Not logged inRybka Chess Community Forum
Up Topic The Rybka Lounge / Computer Chess / WCCC vs TCEC Chess Eng Competitions..Calls To End Disputes
- - By supersharp77 (*) Date 2019-06-12 17:32
By chance happend across the WCCC tourney details and had No Idea Stockfish has never won the competition and that Komodo is the defending champion (3 Times Running) Also had no idea Shredder has been boycotting the TCEC competitions (although I thought they were entered into this years Premier Div..Not sure) Quite a lot of elite names of Chess engines have won the WCCC Chamiponship including Kaissa, Cray Blitz, Hiarcs, Deep Junior, Sjeng, Komodo, Rybka (DQ'ed) Shredder Fritz Deep Thought etc.. but No Houdini or Stockfish...Surprising and very controversial. They demand "Equal Hardware Specs" How will the LC0 & similar NN engines fit into these specifications? Interesting days ahead!

1 1974  Stockholm  13 Kaissa
2  1977  Toronto  16  Chess 4.6[2]
3  1980  Linz  18 Belle
4  1983  New York  22  Cray Blitz
5  1986  Cologne  22  Cray Blitz
6  1989  Edmonton  24  Deep Thought
7  1992  Madrid  22  ChessMachine (Gideon)
8  1995  Hong Kong  24  Fritz
9  1999  Paderborn  30  Shredder
10  2002  Maastricht  18  Deep Junior
11  2003  Graz  16  Shredder
12  2004  Bar-Ilan U  14  Deep Junior
13  2005  Reykjavík  12  Zappa
14  2006  Torino  18  Junior
15  2007  Amsterdam12  Zappa[wccc 1]
16  2008  Beijing  10  HIARCS[wccc 1]
17  2009  Pamplona  10  Junior, Shredder, Sjeng[wccc 1]
18  2010  Kanazawa  10  Rondo, Thinker[wccc 1]
19  2011  Tilburg  9  Junior
20  2013  Yokohama  6  Junior
21  2015  Leiden  9  Jonny
22  2016  Leiden  6  Komodo
23  2017  Leiden  4  Komodo
24  2018  Stockholm  8  Komodo
^ Jump up to: a b c d Rybka originally won the WCCC in 2007 -2010, but was later disqualified for plagiarising code in a controversial decision.
World Chess Software Championship[edit]
From 2010 a new tournament was introduced and held at the same location and during the same period as the World Computer Chess Championship. The rules for the World Chess Software Championship state that competing programs must run on machines with identical hardware specifications. Time control is game in 45 minutes with 15 second increment.[3][4]

Event #  Year  Location  Participants  Winner  Hardware
1  2010  Kanazawa  9  Shredder[5]  Intel quad core Xeon 2.66 GHz, 8MB Hash
2  2011  Tilburg  5  HIARCS  Intel Core2 Duo, 1.7 GHz, 2MB Hash
3  2013  Yokohama  6  HIARCS  Intel quad core i7, 2.7 GHz, 16MB Hash
4  2015  Leiden  8  Shredder [1]  Intel quad core i7, 2.7 GHz, 16MB Hash
5  2016  Leiden  7  Komodo  Intel quad core i7, 3.4 GHz, 16MB Hash [2]
6  2017  Leiden  7  Shredder  Intel quad core i7, 3.4 GHz, 16MB Hash
7  2018  Stockholm  9  Komodo  Intel quad core i7, 1.8 GHz, 16MB Hash

Re: WCCC vs TCEC Chess Engine Competitions..
Post  by bob » Wed Jun 12, 2019 2:24 am

I would like to add something here, since "some" are trying to keep various types of "bitterness" alive and well.

Computer chess went through three really distinct phases.

Phase 1 started in 1970 and continued into the early 90's, where we had annual ACM computer chess tournaments (and every three years a WCCC event). These were "in person" events where the programmers showed up with either a computer or a terminal to access their remote machine. They were about the tournament, AND the paper presentations from the authors. This was what I would call "the fun times" of computer chess. This was the "run whatcha brung" and included everything from $50m Crays to basic microprocessor systems to custom designed chess hardware including Belle, HiTech, Deep Thought, and even a few not-so-custom things like ChessMachine Schroeder and such. Clusters - welcome. Supercomputers - welcome. Z80's - welcome. Etc.

Phase 2 started in the middle 90's after the annual ACM events ended (last was 1994). At that point we were organizing online events using the original computer chess club (which morphed into the Internet Computer Chess [ICC] most of used for many years (no relationship to the current computer chess club message board). These events were more fun, because they could have more rounds, required ZERO travel/expenses, but we lost the "meetings" and "papers" part of the old ACM era. Yes, we had channel 64 discussions, but while those were often interesting and useful, they were not quite the same. There was much less exotic hardware used with one notable exception that clusters were working their way into the mix. They are really "poor man's supercomputers" at some level, but there was very little additional effort expended in custom hardware design once deep thought put away Kasparov.

Phase 3 is where TCEC currently is. Here NO programmers are involved in operating for the games or anything. NO limits on what programs are allowed (near 100% clones are apparently OK) and such. Obviously, TCEC is "the lowest common denominator" from a hardware perspective. They have seen fit to add some GPU hardware, but everyone gets the same basic hardware, whether they can use it all or not.

As we have progressed, the programmers were slowly fading away from the competition. Just a matter of technology changing the way things happen. I won't begin to say things are worse (or better) than they were 20 and 40 years ago. But they are most definitely different.

I will close with a short story to show how things have changed. When I won my first WCCC (1983) we used a dual-cpu Cray machine. It took a LOT of effort to use two processors in a parallel search. The next year we had 4 processors. When the 1986 WCCC event came along, and we won our 2nd consecutive title, we were running on an 8-cpu Cray YMP. People used to talk about how much time I wasted working on the parallel part of the search rather than the alpha/beta part of the search (and I actually did a LOT of work on both parts). Their complaint was "who cares? that hardware is too expensive and will never become "main stream"". After the 1994 ACM event in Cape May, I decided to do a complete re-write of Cray Blitz. Converting from Fortran to C, but more importantly converting from a vector-based super computer to a micro-computer (initially the Intel Pentium, and then the pentium pro). But interestingly, our local computer store dragged a new dual-cpu pentium II box out to my office (most of the employees there had taken classes under me at UAB). So, a brief detour to make Crafty a parallel search program. I remember comments on ICC at the time, "Wow, nice NPS, but who cares? that will never become main stream..." I gave my usual "never is a really long time..." A few years later, when the dual/quad CPU machines were becoming common, the next comment I heard was "who cares about these multiple CPU micro-computers? The future is the hand-held cell phone market and they will never be multiple CPU versions of those, they don't need the power." Need I mention how long "never" is.  :) So what I was doing in 1983 is STILL being done today (parallel search). If we had been "uniform platform" in the 80's, what a different world computer chess would be today.

Re: WCCC vs TCEC Chess Engine Competitions..
Post  by Dann Corbit » Wed Jun 12, 2019 6:26 am

It is true that Rybka was "chock full" of fruit ideas. It even said so in the release notes.
There were lits of clones before, and they received small sensible punishments, even though the cloning aspect was FAR more outrageous than Rybka.
It was the absurd, WAY over the top punishment that forever killed the wmccc in my eyes.

Personally, I would like to see everyone apologize.
I will start.
"I am sorry to carry a grudge so long. I am sorry computer chess was severely injured. I am sorry this wound has never healed. I am sorry that I cannot seem to let it go. I am sorry to see so many intelligent, caring, passionate people at each others throats. I wish I could fix it, but I have no idea how to even get started."
Parent - By Labyrinth (*****) Date 2019-06-12 22:03
WCCC is toast. Very little interest in it now. They can crown whoever they want as champion and it doesn't matter. They can crown Chenard the world computer chess champion for all I care.

>Personally, I would like to see everyone apologize.

Are you familiar with terms like 'when hell freezes over'? :-p
Up Topic The Rybka Lounge / Computer Chess / WCCC vs TCEC Chess Eng Competitions..Calls To End Disputes

Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill