A Crossroad in Computer Chess; Or Desperate Flailing for Relevance

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

Moderators: hgm, Dann Corbit, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Dann Corbit
Posts: 12141
Joined: Wed Mar 08, 2006 7:57 pm
Location: Redmond, WA USA
Contact:

Re: A Crossroad in Computer Chess; Or Desperate Flailing for Relevance

Post by Dann Corbit » Thu Oct 01, 2020 6:02 am

I guess at some point we will have 256 cores and 512 or more threads on the average desktop.
Why not 128 threads doing AB, 128 threads doing MCTS and the other 256 threads doing the better, more efficient search that gets discovered next year? After 128 threads, we seem to be adding less and less anyway.
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.

User avatar
Guenther
Posts: 3938
Joined: Wed Oct 01, 2008 4:33 am
Location: Regensburg, Germany
Full name: Guenther Simon
Contact:

Re: A Crossroad in Computer Chess; Or Desperate Flailing for Relevance

Post by Guenther » Thu Oct 01, 2020 6:08 am

gladius wrote:
Wed Sep 30, 2020 11:24 pm
AndrewGrant wrote:
Tue Sep 29, 2020 7:33 am
I could be all wrong. I could be out of touch. But if I'm not, then the future of computer chess, the future of unique and diverse engines, depends upon all of us, as individuals, to encourage and promote new ideas while discouraging those who take from Stockfish without trying their hand at the problem. I'm already concerned when I see engines with Stockfish nets being placed onto rating lists.
There are many motivations for being involved in computer chess. IMO, most do it for the enjoyment. Very few engines hit the level where cloning even starts to be a consideration.

...
This is wrong, history has shown that people clone engines nearly of all levels.
https://rwbc-chess.de
'chessqueen' 2018-present, aka: 'George' 2013-2016, 'pichy' 2006-2013, 'Jorge Pichard' 2000-2006 (old forum)
Troll barometer:
https://docs.google.com/spreadsheets/d/ ... KSptBx9AUs

smatovic
Posts: 1680
Joined: Wed Mar 10, 2010 9:18 pm
Location: Hamburg, Germany
Full name: Srdja Matovic
Contact:

Re: A Crossroad in Computer Chess; Or Desperate Flailing for Relevance

Post by smatovic » Thu Oct 01, 2020 6:24 am

Dann Corbit wrote:
Thu Oct 01, 2020 6:02 am
I guess at some point we will have 256 cores and 512 or more threads on the average desktop.
Yes, might be that Moore's Law will peak with 256 or 512 cores per socket, and then some new technology steps in with new opportunities for the next thing.
Dann Corbit wrote:
Thu Oct 01, 2020 6:02 am
Why not 128 threads doing AB, 128 threads doing MCTS and the other 256 threads doing the better, more efficient search that gets discovered next year? After 128 threads, we seem to be adding less and less anyway.
Currently we seem to have a kind of sweet-spot for office PCs and mobiles with about 4 cores or so, chess enthusiasts play in another league of course ;-)

Gaming will grow maybe up to 8k resolutions, so NNs in CC will still profit a while from GPUs.

The next thing in hardware and the next thing as killer-application may define what the next thing in CC will be, or alike.

--
Srdja

User avatar
mvanthoor
Posts: 982
Joined: Wed Jul 03, 2019 2:42 pm
Location: Netherlands
Full name: Marcel Vanthoor
Contact:

Re: A Crossroad in Computer Chess; Or Desperate Flailing for Relevance

Post by mvanthoor » Thu Oct 01, 2020 8:04 am

When reading some of the latest threads revolving about idea's and where they come from, I'm almost afraid to start writing an evaluation function.

I'm a decent chess player, but except if it's a -really- serious game, I'm also sometimes lazy with calculation and applying the rules and best practices. That doesn't mean I don't know them, however. Therefore I feel that I could get into the position where people say: "You stole evaluation term from <engine> and you didn't put that in your credits!", even though the only thing I did was implement what I would normally do (or -should- normally do) over the board.
Author of Rustic.
Releases | Code | Docs

smatovic
Posts: 1680
Joined: Wed Mar 10, 2010 9:18 pm
Location: Hamburg, Germany
Full name: Srdja Matovic
Contact:

Re: A Crossroad in Computer Chess; Or Desperate Flailing for Relevance

Post by smatovic » Thu Oct 01, 2020 8:55 am

selfdeleted, nevermind.

--
Srdja

Dann Corbit
Posts: 12141
Joined: Wed Mar 08, 2006 7:57 pm
Location: Redmond, WA USA
Contact:

Re: A Crossroad in Computer Chess; Or Desperate Flailing for Relevance

Post by Dann Corbit » Thu Oct 01, 2020 12:50 pm

Guenther wrote:
Thu Oct 01, 2020 6:08 am
gladius wrote:
Wed Sep 30, 2020 11:24 pm
AndrewGrant wrote:
Tue Sep 29, 2020 7:33 am
I could be all wrong. I could be out of touch. But if I'm not, then the future of computer chess, the future of unique and diverse engines, depends upon all of us, as individuals, to encourage and promote new ideas while discouraging those who take from Stockfish without trying their hand at the problem. I'm already concerned when I see engines with Stockfish nets being placed onto rating lists.
There are many motivations for being involved in computer chess. IMO, most do it for the enjoyment. Very few engines hit the level where cloning even starts to be a consideration.

...
This is wrong, history has shown that people clone engines nearly of all levels.
TSCP is a case in point.

And the odd part is, if you make a fork and give credit, most authors won't mind.
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.

User avatar
Guenther
Posts: 3938
Joined: Wed Oct 01, 2008 4:33 am
Location: Regensburg, Germany
Full name: Guenther Simon
Contact:

Re: A Crossroad in Computer Chess; Or Desperate Flailing for Relevance

Post by Guenther » Thu Oct 01, 2020 2:59 pm

Dann Corbit wrote:
Thu Oct 01, 2020 12:50 pm
Guenther wrote:
Thu Oct 01, 2020 6:08 am
gladius wrote:
Wed Sep 30, 2020 11:24 pm
AndrewGrant wrote:
Tue Sep 29, 2020 7:33 am
I could be all wrong. I could be out of touch. But if I'm not, then the future of computer chess, the future of unique and diverse engines, depends upon all of us, as individuals, to encourage and promote new ideas while discouraging those who take from Stockfish without trying their hand at the problem. I'm already concerned when I see engines with Stockfish nets being placed onto rating lists.
There are many motivations for being involved in computer chess. IMO, most do it for the enjoyment. Very few engines hit the level where cloning even starts to be a consideration.

...
This is wrong, history has shown that people clone engines nearly of all levels.
TSCP is a case in point.

And the odd part is, if you make a fork and give credit, most authors won't mind.
You only know and remember a tip of the iceberg...some you don't even know, because I stopped them
before getting known to the wider public. And some I haven't even revealed, except privately.
(proofs still on my HD)
https://rwbc-chess.de
'chessqueen' 2018-present, aka: 'George' 2013-2016, 'pichy' 2006-2013, 'Jorge Pichard' 2000-2006 (old forum)
Troll barometer:
https://docs.google.com/spreadsheets/d/ ... KSptBx9AUs

User avatar
emadsen
Posts: 277
Joined: Wed Apr 25, 2012 11:51 pm
Location: Oak Park, IL, USA
Full name: Erik Madsen
Contact:

Re: A Crossroad in Computer Chess; Or Desperate Flailing for Relevance

Post by emadsen » Fri Oct 02, 2020 12:54 am

AndrewGrant wrote:
Tue Sep 29, 2020 11:14 pm
Before working on Chess, I played League of Legends in the top few hundred players. Sometime between Chess and now, I played World of Warcraft in one of the top 10 guilds in North America. I think, a general understanding of myself, is that I do things to satisfy competitive urges. I tend to pick one thing at a time in life, and ruthlessly dedicate myself to it.
Thank you for sharing your thoughts, Andrew. It's refreshing to hear someone speak so directly from the heart. I marvel at what you've built with Ethereal. You should be very proud of yourself. I don't feel qualified to weigh in on the technical aspects of neural networks and how, combined with open source projects, they're altering our cherished hobby of chess engine programming. I'll address the emotional issue instead- in your words, feeling "disheartened."

For the most part, I believe a competitive nature is a good personal quality. It drives a person to improve themselves. That was on vivid display a few months ago, in a trip down memory lane, watching Michael Jordan in The Last Dance documentary. Reminded me how absolutely in awe I was of his talents and his competitive drive. I have never seen an athlete as competitive and focused as Jordan, nor have I seen another athlete who could raise the level of play among his teammates- not to the degree Jordan could. Yes, the documentary illustrates some corrosive aspects of his personality, but what's depicted is overwhelmingly positive. So yes, competitiveness is admirable.

However, it is inextricably connected to a need to be praised. This reliance on the attitudes- or obsequiousness- of others can be harmful. You're basically giving away the keys to your self-esteem to other people. People who may not have your best interest in mind. People who may cheat to get the praise they desire- stealing it from you. So, like most things in life, it's a balancing act: Find a profession or hobby where you can direct your energies towards competition. Find friends, a spouse, family where you can give instead of conquer. Balance the two or you'll be miserable.

I mention this because it's connected to something you said in your original post.
I worry about the future of Computer Chess. I see a timeline where a dozen engines use something very similar to Stockfish's methods. They all shoot up in elo. New developers, people working on their own innovations, are disheartened. They ask themselves why they toil away on their new ideas and tweaks, when one can just embrace the NNUE and be on an equal playing field with the top tiers of engines. So they leave.
Are they disheartened, though? Do they leave? I believe this is only true at the upper echelons. Easy for me to say with a middling 2500 engine, while you've built Ethereal into a monster and are up in the stratosphere competing with a huge crowd-sourced engine, a historically strong commercial engine, and unacknowledged clones. What you experience up there in the clouds may not be representative of us down here. I can only speak for myself: I dabble with my engine occasionally to satisfy my intellectual curiosity. To engage in a form of programming very different from my day-to-day business programming.
I've been working on Ethereal for 60hrs a week for the last 6 years.
This concerns me. Are you employed? You obviously are a talented programmer, so I assume you are. If so, this suggests you're burning the candle at both ends. How long can that last?

Sorry to get all philosophical on you. I hope in the near future you can find a healthy mindset and mix of activities (chess programming or otherwise) to satisfy your desire to compete and win and a need to connect and share. Because, like it or not, redressing the wrongs that have occurred in competitive chess engine programming is likely a futile exercise. People will always cheat to boost their fragile self-esteem, harming you in the process.

Those of us that have written chess engines from scratch know what you've achieved. The posers don't understand. They just argue ceaselessly.
My C# chess engine: https://www.madchess.net

RubiChess
Posts: 297
Joined: Fri Mar 30, 2018 5:20 am
Full name: Andreas Matthies

Re: A Crossroad in Computer Chess; Or Desperate Flailing for Relevance

Post by RubiChess » Fri Oct 02, 2020 5:30 am

AndrewGrant wrote:
Tue Sep 29, 2020 7:33 am
...
At the same time, this is me jumping up and down, waving my hands, saying "Hey, I've been working on Ethereal for 60hrs a week for the last 6 years."
...
"Come on, get a life!" is what comes to my mind when I read this. :-)

Reading all these controversial threads (well, I don't read them completely, it is just too much) of the last weeks makes me feel that most of us (including me sometimes) take it way too seriously. It is just computerchess, something that 99.9% of people never heard about. And there are other things to be worried about out there.

My 2 ct.
Andreas

Edit: Noticed that this is mostly a repost of Erik Madsen :-)

Tony P.
Posts: 216
Joined: Sun Jan 22, 2017 7:30 pm
Location: Russia

Re: A Crossroad in Computer Chess; Or Desperate Flailing for Relevance

Post by Tony P. » Fri Oct 02, 2020 6:48 am

Andrew, as an ex-esportsman, you know very well that the meta is a rapidly moving target even while the rules are left intact (the latter is, btw, not a given in chess after the latest Alpha0 paper). CC is no different :P It's very competitive. Fortunately, it's just a game too.

60 hrs a week are indeed too much to dedicate to a hobby, to my taste. Such dedication is fine on a vacation but not all year round.

Post Reply