When did chess engines running on common home pc hardware be

[Leo]

20-30 years more until top engines catch up with best humans in terms of strength

In what sense? A top engine on good hardware beats any human at any time control now.

I don't understand the statement either.

[tmokonen]

I don't understand the statement either.

It's trolling, designed to get attention. Pretty easy to understand.

[JohnW]

The question is -> how long until engines can catch up on opening understanding.

Engines are given opening book to compensate for relative mild opening play.

And how many human players simply memorize openings? No difference in my opinion.

[Lyudmil Tsvetkov]

I don't understand the statement either.

It's trolling, designed to get attention. Pretty easy to understand.

if you call trolling my positive score against SF(with small time odds in my favour, but fast TC), with more than 500 winning games available in my database, then I am fine with it.

[Lyudmil Tsvetkov]

When did chess engines running on a common home pc hardware became stronger than the strongest human?
Was it around 2006?
By common home pc hardware I mean the kind of hardware people would buy (like Intel core 2 duo E6300 back in 2006). I am excluding server machines.

I am interested in which hardware, which human and which chess engine you consider in your reply.

Thank you!

Hi,

there used to be a few human vs comp databases floating around. There is one by G. Tsavdaris at Ed's site but I can't reach the site that hosts it.
The link is:
http://www.top-5000.nl/gm-comp/dgt_chesstheatre.htm


Another one is hosted here:
http://privat.bahnhof.se/wb432434/welcome.htm

It would be great if someone could repost the G. Tsavdaris db.
Thanks,
CL

PS. To answer the original question, I think the top engines reached parity [at STC] with the strongest humans in the 2001-2006 period, and went beyond it after 2006 (Rybka being a notch or two stronger).

most importantly, my 1000+ won games against SF will be missing.

not very representative database in this case, I think.

By all means, Lyudmil, I hope they can include all your games.

thanks Carl.
a highly appreciated opinion by a real connoisseur.

[tmokonen]

if you call trolling my positive score against SF(with small time odds in my favour, but fast TC), with more than 500 winning games available in my database, then I am fine with it.

What database, trollmeister Tsvetkov?

[Lyudmil Tsvetkov]

if you call trolling my positive score against SF(with small time odds in my favour, but fast TC), with more than 500 winning games available in my database, then I am fine with it.

What database, trollmeister Tsvetkov?

Des jungen Trollmeisters Tsvetkovs ccc-Leiden und -Leidenschaften.

my personal database, man, my personal one, which, btw., is growing with each and every day.

I never finish another day until I obtain at least 2 wins from SF.

so, tomorrow, there will be 2 more.

[Werewolf]

I can't see your FIDE rating on their site.

Perhaps you have a special way of playing against engines?

I'm basically saying: if we sat you down - unaided and under controlled conditions - against latest SF on good hardware with a book you hadn't seen before, and played a match at tournament time controls, I'd be amazed if you scored more than 5%.

[Lyudmil Tsvetkov]

I can't see your FIDE rating on their site.

Perhaps you have a special way of playing against engines?

I'm basically saying: if we sat you down - unaided and under controlled conditions - against latest SF on good hardware with a book you hadn't seen before, and played a match at tournament time controls, I'd be amazed if you scored more than 5%.

my FIDE rating is just 2100+, Bulgarian rating 2200+, but I have not played for 12 or so years.

SF is no match for me, of course I will score more than 50% against it under normal, Fischer conditions.

by Fischer conditions I understand a back room, with no buzzing cameras, talking people and obtrusive light.

I never play under any other conditions, and as such conditions are mostly unavalable, I never play in public.

[Milos]

Tsvetkov is actually an amazing phenomenon. He achieved a perfect chess anti-correlation, namely his claims have 100% fail rate which is really astonishing considering they are related to binary problems with only two possible outcomes and the expected random rate should be 50%.
So in an essence he could be used as a perfect chess oracle and is certainly better than SF or any modern engine.
The only correction that one has to make for the sign of "Tsvetkov's evaluation" but that's just a very minor correction .