Forums

My nephew beat HOUDINI?!?!

Sort:
schachfan1

You know, when you work hard on some variation, when you give Houdini (and not only Houdini) time control 40 hours/40 moves when you check some possible and interesting ideas (not only ideas which strong engines suggest, but your own ideas, too), when it takes 1-2 hours (sometimes even more) for Houdini to make one move with that time control, and when in that way after one or two weeks you invent one single move and put it carefully into your notes while working on that variation, and so on move after move - after several months you can of course give Houdini, or Stockfish, or Rybka the usual time control of 2 hours/40 moves and just see what those chess monsters will do against your elaborations

OldChessDog

I beat Houdini the other day myself. Of course it was set at 1400... ;-)

LegoPirateSenior
Indubioproaggredi wrote:

[...] if you are seriously i can give you the phone number. i think it is a good idea to let him play more games on that level - or notate some games he plays vs other strong players and send these games first. i also would like to get these games - i know some high rated GMs and maybe they will have the time to analyse it - also his lost games. my own level is not high enough - i just started about 10 months ago and do it more seriously since 3 months now. if you want to send some of these games then send it to: info@chessplace.net - or info@greweling.com.

This reminds me of a little known fact: the word "gullible" is not listed in either Merriam-Webster or in the Oxford English dictionaries (neither in the printed nor in the online versions).

bigpoison

Ha!  That's funny.  Nor is there a definition of credulous.

samtoyousir

I know someone who looked it up twice... XD

samtoyousir

Was that second post marka marks last post? Because he logged on last month. XD

PearlFey
LegoPirateSenior wrote:
Indubioproaggredi wrote:

[...] if you are seriously i can give you the phone number. i think it is a good idea to let him play more games on that level - or notate some games he plays vs other strong players and send these games first. i also would like to get these games - i know some high rated GMs and maybe they will have the time to analyse it - also his lost games. my own level is not high enough - i just started about 10 months ago and do it more seriously since 3 months now. if you want to send some of these games then send it to: info@chessplace.net - or info@greweling.com.

This reminds me of a little known fact: the word "gullible" is not listed in either Merriam-Webster or in the Oxford English dictionaries (neither in the printed nor in the online versions).

Someone who was gullible wouldn't look it up, they would reblog it or something.

HaveAnotherGame

It should be.

mrtampa2

Lol. I don't believe the post, sorry. Noone playes over 3000 strength at age 8. If it was true he would crush all the current GMs too.

mrtampa2

Even If he was Morphy in his previous life he still would be not smart enough to beat Houdini. May be he is an android from the future... I put Stockfish 5, Commodo against Houdini but still Houdini came out better most of the time. 

pdela

My god, the OP is a post or War and Peace

MisterBoneman

haven't checked all 22 pages to see if markamark came back, however, a few other points should be introduced. A 3000 player, child or adult, might not be known if he plays only White...or, as is known to happen at times, the person suddenly goes off, knocking over pieces and board in a nightmare reaction to his own thoughts. That last does not go over well at Chess matches at all, I think. Especially when you consider all the people who complain bitterly about things such as "he didn't reply fast enough...I could have won sooner..."

Another thought not so indelicately repeated is reincarnation. Houdini? HA! Chess? DOUBLE HA! Mozart. Yeah...THAT Mozart. You know... The bartender in Springfield. no...not the bartender.
Shameless copy of Mozarts older sister's recollection, from wikipedia,
"... In the fourth year of his age his father, for a game as it were, began to teach him a few minuets and pieces at the clavier.... He could play it faultlessly and with the greatest delicacy, and keeping exactly in time.... At the age of five, he was already composing little pieces, which he played to his father who wrote them down"

I dare anyone here to sit down and write a two page song let alone minuets, studies, and themes. Maybe one in tenthousand can? There are ONLY 271+ billion move variations in the first four moves of Chess...there are no ends to how many eight note tunes can be found in music (eight because four moves is White four + Black four) It is infinitely harder.

But the sad part is, if true, the child will at some point be left in a very lonely position of not having anyone to play at all. None will be able to, nor will they want to because of the implications on their "worthiness."

PearlFey

Bears shouldn't drink kool-aid.

sean42
LegoPirateSenior wrote:

Here are results of a quick analysis of that game, using Houdini 1.5a, 512MB hash, 10s/move, at least 14 plies, analysis starting from move 26 (per pfren's diagnosis of 25 book moves). CPU: 2.66 GHz i7, single processor used.

For those unfamiliar with top-3 methodology for detecting engine use, top-1 is the percentage of positions where the 1st egine choice was played, top-2 is the percentage of positions where either the 1st or 2nd engine choice was played, et.c. 

Adam Top1 match: 24/36 = 66.67 %
Adam Top2 match: 33/36 = 91.67 %
Adam Top3 match: 35/36 = 97.22 %

Houdini 3 x64 Top1 match: 25/35 = 71.43 %
Houdini 3 x64 Top2 match: 32/35 = 91.43 %
Houdini 3 x64 Top3 match: 33/35 = 94.29 %

The 10s/move analysis time was based on the OP's statement of having given Houdini 5 minutes for the entire game. This setting is not really correct, but not much can be done without knowing the exact time that the opponents took for each move. I have only the free version of Houdini, so that's another source of inaccuracy.

Given the inaccuracies and small sample, there's no way to draw firm conclusions, however, the numbers are consistent with what might be expected from analyzing a Houdini-Houdini match.

Interesting. Adam is more Houdini-like than Houdini!  

BS!!!!

MrDamonSmith

Wow, you mean he out-Houdinied Houdini? That's almost Borislavish like. Oh......wait.......hmm.....ok......

sisu

This old chestnut thread still going?

It's already been summarised on page 18 comment 341.

Shah_Maht

Its a well known fact to any competent computer scientist chess progammer that the perfect version of any chess playing computer plays only slightly better than a person playing randomized moves if its opening and engame database are off. The computer cannot properly calculate the entire string of possibilities and depending on the time control it will stop the thorough analysis well before any concrete desicion can be determined than it almost "randomly" will select a ove. It's not random but it's based on it's highest move rating which often is silimar to picking the move that delays any end as long as it can. Also it always plays for a draw if no forced mate /win is in sight. In fact I performed ten studies where i played the most advanced houdini and fritz on the highest rating (w/o any preprogrammed databases on of course) and selected my moves based on a nearby battleship game as well as by using the random # button in the calculator to determine my moves. I played them each 5 games. The somewhat predictable result was that i won three games (two to Houdini and one to Fritz) and lost seven. The drawish games against Fritz were drawn since Fritz offered and against Houdini, Houdini won.

PearlFey

Its a well known fact to any competent computer scientist chess progammer that the perfect version of any chess playing computer plays only slightly better than a person playing randomized moves if its opening and engame database are off.



[Citation needed]


PearlFey

https://www.youtube.com/watch?v=CrLbB4ieYfI

Sure looks like it's playing random moves to me.

EscherehcsE
PearlFey wrote:

Its a well known fact to any competent computer scientist chess progammer that the perfect version of any chess playing computer plays only slightly better than a person playing randomized moves if its opening and engame database are off.



[Citation needed]


[Citation will never be supplied, because it's complete B.S.]