Philidor/Parker Chess for PC (1983)

lstlaure · Post by **lstlaure** » Mon Jun 15, 2020 8:18 pm

Dostojewsky wrote:Sorry to tell you the naked truth but Philidor is not such an sophisticated piece of chesssoftware as the Mark V was.
...
So it is time for me to say goodbye to Philidor. Sorry for the brutal truth.
Here is the whole game. Philidors shows that he is no opponent for Zarkov 3

You are absolutely right! Even Mr. Broughton would agree with you. Philidor was a lot weaker than the Mark V. But comparing Philidor to Zarkov 3 is a bit unfair. Philidor for DOS was written in 1984 with design constraints (memory limit, for example) that Zarkov 3 didn't have, as it came out 10 years later. 10 years in the computer world is an eternity!

Even if Philidor is weaker than plenty of other programs that came out shortly after, it is still an important part of the history of computer chess programming. David Broughton was a pioneer, and must not be forgotten.

Post by **Fernando** » Mon Jun 15, 2020 8:44 pm

Weak and strong are relative terms. Just take into account the games Philidor played and won against respectable old dedicated computers. Of course Zarkov 3 defeated Philidor as also could defeat those computers.
My guess is that those dedicated units are around 1800 Elo, Philidor around 1900 running in current hardware and zarkov 3 2200 in current hardware.
After that is a matter of taste to talk of weak or strong.

relative regards
Fern

mclane · Post by **mclane** » Mon Jun 15, 2020 9:54 pm

lstlaure wrote:
Dostojewsky wrote:Sorry to tell you the naked truth but Philidor is not such an sophisticated piece of chesssoftware as the Mark V was.
...
So it is time for me to say goodbye to Philidor. Sorry for the brutal truth.
Here is the whole game. Philidors shows that he is no opponent for Zarkov 3
You are absolutely right! Even Mr. Broughton would agree with you. Philidor was a lot weaker than the Mark V. But comparing Philidor to Zarkov 3 is a bit unfair. Philidor for DOS was written in 1984 with design constraints (memory limit, for example) that Zarkov 3 didn't have, as it came out 10 years later. 10 years in the computer world is an eternity!

Even if Philidor is weaker than plenty of other programs that came out shortly after, it is still an important part of the history of computer chess programming. David Broughton was a pioneer, and must not be forgotten.

We can try out if Philidor is stronger or weaker then mkv / mkVI.

IMO the algorithms have changed between the dedicated chess computers and the pc engine, as I have shown you by giving you the source about the new algorithm „interest search“ on the AI factory website.

http://aifactory.co.uk/newsletter/2014_ ... inimax.htm

The programmer I spoke with told me that with these changes the software made a jump of at least 100 ELO without beeing more brute force.

Philidor is still a selective engine but plays stronger.

As far as I see it the algorithms used in the dedicated chess computers were published in ICCA journal and websites under the name: SEX algorithm.

https://www.chessprogramming.org/SEX_Algorithm

So the difference between dedicated units and pc engine is the difference between „sex algorithm“ and „interest search“.

It is often said that MkVI is weaker then mkV.

I also want to disagree here too.

Philidor MK VI dedicated chess computer plays different then its predecessor.
But not weaker overall.

I made a dedicated chess computer tournament with 40/120 time control that had the following results after 13 rounds:

Code: Select all

    Motor                                                      Punkte ScChScMeScFiScElMeMiYeMeMeNo    S-B
01&#58; SciSys Superstar 28K &#123;Brettcomputer&#125;                       10,5/13 · = 1 0 1 1 1 1 0 1 1 1 1 1   61,25
02&#58; Chafitz Destini Prodigy &#123;Brettcomputer&#125;                    10,5/13 = · 0 = 1 1 = 1 1 1 1 1 1 1   58,25
03&#58; Scisys MKVI Philidor &#123;Brettcomputer&#125;                       9,0/13  0 1 · 1 0 1 0 = 1 1 = 1 1 1   50,50
04&#58; Mephisto III ESB 6000 &#123;Brettcomputer&#125;                      8,5/13  1 = 0 · 1 0 0 1 1 1 1 1 0 1   50,75
05&#58; SciSys Chess Companion III &#123;Brettcomputer&#125;                 8,0/13  0 0 1 0 · = = 1 1 = 1 1 = 1   42,25
06&#58; Fidelity Chess Challenger Sensory Champion &#123;Brettcomputer&#125; 8,0/13  0 0 0 1 = · 0 1 1 = 1 1 1 1   39,00
07&#58; Scisys MKV &#123;Brettcomputer&#125;                                 7,5/13  0 = 1 1 = 1 · 0 1 = 0 0 1 1   46,75
08&#58; Elektronika IM05 &#123;Brettcomputer&#125;                           6,5/13  0 0 = 0 0 0 1 · 0 1 1 1 1 1   29,00
09&#58; Mephisto ESB II &#123;Brettcomputer&#125;                            5,5/13  1 0 0 0 0 0 0 1 · 0 1 1 1 =   28,00
10&#58; Milton Bradley &#123;Brettcomputer&#125;                             5,0/13  0 0 0 0 = = = 0 1 · 0 1 1 =   23,75
11&#58; Yeno 301 XL &#123;Brettcomputer&#125;                                4,5/13  0 0 = 0 0 0 1 0 0 1 · = = 1   21,75
12&#58; Mephisto Junior &#40;Tasten&#41; &#123;Brettcomputer&#125;                   3,5/13  0 0 0 0 0 0 1 0 0 0 = · 1 1   13,75
13&#58; Mephisto Supermini &#123;Brettcomputer&#125;                         2,0/13  0 0 0 1 = 0 0 0 0 0 = 0 · 0   14,75
14&#58; Novag Opal II &#123;Brettcomputer&#125;                              2,0/13  0 0 0 0 0 0 0 0 = = 0 0 1 ·    7,25

91 Partien von 182 gespielt

SEE here for the games:

https://www.schachcomputer.info/forum/s ... or&page=14

We could try to let the dedicated units run vs. The pc engine by reducing the hardware of the pc to the 6502/2 MHz hardware level.

So we have 3 stages:

1. the dedicated chess computers,
2. Parker brothers engine
3. Later Cyrus 68000 engine developed by levy/Taylor.

Stage 3 was IMO without David Broughton.

He had a new job for the television company’s producing video effects for the commercial television channels.

But levy and Taylor continued on the motorola 68000 hardware that was state of the art at that time.

I am not sure if the software was ever sold in another way then CXG Sphinx dedicated chess computer,.

Dostojewsky · Post by **Dostojewsky** » Tue Jun 16, 2020 12:03 pm

For the record: A)All my games were played under Qemu with a FX8320 at 3,7 Ghz and B) The whole last game can be repeated with Philidor B1502.

My last question is who has conncetion to Mark Taylor?

i think he is the man who knows the answer to our question.

Maybe he has a pc-software of Mark V.

mclane · Post by **mclane** » Fri Jun 26, 2020 6:42 pm

The major question now... how strong is Philidor on fast pc ?

Volodymyr · Post by **Volodymyr** » Fri Jun 26, 2020 7:59 pm

Hi mclane
I did a little test. Virtualbox is about 10 times faster than Qemu on my i5-3470.
Most older programs run on this emulator. Psion, Cyrus, Philidor work.
But I don’t know how stable. For example, on the also fast Virtual PC and VMware Workstation, I couldn’t start these programs.
I recommend Virtualbox for tests, now it is a very good and fast emulator.

mclane · Post by **mclane** » Fri Jun 26, 2020 9:32 pm

Hi virtual Box ??

Is this for Windows ??

An app ?

Are you really living in ucraine ?
Where in ucraine.

Isn’t ucraine a splittend country because putin did a war there ?

Volodymyr · Post by **Volodymyr** » Fri Jun 26, 2020 10:10 pm

Yes, I live in the west of Ukraine. I am closer to Poland and Slovakia than to Russia.
You can run different OS DOS, Windows, Linux, Android, Warp.
But I have not tried everything.
I only have DOS 6.22 and Windows 98 installed.
https://www.virtualbox.org/wiki/Downloads

TracySMiller · Post by **TracySMiller** » Tue Jun 30, 2020 10:58 pm

I tried Volodymyr's suggestion of running VirtualBox to get a big speed gain. I was curious as to how that would play out over the board, so I thought I'd play Philidor against Mephisto Master chess again, only this time running Philidor on Virtual Box instead of DOSBox or QEMU. You may recall, from my limited testing Philidor seemed a little stronger than the Novag Constellation (1847), but considerably weaker than the Mephisto Master Chess, losing both games to the 2161-rated Master. We concluded perhaps running at that speed it was maybe in the 1900-2000 range.

Well, check out this first game of Philidor vs. Mephisto Master Chess. I know one game is not much, but it is clearly on a different level than the slower Philidor. Game 1 goes to Philidor, and I played it to mate to make sure it could handle to K+R vs. K ending.

Here's the complete game: https://lichess.org/g6o2o3Gs

mclane · Post by **mclane** » Tue Jun 30, 2020 11:48 pm

How to get this virtual machine running ?

TracySMiller · Post by **TracySMiller** » Wed Jul 01, 2020 12:12 am

mclane wrote:How to get this virtual machine running ?

You have to install VirtualBox, then install DOS 6.22 on Virtual Box. Then you have to use a program (like WinImage) to convert the DOS chess game into an image, which you then load into VirtualBox. Sounds involved, but I was able to get it running fairly quickly, probably less than an hour, and I'm not that great with computers.

TracySMiller · Post by **TracySMiller** » Wed Jul 01, 2020 10:21 pm

Game 2 featured an exciting tactical battle in which Philidor won the exchange, but Mephisto Master Chess had an advanced passed pawn as compensation. The game reached a bit of a standoff in which White (Mephisto Master) was sitting on Mate in 1, but Black (Philidor) had to try to get perpetual checks. Here's the critical position:

[fen]1Q3rk1/P4ppp/3B4/8/4p1P1/1P4K1/P6P/5q2 w - - 2 36[/fen]

According to Stockfish, 36...Qf3+ will do the trick, and shows a dead draw. Unfortunately, Philidor picked 36...Qg1+, and the evaluation jumped to almost +8.00 for White.

Here's the entire game: https://lichess.org/TkNtY5aB#0

Thus, Mephisto Master Chess wins game two, tying the match with Philidor at 1.0-1.0. Any guesses as to the outcome of a 10-game match?

mclane · Post by **mclane** » Wed Jul 01, 2020 11:29 pm

Philidor is B strategy. This means it is NOT calculating all branches.

Post by **Fernando** » Thu Jul 02, 2020 1:37 am

mclane wrote:Philidor is B strategy. This means it is NOT calculating all branches.

Well, that does not speak very favorably of B strategy programs. I am not interested in a program that does not see a simple thing because he is engaged in high ruminations.

BTW, I will use that explanation each time I could be beaten: "well, sir," I will say, "you beat me just because i was thinking in so much high matters than a vulgar checkmate..."

strategic regards
F

Post by **Fernando** » Thu Jul 02, 2020 5:26 am

TracySMiller wrote:
mclane wrote:How to get this virtual machine running ?
You have to install VirtualBox, then install DOS 6.22 on Virtual Box. Then you have to use a program (like WinImage) to convert the DOS chess game into an image, which you then load into VirtualBox. Sounds involved, but I was able to get it running fairly quickly, probably less than an hour, and I'm not that great with computers.

Tracy, it would be great if you can tell me step by step the full process. I have never used those virtual box things, etc. Just D-fend, which is easy but not that efficient.

Old man in his last days asking help regards
Fern

HIARCS Chess Forums

Philidor/Parker Chess for PC (1983)

Re: This is not a sophisticated chessprogram

Re: This is not a sophisticated chessprogram

Mephisto Master Chess wins game 2 against Philidor