Equinox 3.20 (free to all)

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Harvey Williamson, bob

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
ernest
Posts: 1842
Joined: Wed Mar 08, 2006 7:30 pm

Re: Equinox 3.20 (free to all)

Post by ernest » Wed Oct 01, 2014 1:20 am

Sedat Canbaz wrote:And here are the latest sim test results by Equinox 3.20
Please correct me if I am wrong, but didn't the author of the sim test say that sim test results less than 60 don't prove anything ?

majortom
Posts: 669
Joined: Mon Nov 04, 2013 9:19 pm
Contact:

Re: The next without readme (? for all)

Post by majortom » Wed Oct 01, 2014 12:26 pm

Frank Quisinsky wrote:The next unprofessional download for the community ...
Sedat Canbaz wrote:And here are the latest sim test results by Equinox 3.20

It seems Equinox is Rybka derivative too!

sim version 3
------ Equinox 3.20 x64mp (time: 100 ms scale: 1.0) ------
59.36 RobboLito 0.085g3 w32 (time: 100 ms scale: 1.0)
58.27 Fire 3.0 x64 (time: 100 ms scale: 1.0)
57.65 Rybka 3 (time: 100 ms scale: 1.0)
56.85 Houdini 4 x64 (time: 100 ms scale: 1.0)
54.31 Critter 0.90 64-bit SSE4 (time: 100 ms scale: 1.0)
Image
Frank Quisinsky wrote:

Code: Select all

Version: 1.09
Engine : Critter 0.90 SSE4 x64
Result : No predecessor / successor tested!
Time   : September 13th - September 17th, 2014
The next unprofessional test for the community ...

Sedat Canbaz
Posts: 3018
Joined: Thu Mar 09, 2006 10:58 am
Location: Antalya/Turkey
Contact:

Re: Equinox 3.20 (free to all)

Post by Sedat Canbaz » Wed Oct 01, 2014 1:13 pm

ernest wrote:
Sedat Canbaz wrote:And here are the latest sim test results by Equinox 3.20
Please correct me if I am wrong, but didn't the author of the sim test say that sim test results less than 60 don't prove anything ?
Yes..with pleasure I will correct that you are wrong ))

Twinfish (100% clone of Stockfish) is clear proof that any engine can be released below than 60 % similarity

Plus I use time: 100 ms scale: 1.0, I mean the similarity is lower than Pia's calculation method

Otherwise...(with Pia's method) Naum will pass 60 % close to Rybka

To be honest,
I think Don's tool is just great...and I am very happy of using it !!
And it's disaster for those...who are copiers ! it detects almost perfectly !!!

It's true that some guys are managed to hack it...
But however, overall...its working just fine...it detects many clones etc...

Even Don's tool is better in detection than many chess engine experts !

For example (by the experts),
Ippolit can not be proven 100% as clone of a Rybka engine
or
Naum is derivative of Fruit

But in reality,
Naum is Rybka derivative (not Fruit derivative, experts should change their minds)))!!

Note also that the sim test tool helps us a lot !!!
As we see.... Ippolit is approx. 60% similar to Rybka 3
Even the first IPPO become on the scene after Rybka 3 release
Even the first IPPO Elo points were near to Rybka 3
I can give you more facts...just let me know...

In other words,
We should be very naive to not see that...

Btw, there is a good saying about this issue:
“It is well for the heart to be naive and for the mind not to be”


sim version 3
------ TwinFish 0.07 (time: 100 ms scale: 1.0) ------
52.74 Stockfish 140614 64 SSE4.2 (time: 100 ms scale: 1.0)
49.70 Stockfish 2.1 JA 64bit (time: 50 ms scale: 1.0)
49.65 Stockfish 2.1 JA 64bit (time: 100 ms scale: 1.0)
48.45 Protector 1.7.0 (time: 100 ms scale: 1.0)
48.14 Gull 3 x64 (time: 100 ms scale: 1.0)
47.96 Equinox 3.20 x64mp (time: 100 ms scale: 1.0)
47.95 Stockfish 1.7 JA 64bit (time: 100 ms scale: 1.0)
47.84 Houdini 4 x64 (time: 100 ms scale: 1.0)
47.65 Stockfish 1.7.1 JA (time: 100 ms scale: 1.0)
47.61 Critter 0.90 64-bit SSE4 (time: 100 ms scale: 1.0)
47.26 Fire 3.0 x64 (time: 100 ms scale: 1.0)
46.82 RobboLito 0.085g3 w32 (time: 100 ms scale: 1.0)
46.78 Rybka 3 (time: 100 ms scale: 1.0)
46.65 Stockfish 1.5 JA 64bit (time: 100 ms scale: 1.0)
46.39 Naum 4.6 (time: 100 ms scale: 1.0)
46.33 Komodo64 2.03 DC (time: 100 ms scale: 1.0)
45.57 Murka 3 x64 UCI (time: 100 ms scale: 1.0)
44.72 Octochess revision 5190 (time: 100 ms scale: 1.0)
44.56 MinkoChess 1.3 x64 (time: 100 ms scale: 1.0)
43.29 Strelka 2.0 B (time: 50 ms scale: 1.0)
41.59 Rodent 1.4 (build 2) (time: 100 ms scale: 1.0)
41.53 Daydreamer 1.75 JA (time: 100 ms scale: 1.0)
41.44 Bobcat 3.25 (time: 100 ms scale: 1.0)
41.19 cheng4 0.36a (time: 100 ms scale: 1.0)
40.17 Cyrano 0.6b17 (time: 100 ms scale: 1.0)
39.85 Fruit 2.1 (time: 100 ms scale: 1.0)
39.43 Vajolet 2.48 (time: 100 ms scale: 1.0)
34.97 Igorrit 0.086v8_x64 (time: 100 ms scale: 1.0)
34.44 Booot 5.2.0(64) (time: 100 ms scale: 1.0)
24.81 Arasan 17.2 (time: 100 ms scale: 1.0)

Sedat Canbaz
Posts: 3018
Joined: Thu Mar 09, 2006 10:58 am
Location: Antalya/Turkey
Contact:

Re: Equinox 3.20 (free to all)

Post by Sedat Canbaz » Wed Oct 01, 2014 1:50 pm

One thing more,
Even one of the latest Stockfish versions is below than 60% similarity
I mean comparing with the older Stockfish ones
That means any engine which is 55% and above should be under lupe ))!!

And my recommendation is that for all engine testers,
Please use my current rule:

•Each participant (chess program) must be an original work
•Clones, derivatives, private engine versions are not allowed
•Derivatives, clones chess engines will be allowed only:
-In case of approx. +100 Elo stronger than based original engine

And any engine with 55% similarity and above should be counted in that group

Note also that,
In this way...We will give a BIG chance only for talented engine programmers
Robert Houdart, Vadim Demichev....managed to do that...why not you ?


sim version 3
------ Stockfish 140614 64 SSE4.2 (time: 100 ms scale: 1.0) ------
55.96 Stockfish 2.1 JA 64bit (time: 100 ms scale: 1.0)
55.84 Stockfish 2.1 JA 64bit (time: 50 ms scale: 1.0)
53.16 Stockfish 1.7 JA 64bit (time: 100 ms scale: 1.0)
52.96 Stockfish 1.7.1 JA (time: 100 ms scale: 1.0)
52.74 TwinFish 0.07 (time: 100 ms scale: 1.0)
52.40 Protector 1.7.0 (time: 100 ms scale: 1.0)
52.16 Houdini 4 x64 (time: 100 ms scale: 1.0)
51.65 Gull 3 x64 (time: 100 ms scale: 1.0)
51.09 Fire 3.0 x64 (time: 100 ms scale: 1.0)
50.98 Stockfish 1.5 JA 64bit (time: 100 ms scale: 1.0)
50.64 Equinox 3.20 x64mp (time: 100 ms scale: 1.0)
50.46 Rybka 3 (time: 100 ms scale: 1.0)
50.27 Critter 0.90 64-bit SSE4 (time: 100 ms scale: 1.0)
50.23 Naum 4.6 (time: 100 ms scale: 1.0)
49.47 Murka 3 x64 UCI (time: 100 ms scale: 1.0)
49.16 RobboLito 0.085g3 w32 (time: 100 ms scale: 1.0)
48.56 Komodo64 2.03 DC (time: 100 ms scale: 1.0)
47.43 MinkoChess 1.3 x64 (time: 100 ms scale: 1.0)
46.01 Octochess revision 5190 (time: 100 ms scale: 1.0)
45.16 Rodent 1.4 (build 2) (time: 100 ms scale: 1.0)
44.94 Strelka 2.0 B (time: 50 ms scale: 1.0)
44.53 cheng4 0.36a (time: 100 ms scale: 1.0)
43.06 Bobcat 3.25 (time: 100 ms scale: 1.0)
42.62 Daydreamer 1.75 JA (time: 100 ms scale: 1.0)
42.43 Fruit 2.1 (time: 100 ms scale: 1.0)
42.17 Cyrano 0.6b17 (time: 100 ms scale: 1.0)
40.76 Vajolet 2.48 (time: 100 ms scale: 1.0)
35.64 Igorrit 0.086v8_x64 (time: 100 ms scale: 1.0)
35.26 Booot 5.2.0(64) (time: 100 ms scale: 1.0)
25.04 Arasan 17.2 (time: 100 ms scale: 1.0)

Sedat Canbaz
Posts: 3018
Joined: Thu Mar 09, 2006 10:58 am
Location: Antalya/Turkey
Contact:

Re: Equinox 3.20 (free to all)

Post by Sedat Canbaz » Wed Oct 01, 2014 2:48 pm

Frank Quisinsky wrote:Hi Sedat,

the question is since which version from Equinox?
Programmer is reading, perhaps he an give us more information?
Possible that he do it?

Same problem with Critter ... but here I find a good way for FCT1.

That is really bad! A nice development up to version x ...
After that suddenly big Elo jumpings and to see is allways the same if you compare it with other derivative engines.

But not very important ...
One derivative engine more or less ... the community like that!
Many people can't have enough derivative engines or clones.

Very difficult to see such things for each of the engines.

Good known rating lists are in a short time full of clones and derivative engines. With the result that the measuremensts are unexactly, not important how many games played.

That was the main reason I lost my interest in testing engines in the past. It's a hard work to try to see more with games and stats only. But here directly all is clear in most of cases!

After all ...
We have so many own engine developments and enough material for testing.

Best
Frank
Hello Frank,

Good points...

A few notes by me,
Some engine programmers are preferring the easiest way, by picking or coping the ideas of Rybka (Ippo)

I think that those chess engines which are 55 % similarity and above....
They have not much own ideas...I mean of course at near Elo points over based original engine

Yes...what a pity we have a lot of naive people...
Sorry to say that but they are in my eyes as amateurs !!

As you know,
After checking their used opening lines...they are losing their valuable time

They don't understand from openings, it does not matter for them this line is weak or strong
Plus they don't care too much...whether this engine is derivative or clone..
For them it's important the beauty....the rest is not important ))!!

Btw,
I am very happy that your rules are very strict (near to SCCT)
We should protect the originality and of course the talented programers too!!
Otherwise..our tournaments will be based mainly on IPPO (Rybka) derivatives !

One thing more,
You should kill 2 engines in your FCT, I assume you know which ones ;)

Best,
Sedat

Sedat Canbaz
Posts: 3018
Joined: Thu Mar 09, 2006 10:58 am
Location: Antalya/Turkey
Contact:

Re: Equinox 3.20 (free to all)

Post by Sedat Canbaz » Wed Oct 01, 2014 5:13 pm

Btw,
BlackMamba 2.0 seems to be another Rybka derivative too !!

Yes...Everybody is free to test any engine, which he likes...

But however without to not mention this I can't,
In case of testings those Rybka derivatives, be sure in that:
- You are testing Rybka with different settings (exception Houdini, Gull...) !!!

Btw, now I remembered Mindbreaker, who created strong settings for Rybka
But at least, he did not change the real Rybka original name !!!

Note that too,
Just a few years ago I tested many Houdini settings, which were stronger than Houdini default
And all these settings were with their original Houdini names too !!!

Note also I also created CM Gladiator settings too, which was approx. 30-50 Elo points stronger than ChessMaster default
And I did not change its original name too, because I love the true !!!

sim version 3
------ BlackMamba 2.0 x64 (time: 100 ms scale: 1.0) ------
63.85 RobboLito 0.085g3 w32 (time: 100 ms scale: 1.0)
60.32 Fire 3.0 x64 (time: 100 ms scale: 1.0)
58.45 Rybka 3 (time: 100 ms scale: 1.0)
57.70 Houdini 4 x64 (time: 100 ms scale: 1.0)
57.06 Equinox 3.20 x64mp (time: 100 ms scale: 1.0)
56.32 Critter 0.90 64-bit SSE4 (time: 100 ms scale: 1.0)
53.91 Naum 4.6 (time: 100 ms scale: 1.0)
52.94 Murka 3 x64 UCI (time: 100 ms scale: 1.0)
52.02 Komodo64 2.03 DC (time: 100 ms scale: 1.0)
51.21 Senpai 1.0 (time: 100 ms scale: 1.0)
51.09 Stockfish 2.1 JA 64bit (time: 50 ms scale: 1.0)
51.06 Stockfish 2.1 JA 64bit (time: 100 ms scale: 1.0)
50.81 Protector 1.7.0 (time: 100 ms scale: 1.0)
50.34 Stockfish 1.7.1 JA (time: 100 ms scale: 1.0)
49.96 Strelka 2.0 B (time: 50 ms scale: 1.0)
49.96 Stockfish 1.5 JA 64bit (time: 100 ms scale: 1.0)
49.90 Stockfish 140614 64 SSE4.2 (time: 100 ms scale: 1.0)
49.84 Stockfish 1.7 JA 64bit (time: 100 ms scale: 1.0)
49.19 Gull 3 x64 (time: 100 ms scale: 1.0)
48.75 MinkoChess 1.3 x64 (time: 100 ms scale: 1.0)
47.24 Bobcat 3.25 (time: 100 ms scale: 1.0)
46.89 TwinFish 0.07 (time: 100 ms scale: 1.0)
46.75 Daydreamer 1.75 JA (time: 100 ms scale: 1.0)
46.14 Fruit 2.1 (time: 100 ms scale: 1.0)
45.16 Octochess revision 5190 (time: 100 ms scale: 1.0)
44.91 Cyrano 0.6b17 (time: 100 ms scale: 1.0)
44.55 cheng4 0.36a (time: 100 ms scale: 1.0)
43.98 Igorrit 0.086v8_x64 (time: 100 ms scale: 1.0)
43.69 Rodent 1.4 (build 2) (time: 100 ms scale: 1.0)
43.31 Vajolet 2.48 (time: 100 ms scale: 1.0)
39.83 Booot 5.2.0(64) (time: 100 ms scale: 1.0)
26.60 Arasan 17.2 (time: 100 ms scale: 1.0)

Sedat Canbaz
Posts: 3018
Joined: Thu Mar 09, 2006 10:58 am
Location: Antalya/Turkey
Contact:

Re: Equinox 3.20 (free to all)

Post by Sedat Canbaz » Wed Oct 01, 2014 5:26 pm

Just I'd like to add,
Fire 3.0 seems to be another Rybka 3 derivative too !!!

And so far ...how many Rybkanians we noticed ??))

sim version 3
------ Fire 3.0 x64 (time: 100 ms scale: 1.0) ------
63.49 RobboLito 0.085g3 w32 (time: 100 ms scale: 1.0)
60.49 Houdini 4 x64 (time: 100 ms scale: 1.0)
60.32 BlackMamba 2.0 x64 (time: 100 ms scale: 1.0)
58.76 Rybka 3 (time: 100 ms scale: 1.0)
58.27 Equinox 3.20 x64mp (time: 100 ms scale: 1.0)
56.92 Critter 0.90 64-bit SSE4 (time: 100 ms scale: 1.0)
54.07 Murka 3 x64 UCI (time: 100 ms scale: 1.0)
53.17 Stockfish 2.1 JA 64bit (time: 100 ms scale: 1.0)
53.08 Naum 4.6 (time: 100 ms scale: 1.0)
52.10 Stockfish 2.1 JA 64bit (time: 50 ms scale: 1.0)
51.94 Gull 3 x64 (time: 100 ms scale: 1.0)
51.78 Stockfish 1.7 JA 64bit (time: 100 ms scale: 1.0)
51.60 Senpai 1.0 (time: 100 ms scale: 1.0)
51.44 Stockfish 1.7.1 JA (time: 100 ms scale: 1.0)
51.09 Stockfish 140614 64 SSE4.2 (time: 100 ms scale: 1.0)
50.87 Komodo64 2.03 DC (time: 100 ms scale: 1.0)
50.23 Protector 1.7.0 (time: 100 ms scale: 1.0)
49.72 Stockfish 1.5 JA 64bit (time: 100 ms scale: 1.0)
47.74 Strelka 2.0 B (time: 50 ms scale: 1.0)
47.26 TwinFish 0.07 (time: 100 ms scale: 1.0)
46.98 MinkoChess 1.3 x64 (time: 100 ms scale: 1.0)
45.96 Bobcat 3.25 (time: 100 ms scale: 1.0)
45.06 Daydreamer 1.75 JA (time: 100 ms scale: 1.0)
45.02 Rodent 1.4 (build 2) (time: 100 ms scale: 1.0)
44.49 Octochess revision 5190 (time: 100 ms scale: 1.0)
44.03 Fruit 2.1 (time: 100 ms scale: 1.0)
44.02 cheng4 0.36a (time: 100 ms scale: 1.0)
43.23 Cyrano 0.6b17 (time: 100 ms scale: 1.0)
41.94 Vajolet 2.48 (time: 100 ms scale: 1.0)
41.68 Igorrit 0.086v8_x64 (time: 100 ms scale: 1.0)
38.15 Booot 5.2.0(64) (time: 100 ms scale: 1.0)
25.95 Arasan 17.2 (time: 100 ms scale: 1.0)
Last edited by Sedat Canbaz on Wed Oct 01, 2014 5:29 pm, edited 1 time in total.

Frank Quisinsky
Posts: 4844
Joined: Wed Nov 18, 2009 6:16 pm
Location: Trier, Germany
Contact:

Re: NEED YOU, 10.500 games tourney with Arasan, ExChess

Post by Frank Quisinsky » Wed Oct 01, 2014 5:29 pm

Hi Sedat,

Need your help ...

After the still running Protector 1.7.0, Stockfish dev. C-17 test-run & Junior 13.3 test run (in case of Junior I will see the third version) I will start my second league with 21 engines.

10.500 games are to play ...
Will be start in ...
2, 4, 4 days for the test-runs I have in ToDo ...
in around 10-12 days!

place 19 of my first league
Gaviota 1.0

place 20
Deuterium

place 21
Tornado

Tested engines: Fizbo, Crafty, Nirvanachess, Rhetoric
= 7 places

and 14 other engines with more as 2.600 Elo.
Thinking on: Exchess, Scorpio, Arasan and 11 others.

The 11 others are the problem!
Only available engines I will used, no clones, derivative engines.
Furthermore, engines still in development can be participant.

WB or UCI not important.

Perhaps you can help a bit to find out 11 engines for this 45 days still running very big tourney.

Today in the evening I will working to search on CEGT, Graham's tourneys for the participants I can used.

Best
Frank

In the time this tourney is still running, I have a lot time to work on my second "new" book version :-)
I like computer chess!

Henk
Posts: 5651
Joined: Mon May 27, 2013 8:31 am

Re: Equinox 3.20 (free to all)

Post by Henk » Wed Oct 01, 2014 5:41 pm

Sedat Canbaz wrote:Just I'd like to add,
Fire 3.0 seems to be another Rybka 3 derivative too !!!

And so far ...how many Rybkanians we noticed ??))

sim version 3
------ Fire 3.0 x64 (time: 100 ms scale: 1.0) ------
63.49 RobboLito 0.085g3 w32 (time: 100 ms scale: 1.0)
60.49 Houdini 4 x64 (time: 100 ms scale: 1.0)
60.32 BlackMamba 2.0 x64 (time: 100 ms scale: 1.0)
58.76 Rybka 3 (time: 100 ms scale: 1.0)
58.27 Equinox 3.20 x64mp (time: 100 ms scale: 1.0)
56.92 Critter 0.90 64-bit SSE4 (time: 100 ms scale: 1.0)
54.07 Murka 3 x64 UCI (time: 100 ms scale: 1.0)
53.17 Stockfish 2.1 JA 64bit (time: 100 ms scale: 1.0)
53.08 Naum 4.6 (time: 100 ms scale: 1.0)
52.10 Stockfish 2.1 JA 64bit (time: 50 ms scale: 1.0)
51.94 Gull 3 x64 (time: 100 ms scale: 1.0)
51.78 Stockfish 1.7 JA 64bit (time: 100 ms scale: 1.0)
51.60 Senpai 1.0 (time: 100 ms scale: 1.0)
51.44 Stockfish 1.7.1 JA (time: 100 ms scale: 1.0)
51.09 Stockfish 140614 64 SSE4.2 (time: 100 ms scale: 1.0)
50.87 Komodo64 2.03 DC (time: 100 ms scale: 1.0)
50.23 Protector 1.7.0 (time: 100 ms scale: 1.0)
49.72 Stockfish 1.5 JA 64bit (time: 100 ms scale: 1.0)
47.74 Strelka 2.0 B (time: 50 ms scale: 1.0)
47.26 TwinFish 0.07 (time: 100 ms scale: 1.0)
46.98 MinkoChess 1.3 x64 (time: 100 ms scale: 1.0)
45.96 Bobcat 3.25 (time: 100 ms scale: 1.0)
45.06 Daydreamer 1.75 JA (time: 100 ms scale: 1.0)
45.02 Rodent 1.4 (build 2) (time: 100 ms scale: 1.0)
44.49 Octochess revision 5190 (time: 100 ms scale: 1.0)
44.03 Fruit 2.1 (time: 100 ms scale: 1.0)
44.02 cheng4 0.36a (time: 100 ms scale: 1.0)
43.23 Cyrano 0.6b17 (time: 100 ms scale: 1.0)
41.94 Vajolet 2.48 (time: 100 ms scale: 1.0)
41.68 Igorrit 0.086v8_x64 (time: 100 ms scale: 1.0)
38.15 Booot 5.2.0(64) (time: 100 ms scale: 1.0)
25.95 Arasan 17.2 (time: 100 ms scale: 1.0)
Where can I download the e-book "How to clone Rybka 3.0 in 24 hours"

Sedat Canbaz
Posts: 3018
Joined: Thu Mar 09, 2006 10:58 am
Location: Antalya/Turkey
Contact:

Re: NEED YOU, 10.500 games tourney with Arasan, ExChess

Post by Sedat Canbaz » Wed Oct 01, 2014 5:55 pm

Frank Quisinsky wrote: Perhaps you can help a bit to find out 11 engines for this 45 days still running very big tourney.
Hello Frank,

Yes, sure...with pleasure I can try to help you my friend !)

So...I can recommend you to use the below top participants:

.
Image



Best,
Sedat

Post Reply