Martin Bryant opinion about Strelka issue

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

Moderator: Ras

Christopher Conkie
Posts: 6074
Joined: Sat Apr 01, 2006 9:34 pm
Location: Scotland

Re: Martin Bryant opinion about Strelka issue

Post by Christopher Conkie »

Dann Corbit wrote:Martin is wrong.
It depends if you know what we know.

For what its worth Martin is quite accurate.

Regards

Christopher
MartinBryant

Re: Martin Bryant opinion about Strelka issue

Post by MartinBryant »

Hi Steve,
personally I don't believe that the positions that produce different results are of any relevance. By simply changing an evaluation weight slightly you can make an engine play very differently in those positions affected by that weight, even to the extent of tactics (because of move ordering).
To my mind it is the positions that produce identical results that are the giveaway.
PauloSoare
Posts: 1335
Joined: Thu Mar 09, 2006 5:30 am
Location: Cabo Frio, Brasil

Re: Martin Bryant opinion about Strelka issue

Post by PauloSoare »

Christopher Conkie wrote:
Dann Corbit wrote:Martin is wrong.
It depends if you know what we know.

For what its worth Martin is quite accurate.

Regards

Christopher
What you and some people know, Christopher?

Paulo Soares
MartinBryant

Re: Martin Bryant opinion about Strelka issue

Post by MartinBryant »

Well now I'm well confused! :D
I'm wrong, but I'm quite accurate, at the same time.
Sounds intriuging.

So what is it that you 'know' then?

Tell us all, please do...
Christopher Conkie
Posts: 6074
Joined: Sat Apr 01, 2006 9:34 pm
Location: Scotland

Re: Martin Bryant opinion about Strelka issue

Post by Christopher Conkie »

MartinBryant wrote:Well now I'm well confused! :D
I'm wrong, but I'm quite accurate, at the same time.
Sounds intriuging.

So what is it that you 'know' then?

Tell us all, please do...
We know how Rybka works.

All of it, right down to the last centipawn (yes i really mean centipawn).

Christopher
MartinBryant

Re: Martin Bryant opinion about Strelka issue

Post by MartinBryant »

Excuse my ignorance, do you work on the Rybka team?
You've seen the source then?
User avatar
fern
Posts: 8755
Joined: Sun Feb 26, 2006 4:07 pm

Re: Martin Bryant opinion about Strelka issue

Post by fern »

Sorry if my post sounded like that, but in fact not even me is absolutely convinced, as I said that you added extra weight. but I did not say I consider this issue already judged.
I hope you are mistaken, because I would like to see a new, original engine, as everybody else here.

STILL waiting a new colossus with a new, gorgeous gui regards
Fernando
Christopher Conkie
Posts: 6074
Joined: Sat Apr 01, 2006 9:34 pm
Location: Scotland

Re: Martin Bryant opinion about Strelka issue

Post by Christopher Conkie »

MartinBryant wrote:Excuse my ignorance, do you work on the Rybka team?
You've seen the source then?
No neither...but I bet you any money Vas would say we know rather alot about Rybka.

Enough for this for sure.

Regards

Christopher
User avatar
mclane
Posts: 18899
Joined: Thu Mar 09, 2006 6:40 pm
Location: US of Europe, germany
Full name: Thorsten Czub

Re: Martin Bryant opinion about Strelka issue

Post by mclane »

paradox
dialectic
MartinBryant

Re: Martin Bryant opinion about Strelka issue

Post by MartinBryant »

Maybe so.
However, with respect, people like Uri and Michael have tried to debate this issue by providing repeatable empirical evidence. All you have provided is mystery and intrigue. Their evidence is something that any person can get their teeth into and make an objective assessment of. It is a pity you cannot contribute something with objective value for all to assess.