Test your cheater detection skills!

[Red Hood]

Without engine help and based solely on my observation I didn't find any engine. They all made tactical and strategical errors. These are not even GM games as far as I can tell maybe i am wrong. Maybe it's an experimental engine, weak, without much knowledge. In all games they made a lot of unnecessary pawn moves and lost pawns due to errors of judgment.

I've looked so called new games. When i start to look the old ones i looked only the first game and it seem it's the same as the first one before in the new ones.

[yanquis1972]

....what???

he says 7 of ten are computer vs human, & 3 are human games. each game is also given a runthru by none other than houdini (albeit 1.5), which approves of virtually all moves played, at least at fairly low depth.

[Red Hood]

I just said what i think. Moves played aren't computer like. Many, many mistakes. Moves are good up to a certain point. I almost thought i caught some of them than a series of bad moves come.

[yanquis1972]

that may well be what you think, but i'd at least do the slightest bit of digging before stating it. i can assure that in at least 2 of the 3 games i looked at (no sleight to a particular poster here who participated in one of them), the humans were titan-level strong by reputation; & this is presumably their play at its sharpest -- as evidenced by the computer analysis.

if its so miserable than you're the next fischer, i guess, & you can make an announcement to the world by displaying what a chump houdini 1.5 apparently is.

[Mike S.]

[Edited by Moderation to add pgn tags]

Thanks. - Please ignore the "Event Date" in the PGN; this was just the date to which the selection was compiled.

[Mike S.]

how'd i do mike?

Quite ok I'd say... I cannot tell more yet, to avoid giving hints.

Actually I have no idea myself, how easy or difficult the detection is each, because of course I knew who is who beforehand. And I need to admit, I made a kind of diabolical choice of the games, with the help of Houdini.

[Mike S.]

No cojones?

Which one was the computer, in all of these games (except 3)? If THAT cannot be proven, then NOTHING can be proven...

[elcabesa]

I tried, started some analisys, but I can't find

[jefk]

[quote="Mike S."]No cojones?

Which one was the computer, in all of these games (except 3)? If THAT cannot be proven, then NOTHING can be proven...[/quote]

well this should be something for the comp detection program this guy Regan made, and i guess then also the time(s) per move should be included.
http://www.cse.buffalo.edu/~regan/

[Mike S.]

(...) i guess then also the time(s) per move should be included.

Thanks, I took a look into it. I am afraid the move times are unavailable. But I am not sure if they would be required for K.Regan's method, or if it's suitable to identify the computer in a single game, each.

Anyway, the PGN posted here already includes blundercheck results calculated by Houdini 1.5a at 18 plies of depth, threshold 0.25.