is becoming a rerun of an earlier thread -

http://www.hiarcs.net/forums/viewtopic. ... ght=#66538

Could Ron Nelson be the conductor rather than the composer? If he wrote the spec, picked a chess engine off the shelf, added I/O and features and tweaked the result he would be responsible for the end result but not the author of the chess engine. Otherwise you believe that after writing Z80 programs up to around 1980/81, he spent the next 15 years not writing a new chess engine only to pop up out of the blue writing 6502 and H8 chess engines in the late 90s. At a time when the manufacturers of these machines in China had programmers and suitable engines on the shelf. Why would anyone do that and not a whisper about the source of the programs for another ten years? Nothing in any publication at the time, or since.

All the best
Mike
User avatar
Mentat
Member
Posts: 68
Joined: Sun Nov 02, 2014 1:11 am
Location: Nis, Serbia

Post by Mentat »

Mike Watters wrote:Hi Jorge

I think it is fair to say that the chess computer community is split down the middle on the question of who programmed those Excaliburs i.e the ones not obviously cloned and therefore known to be programmed by Kittinger, Taylor, Horvath or Morsch. So this debate is becoming a rerun of an earlier thread -

http://www.hiarcs.net/forums/viewtopic. ... ght=#66538

Could Ron Nelson be the conductor rather than the composer? If he wrote the spec, picked a chess engine off the shelf, added I/O and features and tweaked the result he would be responsible for the end result but not the author of the chess engine. Otherwise you believe that after writing Z80 programs up to around 1980/81, he spent the next 15 years not writing a new chess engine only to pop up out of the blue writing 6502 and H8 chess engines in the late 90s. At a time when the manufacturers of these machines in China had programmers and suitable engines on the shelf. Why would anyone do that and not a whisper about the source of the programs for another ten years? Nothing in any publication at the time, or since.

All the best
Mike
Mike,

nice to have you in this discussion. It's always exciting to try to unearth a fact that's buried in fallible human memory. For instance, what is not completely obvious here (and mind you, I do respect Monsieur Plastique's elucidation, of course) is the huge, really huge effort that Ron Nelson must have put in translating the Z80 assembly code of the Challenger series into another chess program written for the 65C02 chip, in a different assembly. These programs had to suit a particular CPU assembly (for instance it took Morsch quite some time to adapt to C++ so as to code his Fritz for Intel, and still had to write some specific routines in assembly; because this transition from assembly to a higher language takes a while). At the end of the day it just may turn out that we do not have complete data, apart from honest witness statements that I do trust.

Puzzled nodding regards,

Djordje
User avatar
spacious_mind
Senior Member
Posts: 4000
Joined: Wed Aug 01, 2007 10:20 pm
Location: Alabama
Contact:

Post by spacious_mind »

Mentat wrote:
Mike Watters wrote:Hi Jorge

I think it is fair to say that the chess computer community is split down the middle on the question of who programmed those Excaliburs i.e the ones not obviously cloned and therefore known to be programmed by Kittinger, Taylor, Horvath or Morsch. So this debate is becoming a rerun of an earlier thread -

http://www.hiarcs.net/forums/viewtopic. ... ght=#66538

Could Ron Nelson be the conductor rather than the composer? If he wrote the spec, picked a chess engine off the shelf, added I/O and features and tweaked the result he would be responsible for the end result but not the author of the chess engine. Otherwise you believe that after writing Z80 programs up to around 1980/81, he spent the next 15 years not writing a new chess engine only to pop up out of the blue writing 6502 and H8 chess engines in the late 90s. At a time when the manufacturers of these machines in China had programmers and suitable engines on the shelf. Why would anyone do that and not a whisper about the source of the programs for another ten years? Nothing in any publication at the time, or since.

All the best
Mike
Mike,

nice to have you in this discussion. It's always exciting to try to unearth a fact that's buried in fallible human memory. For instance, what is not completely obvious here (and mind you, I do respect Monsieur Plastique's elucidation, of course) is the huge, really huge effort that Ron Nelson must have put in translating the Z80 assembly code of the Challenger series into another chess program written for the 65C02 chip, in a different assembly. These programs had to suit a particular CPU assembly (for instance it took Morsch quite some time to adapt to C++ so as to code his Fritz for Intel, and still had to write some specific routines in assembly; because this transition from assembly to a higher language takes a while). At the end of the day it just may turn out that we do not have complete data, apart from honest witness statements that I do trust.

Puzzled nodding regards,

Djordje
I think that you can also look at it from the perspective that Excalibur would have been sued if a program were to look remotely close to a Fidelity, Mephisto or Saitek product. Especially after the way Fidelity went down and in a way shafted Mephisto. I doubt very much that Nelson could have used anything that belonged to the other companies including his own early Z80 program. Which is proven by the fact that you don't see anything remotely close to Mephisto, Fidelity or Saitek in their products.

It would have been much easier to use his program in the weaker Excalibur programs instead using Levy if that were the case. Therefore Nelson writing a program from scratch while at the same time working on Product Design, selling, advertising, marketing, payroll, HR, Purchasing etc etc seems to me a long stretch. When Excalibur was started you can bet on it that it was staffed to the barest minimum so everyone working there was multifunctional in what they had to do to get it off the ground and keep it going forward.

I know how hard I work and what I can achieve in a day, therefore unless he is superhuman I doubt he had the time even if he wanted to, for creating a new program from scratch that coincidentally, continuously plays at a similar play strength to programs developed by both Danielsen and Horvath.

If any of you have run a business then I think you would know exactly what I am talking about here. You cannot do everything, it is impossible.

Besides which business heads in their right mind would actually tell you who did what, why, where and when? Of course they wouldn't. They tell you what they think you would like to hear instead.

Best regards

Nick
Nick
SirDave
Full Member
Posts: 610
Joined: Thu Sep 09, 2010 4:59 am
Location: Southern California USA

Post by SirDave »

Okay, now we've heard all the theories and evidence. Can someone put this together in a nutshell as to what the consensus is as to who programmed the King Arthur & other Excaliburs of that period?
User avatar
Steve B
Site Admin
Posts: 10140
Joined: Sun Jul 29, 2007 10:02 am
Location: New York City USofA
Contact:

Post by Steve B »

SirDave wrote:Okay, now we've heard all the theories and evidence. Can someone put this together in a nutshell as to what the consensus is as to who programmed the King Arthur & other Excalibur's of that period?
i cannot add anything new to this debate other then what i said in the last debate regarding the later Excaliburs
Nelson told me he programmed them..period
for this ...i was raked over the coals ..dragged though the mud...called a liar an idiot and an exaggerator ..and that was some of the more gentler/kinder comments aimed my way
:P

its nice to just sit back and read all of the theories as they are bandied about

Laying Low Regards
Steve
Last edited by Steve B on Tue Nov 25, 2014 12:21 am, edited 1 time in total.
User avatar
spacious_mind
Senior Member
Posts: 4000
Joined: Wed Aug 01, 2007 10:20 pm
Location: Alabama
Contact:

Post by spacious_mind »

SirDave wrote:Okay, now we've heard all the theories and evidence. Can someone put this together in a nutshell as to what the consensus is as to who programmed the King Arthur & other Excaliburs of that period?
Well Dave my theory is:

All the weak ones = Levy origin to about Kingmaster II
Mid Range = Danielsen origin
Top end = Horvath origin

Of course there is going to be some overlap between them as well.

Danielsen for Example has 4 play styles. Typically they are Normal, Aggressive, Defensive and Random. Sometimes they are called differently like Desperado.

E-Chess for example not so obvious also has 4 play styles, cleverly called differently:

1) Fast ON Random OFF
2) Fast ON Random ON
3) Fast OFF Random ON
4) Fast OFF Random OFF

These type of themes are continuous when you study them closer and no surprise the play strength always ends up being pretty similar. Especially when you start comparing them to the Lexibooks, Millenniums, Kryptons, Systemas and CXG.

Best regards

Best regards
Nick
User avatar
Monsieur Plastique
Senior Member
Posts: 1014
Joined: Thu Jul 03, 2008 9:53 am
Location: On top of a hill in eastern Australia

Post by Monsieur Plastique »

spacious_mind wrote:Therefore Nelson writing a program from scratch while at the same time working on Product Design, selling, advertising, marketing, payroll, HR, Purchasing etc etc seems to me a long stretch. When Excalibur was started you can bet on it that it was staffed to the barest minimum so everyone working there was multifunctional in what they had to do to get it off the ground and keep it going forward.
I don't feel that such a scenario is that long a stretch. Nelson was employed right from the very beginning in a senior technical capacity and there were other ex-Fidelity employees (apparently, according to the CCR of the time) to serve in other capacities.

When I worked for a large telecommunications firm I could argue that my experience there was similar to Mr. Nelson's at Excalibur. When there was a major company restructure in the early 90s I relocated to their head office finance group in a technology related role. As there was simply no off the shelf reporting system in existence for certain accounting / analysis tasks, I had to develop an extremely complex financial reporting system absolutely 100% from scratch. Not one single line of code existed. For the first 12 months I simply worked like mad, often putting in 70 hour weeks until the basic system was up and running. I had other duties relating to system administration, analysis and reporting (my "normal") job, but I knew that once this new system was up and running, we would all be in a sort of "maintenance and ongoing development" mode thereafter.

There was one other period around 4 years later where again there were major changes to the company and their associated systems and databases. This obviously required a second effort to effectively overhaul the system, though the effort was not to the extent of the first time around.

Sure enough, in both cases, once the system was up and running further work would mainly be tweaks, normal maintenance and the addition of minor rather than major features. Although the initial onslaught creating the thing was indeed horrible in some respects, once it was actually there it was well within my ability to enhance it in an ongoing capacity. At worst work would be needed in short bursts - something most people are easily capable of.

If you translate this to Excalibur, it may have been a case that Excalibur wanted their "own" chess engine to be created in house and exclusive to the other agreements they were going to have for product sharing, etc. I don't think it is unreasonable that this engine was created in the early days and that Ron did infact find the time to keep enhancing it over time and to adapt it to the various models they produced.

Remember too that writing chess programs was originally Ron Nelson's hobby. It was not as if you had to drag him into an office and force him to do it. When something begins as a hobby, there is usually less motivation required to do it well on a professional basis and to achieve greater productivity then for someone who just views it as a chore they need to complete in order to pick up the fornightly pay cheque. It is also easier to put in extended hours.

Even when I was sitting at my computer writing these systems from scratch - sometimes even on Saturday nights and Sundays - there was always an element of fun and excitement involved, because I had used computers as a hobby earlier in life and - just like a chess engine programmer - you get a real kick out of finally resolving a bug or finding a more efficient way of doing something. Or indeed, solving a difficult issue to begin with. And it is not hard for me to imagine that developing a chess engine is a lot more fun and exciting than developing a financial reporting system...

To be perfectly honest, some of the best evidence that all these creations bar the "better" Igor and GM were from the keyboard of Nelson comes down to the playing strength achieved with the hardware he was using. The big name programmers always achieved results for given hardware within a relatively limited ELO range - at worst maybe 150 points from "best" to "worst", whereas these Excalibur programs were a good 300 - 400 points behind or so. This gap really is not any different to the early 80s when you had 4K Nelson programs in Fidelity low-end products (i.e Challenger 8, Gambit) and the 4K Danielson program written for Saitek which found it's way in various guises into Companion II, the portable Explorer, Concord and then many years later into the Lexibooks with late generation micro controllers and with more features such as the level choices eluded to earlier.

If these Excalibur machines had all played a couple of classes stronger, I would seriously doubt that Nelson had anything to do with them. But their playing strength just in itself is perhaps the strongest evidence you can have, assuming you are not swayed by the communications from the man himself claiming authorship. It doesn't matter if you are talking about the Fidelity Challenger 8, the earliest King Arthur or the latest Alexandra: the relative capabilities, character and deficiencies of these programs in comparison to their contemporaries is always the same.

I would love for Ron Nelson to write an autobiography. His journey would have been a very fascinating one through a unique era.
Chess is like painting the Mona Lisa whilst walking through a minefield.
User avatar
spacious_mind
Senior Member
Posts: 4000
Joined: Wed Aug 01, 2007 10:20 pm
Location: Alabama
Contact:

Post by spacious_mind »

Monsieur Plastique wrote:
spacious_mind wrote: I would love for Ron Nelson to write an autobiography. His journey would have been a very fascinating one through a unique era.
Me too.

Let me ask you in this way. Do you disagree with the fact that every chess computer that Excalibur sold in the first few years came from Krypton and CXG?

If you agree with this then you would also have to agree there is no question that Excalibur had the programs of other programmers in their hand every single day from inception to at some point four years later when the supposed Nelson program was invented. How about you telling us when this occurred. Which was Nelson's program?

ps. I totally disagree with you. You make it sound that all Nelson had to do at Excalibur was write a program. Therefore as director of the company that was his station in life to be a programmer.

Pay a man 6 figures plus per year plus give him company shares to create a program from scratch, when you have good programmers in Hungary or somewhere else whose program you could buy cheap at Eastern European rates or not even buy at all since they were previously bought by CXG and Krypton. That makes true business sense.

ps. you do realize that Nelsons programs were totally inadequate with complete inability to check mate. Try to set up Nelsons best Sensory Voice Challenger with King against King and Queen and see if it achieves mate in 50 moves. His best program is inadequate, now after 15 years of non programming he finds the ability to create a good program from scratch.

Best regards
Nick
Post Reply