Houdini now included in CEGT rating List

[Prima]

It looks like Houdini 1.03a and 1.5 have been tested. Houdini 1.03a is in the CEGT list though. Here's a Talkchess thread:
http://www.talkchess.com/forum/viewtopi ... 18386e9caa

[kingliveson]

After this interesting turn of event, Rybka quickly modified its forum policy from:

For those of you, who want to check which engines are clones and which aren't, I recommend looking at the professional rating lists like CEGT (http://www.husvankempen.de/nunn/) or CCRL (http://computerchess.org.uk/). They check new engines, don't test clones and are independent. The lists are updated regularly, so you can expect them to be up to date. Also the engine's appearance (description on the website, engine name and so on) indicates if it's a serious engine or just a clone.

to:

For those of you, who want to check which engines are clones and which aren't, I recommend looking at the professional rating lists like CCRL (http://computerchess.org.uk/). They check new engines, don't test clones and are independent. The lists are updated regularly, so you can expect them to be up to date. Also the engine's appearance (description on the website, engine name and so on) indicates if it's a serious engine or just a clone.

http://rybkaforum.net/cgi-bin/rybkaforu ... ?tid=14541

I think this could compromise CCRL status as an independent ratings list -- something that has been suggested. Especially when everything is put in context including one of the testers attitude on the forums. But again, it's a private group and can do as it wishes.

[thorstenczub]

hehe

the best would be to have a rating list for each commercial
engine.

lets say: ccrl for rybka fans.
inwoba for shredder fans.
cegt for houdini fans.
and and and.
which rating agency is willing to bring hiarcs on top of his list. i am sure
harvey would be pleased to announce them in the hiarcs forum.

[Prima]

After this interesting turn of event, Rybka quickly modified its forum policy from:

For those of you, who want to check which engines are clones and which aren't, I recommend looking at the professional rating lists like CEGT (http://www.husvankempen.de/nunn/) or CCRL (http://computerchess.org.uk/). They check new engines, don't test clones and are independent. The lists are updated regularly, so you can expect them to be up to date. Also the engine's appearance (description on the website, engine name and so on) indicates if it's a serious engine or just a clone.

to:

For those of you, who want to check which engines are clones and which aren't, I recommend looking at the professional rating lists like CCRL (http://computerchess.org.uk/). They check new engines, don't test clones and are independent. The lists are updated regularly, so you can expect them to be up to date. Also the engine's appearance (description on the website, engine name and so on) indicates if it's a serious engine or just a clone.

http://rybkaforum.net/cgi-bin/rybkaforu ... ?tid=14541

I think this makes CCRL look bad, and confirming the idea suggesting that the rating list is not independent.

Agreed that CCRL is NOT independent. More so, the statement made by Felix Kling? stating, " I recommend looking at the professional rating lists like CCRL...." confirms what some (myself included) have been saying all along that CCRL is an exclusive supporter of Rybka. It shouldn't shock anyone when I make the statement that CCRL stands to benefit from this and vice versa.

To sum;

1. Pure selfishness, greed, "cover ups",and monopoly by author of Rybka. He doesn't want to see other authors succeed, even if we assume they did the same thing he did with Fruit/Rybka. Of course, I'm NOT saying other authors did this but it shows the hypocricy of Vas.
2. And pure cover ups, "principle & moral helplessness" , and double standard by the CCRL - in light of Fabien confirming Rybka is Fruit-clone.

On this note, let me reiterate: My opinion of CCRL is based on long established patterns of "how they do things" and should not be misconstrued as me caring if CCRL includes Houdini in their list. I prefer other honest rating lists to the CCRL's and quite content with non-CCRL lists. These honest, non-CCRL rating lists individuals can think for themselves, are not biased, see the truth or facts objectively and refuse to be corrupted by Rybka's monopoly/commercial interests.

[BB+]

Here is a suggestion (largely to CCRL): list the engines alphabetically (at least as the default) rather than by strength. After all, if strength is not typically the main concern, why then are the engines sorted in this manner? Perhaps I am hoping too much from the website capabilities, but you could then also have "Show engines near this one in strength" (for instance, you click on Crafty, it gives you the 10 closest engines in either direction). This would hopefully get rid of much of the animosity toward perceived CCRL/Rybka links.

[Prima]

Following the TalkChess on Houdini now tested in CEGT, here's an excerpt of what Roger Brown stated:

I think that you are placing Graham in a particularly evil position.

He should not be the one to explain why Rybkachess did what it did.

Rybkachess was set up to promote a specific commercial chess engine, nothing more, nothing less. Lists that show that engine in less than a dominating light are going to be removed. I would not find that behaviour the least bit surprising on any other commercial based site.

I know Roger Brown is trying to provide a reasonable explanation for the obvious censorship/banning by the Rybka forum moderators and "Why" CCRL won't test engines better than RYbka because of commercial conflicts of Rybka's interest (my interpretations).

But, it still begs the question: Why didn't the Rybka moderators/Vas just say they're interested in seeing Rybka as #1 on the Rybka-based rating list (CCRL) to ensure selling Rybka in high volumes, as opposed to the false labeling of an engine stronger than Rybka as "Rybka-clones"?

Was this what the unfounded

"IppoLit/RobboLito...etc are Rybka3 clones"

all about?

As for Graham Banks being put in a particularly evil situation, he placed himself in that position. No one else did. He took the words of Vas without actual substantiation and also censored many during his term as a CCC moderator, among other things he did.

Now Fabien Letouzey, from which Rybka came forth from, has stated that Rybka is derived from Fruit. Others computer science/programming experts have been saying the same thing for years. I wonder if Graham Banks will truly live by his "holy principles" and take the words of Fabien Letouzey over Vas's that Rybka is indeed Fruit-derived/cloned.

This is the kind of double standard "evil situation or position" Graham Banks places himself in. Let's see him get out of his entrapment.


http://www.talkchess.com/forum/viewtopi ... 894799943b


PS: Roger Brown, I'm not attacking you or attempting to make you look bad. Far from it. I do think that people need to understand how Rybka forum moderators/CCRL react & operate in response to engines, especially free engines, better than Rybka in the commercial field.

[Jeremy Bernstein]

When all else fails, blame the users. Thanks for the laugh, H.G. Muller:

Well, if I were a CCRL member, I would _never_ test Houdini, no matter what the legal status of it turned out to be. Simply for the reason that most Houdini users present themselves here as a bunch of indecent, demanding bullies, that think they have the right to tell others what to do, and insult or accuse them of whatever they please when they don't get their way. I would never want to have anything to do with such riphraph, or with the engine they support!

CCRL is a great service to us, 'small engine' authors. I hope it will always stay that, and that they will resist pressure to waste valuable testing time to please people that do not deserve it. (And are not generally useful by any standard, as they did not author any engine at all, not even a 'small' one, their only 'accomplishment' in life apparently telling others what they want done...)

[kingliveson]

Here is a suggestion (largely to CCRL): list the engines alphabetically (at least as the default) rather than by strength. [...]

Not a good idea.

[alfons]

When all else fails, blame the users. Thanks for the laugh, H.G. Muller:

I didn't laugh and I think he's right on the spot (but i'm not so interested in rating lists either...).

Don't buy that story of Houdini*/Houdartmeister btw.:
Original? Didn't copy anything to make that thing work?

I don't think so...and Grahambanks should certainly be free to do whatever he wants on his island.

*which is just another blackbox/code-sucker

[MichaelIsGreat]

Hello to All,

I would advise you to read my own post "CCRL Forum censored my posting on Houdini 1.5 unjustifiably!" at http://www.open-chess.org/viewtopic.php?f=7&t=887 You will be astonished and shocked to discover like I did that both the CCRL and the CEGT chess engines rating lists have a hidden agenda of protecting their pet engines against new stronger chess engines! Hard to believe but true!
I checked four sites where you can find reliable rating lists for the strongest chess engines currently available. See further below.

The only reason why the CEGT rating lists have started, only very partially, to include in their results the latest two versions of Houdini is because they know that they have been uncovered by me and others too as having a shameful agenda of protecting their pet engines from stronger new chess engines!

WARNING:
Any chess engine rating list that avoids including the strongest chess engines (in particular Houdini 1.5!!) has basically a hidden agenda of wanting to avoid dethroning their pet chess engines (very often Rybka 4)! It is the case with the CCRL and the CEGT chess engines rating lists!! In such a case, you know that these chess engines rating lists are completely unreliable and nothing less than dishonest!! Avoid them at all costs!!
By carefully selecting dishonestly which chess engines to test and which ones to avoid testing, one could create completely bogus chess engines rating lists that are completely unreliable, with the only goal of keeping at the top someone's pet chess engines that have been dethroned by stronger and more recent chess engines!! Keep that in mind when you decide which chess engines rating lists to visit regularly!
That is not very much different than what has been done, for quite some time, by the CCRL and the CEGT chess engines rating lists!!! By being denounced by several people and by me in particular on several chess forums, they now realize that they have to stop producing dishonest and completely unreliable rating lists!! That is the only reason why you now see Houdini 1.5 and its previous version being stated, at least partially so far, on the CEGT latest rating lists!


I suggest four reliable and trustworthy chess engines rating lists. Reliable and trustworthy because they compile their chess engines rating lists honestly and they do not have a hidden agenda of wanting to avoid testing particular strong chess engines in order to keep their pet engines at the top!

1) The TCEC (Thoresen Chess Engines Competition) web site (40 MOVES IN 100 MIN+NEXT 20 MOVES IN 50 MIN+20 MIN FOR THE REST+10 S ADDED PER MOVE AT THIS LAST TIME CONTROL; 6 CORES; TABLEBASES USED; PONDER OFF; HASH USED; ALL OPENING MOVES ARE RANDOMLY FETCHED FROM A PGN FILE WHICH CONTAINS 200.810 DIFFERENT OPENINGS AND THEY ARE ALL FIXED TO 12 MOVES / 24 PLIES) at http://www.tcec-chess.org/
This site is outstanding! It is not a chess engine rating list but it should give you a very good idea of which chess engines are currently the best, as it does several tournaments with the best chess engines available. Check the listing of the chess engines in each category (Elite, Division I, II, and III) and you have a very good idea which chess engines are currently the best.
Highly recommended.
In particular, check the current tournament for the Elite Match between Houdini 1.5a and Rybka 4.0 (that is for the top two chess engines in the world). Very interesting games have been played so far.

2) The G/90mins Ratings (90 MIN TO PLAY ALL THEIR MOVES; 2 CORES; TABLEBASES USED; PONDER ON; HASH USED; OWN OPENING BOOK USED) at http://www.brinan.pwp.blueyonder.co.uk/ ... atings.htm
This web site is outstanding! Highly recommended.
Visit in particular the other links on this web site.

3) The IPON-Rating-List (5 MIN/GAME+3 S/MOVE; 1 CORE; TABLEBASES USED; PONDER ON; HASH USED; 50 DIFFERENT OPENING POSITIONS [NO BOOKS] ) at http://www.inwoba.de/
For those more interested at how chess engines perform at playing quick games.

4) The SWCR Rating (40 MOVES IN 10 MIN; 1 CORE; TABLEBASES USED; PONDER ON; HASH USED; RANDOM OPENING BOOKS USED) at http://www.amateurschach.de/
Again, for those more interested at how chess engines perform at playing quick games.


Best Regards to All
MichaelIsGreat