Developer tests of Stockfish need Stockfish 8 instead of Stockfish 7

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.
sovaz1997
Posts: 198
Joined: Sun Nov 13, 2016 9:37 am

Re: Developer tests of Stockfish need Stockfish 8 instead of Stockfish 7

Post by sovaz1997 » Tue Jun 12, 2018 5:41 pm

corres wrote:
Tue Jun 12, 2018 5:19 pm
sovaz1997 wrote:
Tue Jun 12, 2018 4:54 pm

I do not know English very well, to understand how polite I am writing.
Patience and politeness do not depend on knowledge of English but one's attitude.
Perhaps but not completely.
Zevra chess engine
Binary, source and description here: https://gitlab.com/sovaz1997/zevra2/tags

syzygy
Posts: 4377
Joined: Tue Feb 28, 2012 10:56 pm

Re: Developer tests of Stockfish need Stockfish 8 instead of Stockfish 7

Post by syzygy » Tue Jun 12, 2018 6:45 pm

corres wrote:
Tue Jun 12, 2018 5:19 pm
sovaz1997 wrote:
Tue Jun 12, 2018 4:54 pm

I do not know English very well, to understand how polite I am writing.
Patience and politeness do not depend on knowledge of English but one's attitude.
But what do you expect if you ask on Talkchess which is not the SF developer forum the SF developers to test against SF8 instead of SF7 when the SF developers basically only test against SF master.

It's not clear where to start if we want to resolve your confusion...

syzygy
Posts: 4377
Joined: Tue Feb 28, 2012 10:56 pm

Re: Developer tests of Stockfish need Stockfish 8 instead of Stockfish 7

Post by syzygy » Tue Jun 12, 2018 6:48 pm

Of course the real problem is that you are confusing this web page with SF development:
https://nextchessmove.com/dev-builds

The NCM site has nothing to do with SF development.

Still, you can try to contact whoever runs that site and try to convince her or him to switch to testing against SF8 or SF9 or whatever. But if you succeed that will have exactly zero effect on SF development.

corres
Posts: 1072
Joined: Wed Nov 18, 2015 10:41 am
Location: hungary

Re: Developer tests of Stockfish need Stockfish 8 instead of Stockfish 7

Post by corres » Tue Jun 12, 2018 8:59 pm

syzygy wrote:
Tue Jun 12, 2018 6:48 pm
Of course the real problem is that you are confusing this web page with SF development:
https://nextchessmove.com/dev-builds

The NCM site has nothing to do with SF development.
You are right, I addressed my post to wrong direction.
But the the relative Elo loss of Stockfish is a real issue.
Patience and politeness are such tings what are due to everybody, I think.
I would be happy if you called attention of me to my fault at first.

Leo
Posts: 704
Joined: Fri Sep 16, 2016 4:55 pm
Location: USA/Minnesota
Full name: Leo

Re: Developer tests of Stockfish need Stockfish 8 instead of Stockfish 7

Post by Leo » Tue Jun 12, 2018 9:56 pm

I guess it is back to Drawfish again. I thought they had the draw problem fixed.
BrainFish 190310 Elo 3568 . AMD fan

sovaz1997
Posts: 198
Joined: Sun Nov 13, 2016 9:37 am

Re: Developer tests of Stockfish need Stockfish 8 instead of Stockfish 7

Post by sovaz1997 » Tue Jun 12, 2018 10:10 pm

corres wrote:
Tue Jun 12, 2018 8:59 pm
syzygy wrote:
Tue Jun 12, 2018 6:48 pm
Of course the real problem is that you are confusing this web page with SF development:
https://nextchessmove.com/dev-builds

The NCM site has nothing to do with SF development.
You are right, I addressed my post to wrong direction.
But the the relative Elo loss of Stockfish is a real issue.
Patience and politeness are such tings what are due to everybody, I think.
I would be happy if you called attention of me to my fault at first.
What you did not like about my communication? Specific lines please write.
Zevra chess engine
Binary, source and description here: https://gitlab.com/sovaz1997/zevra2/tags

syzygy
Posts: 4377
Joined: Tue Feb 28, 2012 10:56 pm

Re: Developer tests of Stockfish need Stockfish 8 instead of Stockfish 7

Post by syzygy » Tue Jun 12, 2018 11:01 pm

corres wrote:
Tue Jun 12, 2018 8:59 pm
syzygy wrote:
Tue Jun 12, 2018 6:48 pm
Of course the real problem is that you are confusing this web page with SF development:
https://nextchessmove.com/dev-builds

The NCM site has nothing to do with SF development.
You are right, I addressed my post to wrong direction.
But the the relative Elo loss of Stockfish is a real issue.
No, it is a question of too few games. SF had a good run and was overrated. Now it has a bad run. That's why many games are needed and that's why error margins are calculated. If you choose to ignore error margins, then that is your choice, but on a forum like this you will then just have to accept that people will point out your unscientific approach in no unclear terms.
Patience and politeness are such tings what are due to everybody, I think.
You insist on ignoring error margins. Others don't need to have patience with that.
I would be happy if you called attention of me to my fault at first.
At first? I am not a nanny continuously monitoring this forum to save you from trouble. Instead, you could do a tiny bit of research before posting (or realising it is better not to post if you want to avoid "unpoliteness"). I mean, if you decide to open a topic to instruct the SF developers what to do, then the least you could do is to do a bit of research first, right? Or am I seeing it all wrong now?

Nay Lin Tun
Posts: 436
Joined: Mon Jan 16, 2012 5:34 am

Re: Developer tests of Stockfish need Stockfish 8 instead of Stockfish 7

Post by Nay Lin Tun » Wed Jun 13, 2018 12:12 am

Leo wrote:
Tue Jun 12, 2018 9:56 pm
I guess it is back to Drawfish again. I thought they had the draw problem fixed.
Most probably, SF in last season was overrated. SF won over Houdini was much beyond expected elo gap from other rating lists.

corres
Posts: 1072
Joined: Wed Nov 18, 2015 10:41 am
Location: hungary

Re: Developer tests of Stockfish need Stockfish 8 instead of Stockfish 7

Post by corres » Wed Jun 13, 2018 7:43 am

syzygy wrote:
Tue Jun 12, 2018 11:01 pm
No, it is a question of too few games. SF had a good run and was overrated. Now it has a bad run. That's why many games are needed and that's why error margins are calculated. If you choose to ignore error margins, then that is your choice, but on a forum like this you will then just have to accept that people will point out your unscientific approach in no unclear terms.

So TCEC is absolute unnecessary thing.
The results of TCEC are depend on the stroke of luck merely - at least under the strict scientific approach.
But tests shows that after a few turns it forms near the final order. Why?
Maybe in the case of a relative few games the formula accepted to be scientific write down inaccurately the facts.
You insist on ignoring error margins. Others don't need to have patience with that.
Patience and politeness are independent from error margins of results.
Rather it depends on the "error margins" of the man who attacks me.
I would be happy if you called attention of me to my fault at first.
At first? I am not a nanny continuously monitoring this forum to save you from trouble. Instead, you could do a tiny bit of research before posting (or realising it is better not to post if you want to avoid "unpoliteness"). I mean, if you decide to open a topic to instruct the SF developers what to do, then the least you could do is to do a bit of research first, right? Or am I seeing it all wrong now?
OK, you are not a moderator now as I see.
You are right, I would be more cautious when I chose the subject of my post, sorry.
But the "impoliteness" is not my mistake.
My post is not instruct the developer of Stockfish. It was only a proposal if it was a bad proposal.

syzygy
Posts: 4377
Joined: Tue Feb 28, 2012 10:56 pm

Re: Developer tests of Stockfish need Stockfish 8 instead of Stockfish 7

Post by syzygy » Wed Jun 13, 2018 6:41 pm

corres wrote:
Wed Jun 13, 2018 7:43 am
So TCEC is absolute unnecessary thing.
TCEC is as unnecessary as the game of chess.
The results of TCEC are depend on the stroke of luck merely - at least under the strict scientific approach.
The winner of TCEC is the TCEC winner. It was never the purpose of TCEC to determine the best engine.

The winner of the FIFA World Cup will be the FIFA world champion for four years and that's about it. Which team would come out on top after 40,000 games we will never know.

Post Reply