amazing rare match - rare showing of resilience by stockfish

topic: a very rare match which doesn't happen often - a rare showing of resilience by a chess engine stockfish like a fairy tale like fish and the little mermaid







from the diagram,
fish is white
and the yellow line in the graph is fish

houdini is black
and the brown line in the graph is houdini

fish was losing more and more in "a"

at the start of phase "b"

fish was losing so much with the point of -3.0

ppl may think okay that's it
the match is totally over
a loss for the fish

but, at the end of "b"
and at the start of "c"
fish was only losing by -0.7 point
from houdini's eval

so I thought, fish can have a draw!
after trailing behind by such a big score

***

so that's the story of resilience of fish

but at the end of "c"
fish lost...
I thought fish could make a tie but...
lost

but the journey is more important than the destination!!!
the process is of importance!
even if fish lost, the fact that fish tried hard to overcome the phase "b" is remarkable!!!  

@thorhammer wrote:

topic: a very rare match which doesn't happen often - a rare showing of resilience by a chess engine stockfish like a fairy tale like fish and the little mermaid







from the diagram,
fish is white
and the yellow line in the graph is fish

houdini is black
and the brown line in the graph is houdini

fish was losing more and more in "a"

at the start of phase "b"

fish was losing so much with the point of -3.0

ppl may think okay that's it
the match is totally over
a loss for the fish

but, at the end of "b"
and at the start of "c"
fish was only losing by -0.7 point
from houdini's eval

so I thought, fish can have a draw!
after trailing behind by such a big score

***

so that's the story of resilience of fish

but at the end of "c"
fish lost...
I thought fish could make a tie but...
lost

but the journey is more important than the destination!!!
the process is of importance!
even if fish lost, the fact that fish tried hard to overcome the phase "b" is remarkable!!!  

Thor, which version of Stockfish did you use? 090713?

Yup stockfish is hopeless in other words (depends on the version)

Vocal.. plz dont quote big post like that, it make my eye bleed.

VocalTechnique wrote:

Thor, which version of Stockfish did you use? 090713?

of the 4 revisions
I used the latest one!  

No Opening Book Advantages! (Put in by Humans!?) No End Game Data Bases (Put in by Humans!?)

Pure!... Computer Chess from the Start!  


https://www.chess2u.com/t7666-the-arb-chess-system-chess-problem-04-10-2013



A.R.B

VocalTechnique wrote:

Thor, which version of Stockfish did you use? 090713?

I think I misunderstood your question!  

I thought you were asking
what revision I used for the tourney I ran in another thread

of the 4 revisions of SF Sep 7th
I used the latest one!

but you were asking what version I used for this thread's post!  

I used the latest one Oct 9  

it is my subjective opinion that
we have some strongest engines our intelligence can brag about for our bragging rights

after watching hundreds of matches between
houdini and the SF Oct 9th

it is my opinion that
houdini is best in eval
the eval of houdini is the most accurate I think
so when I do analysis the eval and the score and the board positions
I use houdini for chess learning and analyzing methods

but it is my opinion that
SF is best in some things like
- the attack
- I'm not sure about the defence of SF
I have a mixed opinion on defence
- positional play & positional awareness
- maticulous small scale tactics in the short run
- perspicacious large scale strategies in the long run
- combinational moves among chessmen


***

for stockfish,
as for the attack
it is extremely powerful and pushing forward
for so many years of watching best engine Vs. best engine matches
I have never seen an engine like SF Oct 9
which can have such attacks on opponent
which has no way but to retreat when SF attacks it
with such pushing and ramming power
it can search for mate very accurately and aggressively
while the general pace at which SF attacks is super duper hyper aggressive

as for the minute tactics and grandiose strategies
SF is very good at tactics
and I have never seen such unimaginable tactics before
and those tactics are done by SF
those tactics are simply unbelievable
and SF can make us of pawn pushing and pawn promoting after advancing very well

SF can confine opponent's pieces

for example,
it is impossible to find the past matches done
in the pgn file
so I tried my best to reminisce about the board position and the match
and I recreated what I witnessed

the following matches were done between SF and houdini
and these pictures are the recreation of what I remember
the match situation was something like the following
something similar like the following
(of course I can't remember exactly how the board position or FEN
looked like
but the following is the recreation of the match from my memory)








***

update:
(elaboration)

I wrote that in my subjective opinion
I was not sure about SF's defence

after watching hundreds of matches between SF Oct 9 and houdini

it is my subjective opinion on defence
that H's attack is generally more powerful than SF's attack

but here is something to note

during the opening and the middle of the match

I think SF's attack is generally weaker but
SF can search deeper
and SF can search for mate and good attacking points which are pivotal
so since SF is adapt at searching for tactical moves and strategical spots
when SF finds a good turning point then SF starts leading the match

but in the ending
it is my opinion that
H seems more powerful
and H is deemed winning a lot in the ending phase
H's ending play is solid and sound I think

***

so here is my analogy on H and SF

if 2 friends play a boxing for practicing
one guy (A) is heavy and
has a very powerful punch which can knock down another with one blow

another guy (B) is lean and light
and "B" doesn't have any poweful punches he can make use of

but the heavy guy "A" can be tired so fast only after 2 minutes of playing
cuz he is aggressive and he only concentrates on one powerful knock out punch

and "B" doesn't get tired easily even after 10 minutes of playing
and "B" can last for a long time
(if he can avoid the devastating one knock out punch from A)
without getting tired

so "A" has a good chance to win early in the match (if we say
the total boxing match will last for 12 minutes)

but if "B" can avoid the knock out punches from A
and if B can survive the first a few minutes (like 6 minutes)
then B has a good chance to win

@thorhammer wrote:

***


***

so here is my analogy on H and SF

if 2 friends play a boxing for practicing
one guy (A) is heavy and
has a very powerful punch which can knock down another with one blow

another guy (B) is lean and light
and "B" doesn't have any poweful punches he can make use of

but the heavy guy "A" can be tired so fast only after 2 minutes of playing
cuz he is aggressive and he only concentrates on one powerful knock out punch

and "B" doesn't get tired easily even after 10 minutes of playing
and "B" can last for a long time
(if he can avoid the devastating one knock out punch from A)
without getting tired

so "A" has a good chance to win early in the match (if we say
the total boxing match will last for 12 minutes)

but if "B" can avoid the knock out punches from A
and if B can survive the first a few minutes (like 6 minutes)
then B has a good chance to win

This is a very elegant way to describe Houdini 3 and Stockfish! Thank you. Btw, I think:

Houdini = A
Stockfish = B

I Will give you a Much Better Way to Describe the Differences between Stockfish & Houdini

Houdini is Stagnant! (Not Developing Fast Enough!)

Stockfish is Developing at the Speed of Light! (Soon to be Ahead of Every Other Program on the Planet!)  


Stockfish 2014 Will be the Most Incredible Program Known to Man & Woman (+ The Other Machines!)  



Anthony.R.Brown (Prediction for 2014!) +

let's just assume that
this is SF's most important match in the entire history of stockfish!

at nTCEC
professional computers experts
and chess analysts
along with GMs are talking about SF

SF must win this to go thru the next stage

if SF loses
SF's chance to be qualified is very slim

but the engine's PV and eval is becoming
in favor of SF

SF is on the brink of elimination and disqualification
and this stage is drawing to a close

ppl may say this match is the most important match for SF!


***







SF Vs. houdini

I ran a simulated engine match
in this position
from move number 34
to predict which will win

so from the move number 34
SF Vs. H

SF is replaced with newer version
and H dev version is stronger than H3
but the engines I used are comparable to the engines
used in the tourney kind of


***


this is the FEN I used

2q2bk1/5pPp/3p4/3PpbP1/3r4/2NP4/1P1Q3P/1KBR4 b - - 0 35

from this position I ran the SF vs H simulation

to find out which will win

the TC is long TC
a long TC is used
for this simulation

this is the pgn



[pgn]
[Event "Computer Chess Game"]
[Site "USER01"]
[Date "2013.11.04"]
[Round "-"]
[White "Houdini 3 Pro w32"]
[Black "Stockfish 021113"]
[Result "0-1"]
[TimeControl "10/60"]
[FEN "2q2bk1/5pPp/3p4/3PpbP1/3r4/2NP4/1P1Q3P/1KBR4 b - - 0 1"]
[SetUp "1"]

{--------------
. . q . . b k .
. . . . . p P p
. . . p . . . .
. . . P p b P .
. . . r . . . .
. . N P . . . .
. P . Q . . . P
. K B R . . . .
black to play
--------------}
1... Kxg7 {+2.38/20 10} 2. Qe3 {-1.06/17 7} Be7 {+2.66/20 2.6} 3. h4
{-1.27/18 21} Qa6 {+2.78/19 3} 4. Kc2 {-0.92/17 5} Bd8 {+2.63/22 7} 5. h5
{-1.61/17 8} Kg8 {+2.63/20 5} 6. Rd2 {-1.31/17 2.8} Ba5 {+2.62/21 4} 7. h6
{-1.58/17 5} Kf8 {+2.88/21 7} 8. Qe2 {-1.65/16 4} Bg6 {+3.23/20 7} 9. Qf1
{-2.11/15 2.8} Qc4 {+3.43/21 4} 10. Qh3 {-2.15/17 3} Rxd5 {+3.13/22 7} 11.
Kd1 {-2.32/17 8} Rxd3 {+3.19/22 11} 12. Rxd3 {-2.31/16} Bxd3 {+3.19/21 4}
13. Bd2 {-2.03/19 6} Ke8 {+2.98/21 6} 14. Ke1 {-1.89/18 13} Bd8
{+3.21/21 8} 15. Qf3 {-1.85/17 11} Qh4+ {+2.90/20 7} 16. Kd1 {-1.85/15} e4
{+3.21/21 3} 17. Qg2 {-1.87/17 5} Qh5+ {+2.76/22 10} 18. Ke1 {-1.74/18 4}
f5 {+3.15/23 6} 19. Nd5 {-2.24/17 5} Bc2 {+3.09/20 1.9} 20. Qe2
{-2.21/19 6} Qh1+ {+3.25/21 3} 21. Qf1 {-2.21/18 0.5} Qxf1+ {+3.39/23 3}
22. Kxf1 {-2.00/1} Kf7 {+3.77/24 5} 23. b4 {-2.01/19 9} Bb3 {+3.91/24 4}
24. Nf6 {-2.22/21 17} Kg6 {+4.06/24 11} 25. Be3 {-2.27/20 5} d5
{+4.06/23 5} 26. Ke1 {-2.76/16 10} d4 {+4.48/22 4} 27. Bxd4 {-2.76/19 8}
Kxg5 {+4.78/24 8} 28. Nxh7+ {-2.75/19 3} Kxh6 {+5.01/25 5} 29. Nf8
{-3.01/19 7} f4 {+5.35/24 7} 30. Nd7 {-3.17/18 2.4} Kg6 {+5.71/25 8} 31.
Nc5 {-3.64/20 14} Bd5 {+6.42/22 3} 32. Na4 {-4.09/20 16} e3 {+8.04/21 1.8}
33. Nc3 {-5.79/18 8}
{User adjudication} 0-1
[/pgn]



SF wins in the simulation


***

it is move # 36

-3.7
-2.5

both evals are in favor of SF

SF wins and that means something of importance

Houdini 9601 is amazing
Houdini 9601 version is undefeated and invincible
and SF is gonna be the first engine to win over houdini 9601
along with SF will move on to the next stage in nTCEC