In support of the IvanHoe authors

[velmarin]

Similarity tool "Author Don Daley" executed on engines named.

Image from the point of view of Ivanhoe 46h (CCRL official, name internal 46b)
Robbolito is the most similar.
Clearly Bouquet away.





Download, Sim3, similarity.data and engines:
http://www.mediafire.com/?h9d3m69o7450czh

Greetings.

[pohl4711]

Mars has not yet executable for 64 no_popcnt..

Thats not correct. The download contains three compiles: 64bit, 64bit popcount and 64bit avx.
And today Trap launched a new try of a 32bit-compile. I cant verify this one, because I dont have 32bit-Windows.

Best - Stefan

Final results of Mars 1 avx and Gull 2.1 Trap avx tomorrow for the LS-ratinglist, if all works correct. After that, the LS-testrun of Gull 2.2 will follow.

[Dr.Wael Deeb]

Within what Ippolit family,
leaving out Houdini, Critter, Equinox, Blackmamba.

Gull, but this is a special case with a tremendous merit.

In the rest in my opinion, what remains is the following:

IvanHoe45, born and died, no work ...
Firenzina equal ....

Robbolite, abandoned, only to growl ...
Fire, equal ...

Mars has not yet executable for 64 no_popcnt or 32-bit,
Fire quick compilation, I have doubts about his future ...

PanChess, Izak hard work, with much enthusiasm,
live project, ...

Bouquet, equal to Panchess, daily work, daily tests, few means to make progress, hopefully, but tired of insults and other things.
Soon I will leave and go into other things, worth might not keep fighting.

After I post similarity test...

Please keep up the good work Jose....

I am a huge fan of Bouquet 1.5 x64 MP and after 3200 games played using 20 minutes + 20 seconds increment it achieved a 3174 Elo in my private rating list......

Just to have a more clear picture,BlackMamba 1.4 x64 has 2969 after the same number of games and against the same opponents
Cheers,
Dr.D

P.S.Some beautiful day during a gorgeous spring I'll publish my rating list

[velmarin]

using 20 minutes + 20 seconds increment

Thank you, Dr..
I did not want to be discourteous to other authors ..
Or impolite with testers (as important) (different forums)
relases output of "Stockfish" almost daily,
tests are being carried LittleBlitter almost always, all super fast, I argue that many engines fail to complete a full search, sending many evals and search incorrect.

So I like CCRL, (CEGT keep forgetting certain engines).
And it is much appreciated what you write, is the minimum for these engines give their full potential.

In the last TCEC thinking engines saw a move make tens of minutes.
In super fast tests for a move in TCEC,
In littlebliter 100 games can happen.

[Dr.Wael Deeb]

using 20 minutes + 20 seconds increment

Thank you, Dr..
I did not want to be discourteous to other authors ..
Or impolite with testers (as important) (different forums)
relases output of "Stockfish" almost daily,
tests are being carried LittleBlitter almost always, all super fast, I argue that many engines fail to complete a full search, sending many evals and search incorrect.

So I like CCRL, (CEGT keep forgetting certain engines).
And it is much appreciated what you write, is the minimum for these engines give their full potential.

In the last TCEC thinking engines saw a move make tens of minutes.
In super fast tests for a move in TCEC,
In littlebliter 100 games can happen.

You're welcome my friend

BTW,here are my conditions applied for my private rating list:

_Hardware:
Intel Core i7-3970X Extreme Edition Sandy Bridge-E 3.5GHz (4.0GHz Turbo) LGA 2011
Hashtables=1024 Mb for each engine.

_Software: Windows 7 Ultimate x64
GUI:Deep Fritz 13.
Nalimov EGTB:Endgame Turbo 3

_Conditions:
Time controls: 20 minutes per game + 20 seconds increment.

Cheers,
Dr.D

P.S.You look great Jose in your avatar

[Graham Banks]

Mars has not yet executable for 64 no_popcnt..

Thats not correct. The download contains three compiles: 64bit, 64bit popcount and 64bit avx.
And today Trap launched a new try of a 32bit-compile. I cant verify this one, because I dont have 32bit-Windows.

Best - Stefan

Final results of Mars 1 avx and Gull 2.1 Trap avx tomorrow for the LS-ratinglist, if all works correct. After that, the LS-testrun of Gull 2.2 will follow.

Does Mars break the GPL license?
Just wondered as I saw this raised elsewhere.

[kranium]

tests are being carried LittleBlitter almost always, all super fast, I argue that many engines fail to complete a full search, sending many evals and search incorrect.

I disagree...
fast testing is proving accurate (and has for years, IMO)...
especially with the multi-core CPU speeds being achieved today

this is an excerpt from Stefan's LightSpeed rating list site:
http://ls-ratinglist.beepworld.de/ls-to ... nament.htm

Thinking-time: 45''+500ms Fischerbonus (= 85-90 seconds per game/engine). That means a average thinking time of 2-3 seconds in the middlegame down to 0.5 seconds (but never below!) in the endgame. Thats fast, but not ultrafast. Average search depths of some engines playing with LS-conditions: Houdini 18 plies, Stockfish 22 plies, Komodo 17 plies, Ivanhoe 17.5 plies. I believe, that should be enough for playing good chess, dont you?

compare the LightSpeed results for the top engines to those of CCRL 40/40 and CEGT 40/120

I think you'll find that they match closely

[Izak Pretorius]

Does Mars break the GPL license?
Just wondered as I saw this raised elsewhere.

hi Graham

Yes this is correct.Norman Schmidt (Kranium) has posted this on immortal forum.

sorry to disappoint guys,
unfortunately, it's neither "new Ippolit",
nor is it a new engine

it appears to simply be Fire 2.2 (or perhaps Firenzina) recompiled with many UCI options removed
...for ex: type "default" or "random" at the console prompt and the engine will produce Fire's fire.cfg
Last edited by kranium; 11th August 2013 at 13:26.

So Mars is a modified re-compile of Firenzina which is GPL
It is just strange to me that now Norman wants this engine to participate
in TCEC even knowing and even being the person that pointed out that this re-compile modification breaks the GPL license agreement.

Very strange indeed.

Best Regards
Izak (Peterpan)

[kranium]

It is just strange to me that now Norman wants this engine to participate
in TCEC even knowing and even being the person that pointed out that this re-compile modification breaks the GPL license agreement.

Very strange indeed.

Best Regards
Izak (Peterpan)

what on earth are you talking about?
that's despicable...why would you post such trash?

i've never suggested any such thing...
but i have gone on record here concerning my TCEC preference for IvanHoe in Martin's topic:

http://www.talkchess.com/forum/viewtopi ... =&start=20

and in fact, this very topic itself split from the one above in order for me to express my support for IvanHoe even further

[Izak Pretorius]

It is just strange to me that now Norman wants this engine to participate
in TCEC even knowing and even being the person that pointed out that this re-compile modification breaks the GPL license agreement.

Very strange indeed.

Best Regards
Izak (Peterpan)

what on earth are you talking about?
i've never suggested any such thing...why are you posting such trash?

but i have gone on record here concerning my TCEC preference for IvanHoe in Martin's topic:
http://www.talkchess.com/forum/viewtopi ... =&start=20

and in fact, this very topic itself split from the one above in order for me to express my support for IvanHoe even further

Perhaps a misunderstanding.I was under the impression this is what you wanted.Perhaps i was thinking about your post regarding the modification Ivanhoe 50kQ,which is clearly not based on the pure Ivanhoe source,and you were promoting it as being pure.

My apologies then,i am then glad we agree not to use GPL breakers in tournaments.