(;GM[1]FF[4]CA[UTF-8]AP[CGoban:3]ST[2]
RU[Japanese]SZ[19]KM[6.50]TM[3600]OT[3x60 byo-yomi]
PW[Zen19X]PB[miao]BR[6d]DT[2016-08-28]PC[The KGS Go Server at http://www.gokgs.com/]C[Zen19X [-\]: GTP Engine for Zen19X (white): Zen version 12.0
]RE[B+Resign]
;B[pd]BL[3594.016]C[miao [6d\]: hi 
miao [6d\]: gg
pookpooi [?\]: it begin
]
;W[dp]WL[3570.869]C[GoldenBear [1d\]: Good luck to Zen and Lukas! Have fun!
]
;B[pq]BL[3583.684]C[gghideki [?\]: gg
gogonuts [3d\]: generous byoyomi
]
;W[dd]WL[3532.794]C[pookpooi [?\]: is zen set to 30 sec per move?
r16 [4k\]: or is it variable according to situation?
]
;B[ql]BL[3543.069]C[KlausB [?\]: Is the lomg byoyomi better for the Humans or the Computer??
GoIngo [?\]: Here we go. Good luck to both players!
GoldenBear [1d\]: And as usual a short word from the sponsor: The challenge will be documented on go.codecentric.de. And you are invited to visit www.codecentric.de, too :-)
gogonuts [3d\]: longer times are better for the human
pookpooi [?\]: meanwhile in another room 9d vs 9d
gogonuts [3d\]: although the bot improves too
GoIngo [?\]: We use the timing of LSD-vs-AlphaGo
]
;W[qf]WL[3489.259]
;B[qe]BL[3538.161]C[polipor: 12.0?
GoldenBear [1d\]: Also, interviews with the players are online at https://blog.codecentric.de/en/2016/08/codecentric-go-challenge-2016/
polipor: :(
pookpooi [?\]: timing of lsd is 2 hrs.?
gghideki [?\]: yes, v12.0
snowmen [1d\]: long time short game
]
;W[pf]WL[3461.249]C[GoIngo [?\]: 1 hour plus 3*60 secs byoyomi
]
;B[nd]BL[3531.33]C[gogonuts [3d\]: GB, is that link brand new?
gogonuts [3d\]: couldn't find anything today
GoldenBear [1d\]: Yes.
GoIngo [?\]: yes
GoldenBear [1d\]: Today
gogonuts [3d\]: thx
gghideki [?\]: v12.1 uses 4 GPUs!
]
;W[pi]WL[3430.88]C[GoldenBear [1d\]: One hour ago.
miao [6d\]: Zen knows adanced joseki :)
]
;B[fq]BL[3502.623]C[gogonuts [3d\]: pro level sofar :-)
Kotonoha [2d\]: pro could play around 5 hours with this time setting ( fully use of time suji ) 
pookpooi [?\]: why 12.1 isn't being used here again? can someone clarify this?
gogonuts [3d\]: b threatens p10, so I expect a w pincer
snowmen [1d\]: Why is this game not rated?
]
;W[cn]WL[3385.711]C[Kotonoha [2d\]: this seems some event 
]
;B[jp]BL[3486.61]C[snowmen [1d\]: evening event
GoIngo [?\]: it is part of the negotiations
]
;W[po]WL[3365.086]C[GoIngo [?\]: by the way, Zen19K ws not expected to be ready already
miao [6d\]: direction is good !
IA [8d\]: o.o
GoIngo [?\]: fj is watching - welcome
fj [?\]: hi
pookpooi [?\]: all star kgs account is here ;)
GoIngo [?\]: also: welcome Aja!
miao [6d\]: the usual follow up is p10 for black insteach of f3
gogonuts [3d\]: was k4 a little passive?
asdewq [?\]: "why 12.1 isn't being used here again? can someone clarify this?"
miao [6d\]: now it is too dangerous
gogonuts [3d\]: w q5 looks unwelcome for b
Kotonoha [2d\]: pookpooi [?\]: is zen set to 30 sec per move? < C6 use up around 75 seconds
pookpooi [?\]: ok kotonoha
GoIngo [?\]: It was negotiated that the "old" Zen is playing (without GPU)
gogonuts [3d\]: I expected that p10 instead of k4
pookpooi [?\]: thanks, ingo
pookpooi [?\]: do you think remi will appear too?
YoungPro [6d\]: is zen19x stronger than zen19k?
thepok2: how old is old?
]
;B[np]BL[3364.283]C[GoIngo [?\]: maybe in anonymity ;-)
gghideki [?\]: No, max time is 60s
pookpooi [?\]: ;)
miao [6d\]: o4 is the most challenging move
GoIngo [?\]: but who knows
pookpooi [?\]: so internet connection problem?
]
;W[no]WL[3331.671]C[Kotonoha [2d\]: sorry 45 :-) 
]
;B[mo]BL[3355.941]C[gghideki [?\]: If a move is prefered so much, it is played very earlier
miao [6d\]: the outcome here is rather unclear
GoIngo [?\]: Some comments on Joseki from the watching experts?
]
;W[nn]WL[3302.202]C[gogonuts [3d\]: is zen time management 60s before byoyomi too?
]
;B[op]BL[3349.972]C[YoungPro [6d\]: so far bad for b
ibd2 [?\]: R7?
YoungPro [6d\]: k4 was mistake indeed
gogonuts [3d\]: i blame k4
YoungPro [6d\]: m4 was better than k4,considering that you have r8 and q11 shape
]
;W[ol]WL[3261.158]C[GoIngo [?\]: So, we have two blamers already
gogonuts [3d\]: this looks pretty awful
YoungPro [6d\]: you can open pro database or 9d tygem database,its obvious
pookpooi [?\]: will there be next year codecentric challenge if zen win?
snowmen [1d\]: r8 can live
ibd2 [?\]: W seems playful today
YoungPro [6d\]: its not about being alive or dead..
GoIngo [?\]: we will decide after the match
YoungPro [6d\]: this result is just so terrible for b
pookpooi [?\]: ;)
gogonuts [3d\]: I already had a bad feeling when b got fancy with q8, but that was ok of course
gogonuts [3d\]: with r8, i mean
pookpooi [?\]: I still don't know the difference between zen 11.9 and 12.0
chaos0 [4d\]: zen19X  19K  19A, which is the best??
gogonuts [3d\]: 0.1
polipor: how strong is zen 12.0 compared to 12.1?
Kotonoha [2d\]: gghideki: I want to know too ^ ^
GoIngo [?\]: Difficult to compare, because Zen19X and Zen19K use different hardware
pookpooi [?\]: so like 12.0 is the 11.9 on new hardware? and 12.1 add value network?
gghideki [?\]: Zen19A and K are run by Yamato and Zen19X is run by me;  that's all
gogonuts [3d\]: letting r8 just get swallowed seems too awful - r10, maybe?
Kotonoha [2d\]: this topic seems like when I were child talking about how strong the character 's fighting power in DragonBall lol 
GoIngo [?\]: lol
]
;B[kc]BL[3147.696]C[gogonuts [3d\]: my power ranger is stronger than yours
GoIngo [?\]: like in Pokemon Go?
Adriano [-\]: over9000
pokamon [2k\]: :-)
gghideki [?\]: clearly 12.0 is stronger than 11.9
miao [6d\]: i aim on daring white to invest another move
Kotonoha [2d\]: well for the game it seems YP is talking about right
]
;W[qn]WL[3220.19]C[gogonuts [3d\]: L17 is very important, but I don't think b had the time
fluislost [?\]: zen19K  is 12.1  currently on kgs
Kotonoha [2d\]: he behaves so serious when there is an admin ^ ^
snowmen [1d\]: SE prefers white by 38
fluislost [?\]: on better hardware
gogonuts [3d\]: now r8 looks cooked
gghideki [?\]: Zen12.1 is the first version using "value network"
Jazzman [1d\]: well, w did invest anotehr move
]
;B[cf]BL[3120.042]C[chaos0 [4d\]: I want to see alphabot v20 play in KGS someday
fluislost [?\]: nice Hieki
Guest2d [1d\]: Is it two hours for each game? 1 game per day for 5 days?
GoIngo [?\]: one game per Sunday
miao [6d\]: i will try to play at many parts of the board and later use r8
miao [6d\]: s12 is still nice aji
Guest2d [1d\]: ok thx
miao [6d\]: also r4 is almost sente
]
;W[fc]WL[3180.385]C[gghideki [?\]: v12.1 uses, sadly for me, 4 GPUs!
]
;B[bd]BL[3117.077]C[pookpooi [?\]: titan gpu
gghideki [?\]: yes, TITAN X
Kotonoha [2d\]: painful to watch miao talking in the game like rsun ^__^ 
gogonuts [3d\]: lol, koto
pookpooi [?\]: but 12.0 has better cpu
]
;W[ic]WL[3148.638]C[gghideki [?\]: this Zen19X uses one GTX 960 which is much cooler :)
GoIngo [?\]: Is Zen19K taking a break?
First [1k\]: why is miao playing
gghideki [?\]: and silent
Jonsoncao [4k\]: SISUplayer is here
gghideki [?\]: ah, really
]
;B[dq]BL[3070.128]C[Jonsoncao [4k\]: is he gonna play against zen?
pookpooi [?\]: 150 observer now
snowmen [1d\]: this game looks like himan game
gghideki [?\]: looks disconnected
miao [6d\]: so far it is hard to say white made any real mistakes
GoldenBear [1d\]: I don't know if am thankful for Lukas commenting life on his game or if I want to hit him to concentrate on the game instead :-)
]
;W[cq]WL[3111.786]C[miao [6d\]: whereas black c14 and o4 might be a bit dubious
]
;B[cr]BL[3063.445]C[First [1k\]: commenting your own games is rude
pookpooi [?\]: not!
Epan: Crazy stone dl gives 46 % to black.
First [1k\]: you should really have anti comment feature
First [1k\]: for players
]
;W[eq]WL[3095.151]C[GoIngo [?\]: I would hint him, but can you?
gogonuts [3d\]: lukas is steadying his nerves :-)
]
;B[dr]BL[3060.768]C[Guest2d [1d\]: P8 is kind of bot move
KlausB [?\]: well, the opponent isn't listening anyway ...
gogonuts [3d\]: players can comment, but not read - very sensible
GoldenBear [1d\]: No, there is no way I have to reach Lukas. And we should't.
]
;W[fp]WL[3069.976]C[miao [6d\]: d3 secures the lower side and takes eom corner
]
;B[er]BL[3058.307]C[pookpooi [?\]: hideki, will deepzengo project challenge pro in even game soon?
]
;W[ep]WL[3055.17]
;B[gp]BL[3057.112]C[GoIngo [?\]: you mean evening game?
miao [6d\]: b3 g5 will follow
pookpooi [?\]: the goal is within 1 year to catch up nature alphago?
GoIngo [?\]: ;-)
]
;W[go]WL[3040.299]C[miao [6d\]: then w has sente to take c12
]
;B[hp]BL[3054.483]
;W[ch]WL[3025.46]C[miao [6d\]: i don´t particularly like g5, but it is also ok
KlausB [?\]: Goingo: the game need not be in the evening ,,,
]
;B[cc]BL[3040.98]C[Epan: 44%
flobby [-\]: is zen as good as october alphago yet?
GoIngo [?\]: ok, maybe I missunderstood the question
]
;W[ce]WL[3010.004]C[miao [6d\]: c4 is a big move
pookpooi [?\]: I'd prefer the game to be in saturday though
gogonuts [3d\]: zen needs to win on 2 stones first, got unlucky last try
pookpooi [?\]: I don't think zen19k can win against fan hui in 1 hr per side game yet
]
;B[be]BL[2999.439]C[GoIngo [?\]: @pookpooi: Lukas has other obligations on Saturday
snowmen [1d\]: Zen already won by 2 stones against pro
chaos0 [4d\]: wow, 1 hours??
]
;W[df]WL[2995.246]
;B[dg]BL[2997.343]C[Meruem92 [3d\]: @flobby: no
pookpooi [?\]: thanks goingo, I appreciate his participate
GoIngo [?\]: Yes, we are happy that he is standing like a man
]
;W[cg]WL[2964.27]C[miao [6d\]: i need to be careful of the center soon though
]
;B[bf]BL[2984.832]C[GoldenBear [1d\]: This is a very fast game compared to what we saw last years.
pookpooi [?\]: he has his pride as german champion too
pokamon [2k\]: zen19k looks like the bot managed to play fierce ko fights, i hop miao test this version for his ko abilitys
gghideki [?\]: Yes, this version won against Cho Hye-Yeon 9p
gghideki [?\]: with 2 hc
gogonuts [3d\]: i think this position is objectively good for w, but what is even worse, is that it's bot-freindly
miao [6d\]: e14 will give me the invasion at d10
chaos0 [4d\]: zen 6 gives 60% to white
miao [6d\]: also d7 is not a bad move to make sabaki
]
;W[ef]WL[2920.955]C[Epan: crazy stones gives 43% to black.
pookpooi [?\]: miao is waiting for game 4 miracle like lee sedol
gogonuts [3d\]: lol
snowmen [1d\]: This game is clearly good for white
Gozira [16k\]: Pachi bot makes same moves as Zen :) 
Gozira [16k\]: interesting 
]
;B[dj]BL[2922.695]C[gogonuts [3d\]: i think they play to 3 wins
gghideki [?\]: But Zen's MC simulation has several weakpoints;  B will have chances, I'm sure
gogonuts [3d\]: so no game 4 miracle
]
;W[ei]WL[2878.949]C[gogonuts [3d\]: miracle has to come earlier
iceweasel [-\]: hideki predicted 3-1 or 3-0
tonkachi [13k?\]: lee sedol gam 4 wasn't a miracle
]
;B[ej]BL[2909.224]C[Gozira [16k\]: e12 hmm,,, now different 
GoIngo [?\]: not predicted bujt expected
tonkachi [13k?\]: was a display of pure genius
gogonuts [3d\]: was a bot blunder
]
;W[fi]WL[2850.526]
;B[ci]BL[2901.372]C[gogonuts [3d\]: lsd tricked the bot
tonkachi [13k?\]: exactly
miao [6d\]: i am trying to sabaki here
pookpooi [?\]: I'm reading codecentric interview now
mafutrct [-\]: great, we agree
GoIngo [?\]: feel free to ask on it, pookpooi
pookpooi [?\]: ty ingo
miao [6d\]: and wondering about s12
]
;W[di]WL[2800.345]C[miao [6d\]: s12 is at least ko
]
;B[bi]BL[2889.649]C[GoIngo [?\]: GoldenBEar asked the questions
miao [6d\]: this is very unpleasent to disconnect
snowmen [1d\]: Zen got tricked
miao [6d\]: for white
]
;W[ck]WL[2766.379]
;B[cj]BL[2884.45]C[GoldenBear [1d\]: That is true. But it seems I did not understand all the answers :-)
miao [6d\]: now i am connected
GoIngo [?\]: no problem
miao [6d\]: but still white has gained thickness
pyramids2 [1k\]: What codecentric interview are you talking about?
GoIngo [?\]: connected to KGS?
miao [6d\]: never underestimate thickness
GoldenBear [1d\]: https://blog.codecentric.de/en/2016/08/codecentric-go-challenge-2016/
GoIngo [?\]: GoldenBear gave the link above
]
;W[mf]WL[2721.342]C[Guest2d [1d\]: no, the group
pyramids2 [1k\]: thanks, GoldenBear
pookpooi [?\]: lol at goldenbear and goingo
GoIngo [?\]: I see
snowmen [1d\]: black catched up a lot
pookpooi [?\]: teamwork chat
gogonuts [3d\]: zen didn't get tricked - its already moving towards safe mode
pokamon [2k\]: b connect
GoIngo [?\]: Yes, we are funny from time to time
chaos0 [4d\]: 19k gives a diffenrent result compare with 6
GoIngo [?\]: moyoesk?
miao [6d\]: i would never come up with a move like this
miao [6d\]: but i don´t understand moyos
miao [6d\]: my upper side is nor exeptionally big
trohde [?\]:  BTW, to whoever it was who said it is rude to comment one’s one’s moves during the game: NOTHING is rude when fighting a robot ;-)
asdewq [?\]: zen says is 29% for black now.. miao already losing grip :)
miao [6d\]: nor easyly convertable into points
GoIngo [?\]: +1 trhode
asdewq [?\]: *39
platinum [5k?\]: Who is he talking to 
Jazzman [1d\]: to all of us
gogonuts [3d\]: n14 denies s12?
GoIngo [?\]: to us
bitzi [-\]: he seems to be wasting time by chatting
platinum [5k?\]: For gods sake
miao [6d\]: second of all it would be a huge gain for white to play a move like k15 and connect his stones
platinum [5k?\]: We don't need to know anything 
gghideki [?\]: 62% for Zen 
pookpooi [?\]: many people will e-cheating using zen!
chaos0 [4d\]: no, still 40%
miao [6d\]: my first instinct is k16
Guest2d [1d\]: w lower left is not in solid shape
platinum [5k?\]: Just play or resigm
GoIngo [?\]: thx hideki for score
miao [6d\]: but if both keep pushing white make some points and i still have a weakpoint at n17
pookpooi [?\]: 62 is quite high
sjh [9k?\]: where can you see zen's evaluation?
gogonuts [3d\]: people will be 9d at e-cheating :-)
hikgs: just do it, pls
miao [6d\]: that is very hard to adress
GoIngo [?\]: hideki is operating zen
pookpooi [?\]: hideki is the zen programmer, he can see everything
sjh [9k?\]: ah
miao [6d\]: m15 is adressing both black issues in a way
pookpooi [?\]: I just stated echeating cause lukas interview give an interesting point about this
miao [6d\]: but after r17 i again  have a hard choice
polipor: i don't get the point of cheating....
asdewq [?\]: looks like b is doing better, but Zen's confidence of 62% is astonishing
miao [6d\]: i might lose aji with r8 or 
gghideki [?\]: This version plays similar to Tencho 6 :)
miao [6d\]: the top 
polipor: if you use zen to play, you don't play, it's zen
gogonuts [3d\]: looks like w is doing better, because w is doing better
GoIngo [?\]: Polipor, it is sort of addictive
GoIngo [?\]: once you tested it yoou can not stop
snowmen [1d\]: Zen is echeating
GoIngo [?\]: ;-)
gogonuts [3d\]: lol, snowmen
NiceGoAuth [-\]: sorry, I did not get everything, has thismversion a value net, and howmdeep?
pookpooi [?\]: what is dgob
miao [6d\]: s12 is thus an interesting idea
First [1k\]:  'b far behind
GoldenBear [1d\]: I suppose it must be relaxing for Lukas to keep up this chatter. It might be for me. But it still strange that he is doing it on his own time.
miao [6d\]: but also a bit risky
GoIngo [?\]: dgob = Deutscher Go Bund
GoIngo [?\]: german go association
pookpooi [?\]: oh
platinum [5k?\]: Polipor has tested everything he doesn't need anymore
pookpooi [?\]: ok
GoldenBear [1d\]: There is a glossary at the end of the interview.
Guest2d [1d\]: b has to do something for r8
pookpooi [?\]: but dutch and german is different country
CrazySt0ne [8d\]: what is this game about?
gghideki [?\]: NiceGOAuth: This version (v12.0) doesn's use value network
NiceGoAuth [-\]: thanks hideki!
gghideki [?\]: v12.1 (current Zen19K) uses it.
gogonuts [3d\]: https://blog.codecentric.de/en/2016/08/codecentric-go-challenge-2016/
miao [6d\]: extremely risky
NiceGoAuth [-\]: yes, Imread his, do you say how deep?
miao [6d\]: maybe i waited a bit long to play s12
gghideki [?\]: Playing style is soooooooooo different, really
miao [6d\]: it is a huge move
pyramids2 [1k\]: CrazyS0ne: Short summary of my understanding: This is an invitation-only computer-human go challenge sponsored by an IT firm.
polipor: gogogo miao
GoIngo [?\]: we just gots 200+ spectatrors
polipor: play it
flobby [-\]: what's zen's % now?
polipor: s12
CrazySt0ne [8d\]: thx
gghideki [?\]: how deep?  Number of layers?
pookpooi [?\]: omg
gghideki [?\]: I remeber 16
NiceGoAuth [-\]: y and firs
gghideki [?\]: 192, maybe
pookpooi [?\]: will aja huang surprise us with alphago winrate?
First [1k\]: b behind a lot
gghideki [?\]: The network is designed by Yamato and so I have no sure numbers
First [1k\]: yamato is a samurai
lofuchi [3k\]: it is a battleship isn't it :P
pookpooi [?\]: yamato is rarely seen in public
]
;B[qp]BL[2344.988]C[NiceGoAuth [-\]: ok, thx
KlausB [?\]: Yamato was a battle ship ...
pookpooi [?\]: hideki is like doing representative duty too
privatePi [2d?\]: quite slow move, now white will invade top
miao [6d\]: i chose s4 to create some more aji and ask white where to continue
sh [-\]: Who is he taking to
sh [-\]: Disgusting
miao [6d\]: also it is a doubble sente in a way
]
;W[mc]WL[2674.444]C[KlausB [?\]: himelf, mostly ...
snowmen [1d\]: chating while playing became popular because of haylee videos
]
;B[md]BL[2336.947]C[Hipparion [-\]: miao is commenting is own game as he goes
GoIngo [?\]: ;-)
miao [6d\]: this is a move i kinda like to see
lofuchi [3k\]: but he couldn't see us talking right?
]
;W[lc]WL[2657.54]C[pookpooi [?\]: it's long before haylee video
Rappelll: because he knows people are listening
Hipparion [-\]: no
GoIngo [?\]: right
miao [6d\]: it reduces me a lot
backpack1 [3d\]: i think its very interesting to follow the players thoughts
TheParrot [1k\]: he is talking to me
]
;B[ld]BL[2330.682]C[Hipparion [-\]: can't see us talking
GoIngo [?\]: we are silent for him
backpack1 [3d\]: much more interesting then just a winrate ;) 
lofuchi [3k\]: perhaps he has another account
First [1k\]: does he think he is a famous streamer?
]
;W[kb]WL[2642.771]C[First [1k\]: yes I know that is an oxymoron
Hipparion [-\]: this is quite rare, to have the thoughts of a player while he is playing
martin97 [-\]: first, he is but inactive
miao [6d\]: but also solidifies my corner
pookpooi [?\]: he should stream himself while playing in youtube
]
;B[kd]BL[2294.718]C[pookpooi [?\]: but he might say in german
Cybercage [5k\]: Tbh I domt think he is doing this for getting ratio. Its more like taking notes
lofuchi [3k\]: the corner now is very small
lofuchi [3k\]: and gote
Rappelll: but middle is weaker now
]
;W[oc]WL[2602.447]C[miao [6d\]: the center is not not a big concern anymore
lofuchi [3k\]: oh yes this is possible
lofuchi [3k\]: i thought must k18
First [1k\]: dear diary
]
;B[od]BL[2256.594]C[asdewq [?\]: 65 for Zen now
First [1k\]: today I played against zen19x
ohyou [-\]: could somebody enlighten me if the Zen configuration changed ever since the last match? or it is the game?
ohyou [-\]: the same*
]
;W[pb]WL[2565.989]C[sh [-\]: Cya
pookpooi [?\]: is score estimator reliable at this point
fluislost [?\]: this zen is an "old" version.  12.1 is out and is being tested as Zen19K
miao [6d\]: is this move helping white so much?
lofuchi [3k\]: so b has nothing left in the corner
Guest2d [1d\]: zen is in attacking mode
GoIngo [?\]: But Zen12.0 is cleary better than Zen from September 2015
First [1k\]: if upgrade not work
lofuchi [3k\]: in contrary to b's expectation
First [1k\]: have to downgrade
First [1k\]: sometimes downgrading improves your game
jhtf: but prevented w to make big center
Jazzman [1d\]: hmm, somehwow, i dont thing b has enough...
lofuchi [3k\]: b even doesn't have 2 eyes, isn't it
lofuchi [3k\]: i think it just helps w making center territory
chaos0 [4d\]: gg
Guest2d [1d\]: exactly
]
;B[nc]BL[2152.928]C[pookpooi [?\]: will mcsts bot (without dcnn) has no future now, even with mcts evolution knowledge-based bot manage to crawl up from 8k to 5k
jhtf: left not big for w top too flat right side is big b has bottom big i think close
pookpooi [?\]: I want to see 9d mcts bot
]
;W[nb]WL[2530.642]
;B[pc]BL[2144.505]C[pookpooi [?\]: dcnn is like cheating
GoIngo [?\]: ??
redreoicy [4d\]: possible with some kind of improvement to mcts maybe
pyramids2 [1k\]: dcnn is merely like adding intuition
gonmf [-\]: i like how you think
]
;W[ob]WL[2511.929]C[pyramids2 [1k\]: not cheating, but greatly improving
]
;B[jb]BL[2133.937]C[pookpooi [?\]: yeah, I'm wrong
pookpooi [?\]: greatly improvement
KlausB [?\]: what stands dcnn for?
GoIngo [?\]: convoblabla
pyramids2 [1k\]: deep, convoluted neural network
GoIngo [?\]: I learned this afternoon
tjl [2k\]: *convolutional*
]
;W[qc]WL[2469.354]C[snowmen [1d\]: hm
GoIngo [?\]: the original name is too long for such an important concept
lofuchi [3k\]: why do they give such name ...deep, convoluted :)
pyramids2 [1k\]: a neural network with position-invariance built-in to reduce the number of weights (variables)
]
;B[lb]BL[2115.178]C[pookpooi [?\]: i like mcts alienate style
gogonuts [3d\]: b r18 was better than k18?
lofuchi [3k\]: a very arrogant name :P
asdewq [?\]: 70% now, miao missing all crucial points
]
;W[mb]WL[2434.65]C[GoIngo [?\]: asdewq: from which bot
lofuchi [3k\]: why not just call it multi layer neural network
]
;B[ka]BL[2112.137]C[asdewq [?\]: zen6
pyramids2 [1k\]: a descriptive, scientific name, lofuchi: Deep = many layers, Convolutional = reduced in a way that more easily allows to have many layers
Tsuke [3d\]: lofuchi, there are neural networks, which typically only have 1-3 layers, but those with very many layers you call deep neural networks
GoIngo [?\]: thx
lofuchi [3k\]: yes i know
Tsuke [3d\]: convolutional, is because you perform a convolution at each layer
lofuchi [3k\]: but then it is just multi layer NN
lofuchi [3k\]: :P
]
;W[kf]WL[2395.525]C[lofuchi [3k\]: call it deep sounds very arrogant :)
YoungPro [6d\]: this is totally game over
lofuchi [3k\]: mayber worth it though given its power
Tsuke [3d\]: yp, who won?
]
;B[rc]BL[2089.936]
;W[rb]WL[2394.862]C[asdewq [?\]: goingo: zen 6 doesn't play that strong, but when you give it a position (for example alphago vs LS) it has the correct winrate...
]
;B[qd]BL[2086.551]C[YoungPro [6d\]: i dont see any chance for b
Tsuke [3d\]: topright w isnt alive tho
Tsuke [3d\]: b can ko it
Epan: Deep networks is rather new (last few years). For many years no one could figure out how to train nets with more than a few layers.
]
;W[qb]WL[2361.317]C[GoldenBear [1d\]: I like the idea that human players feel convoluted by bots using DCNN...
Rappelll: looks alive to me
Guest2d [1d\]: maybe b should attack lower left?
jhtf: nice ko
Tsuke [3d\]: rapelll, b can t18 for ko
gogonuts [3d\]: lukas must be feeling pretty convoluted right now
miao [6d\]: i have some connection issue will be bac asap
miao [6d\]: sorry guys
snowmen [1d\]: convolution = curve
pookpooi [?\]: ok
pokamon [2k\]: s15 now?
Tsuke [3d\]: no way b plays s15
KlausB [?\]: connection issues on the board as well as online ...
Tsuke [3d\]: safe t18 ko, or play it now
pookpooi [?\]: his connection is really convoluted
Tsuke [3d\]: save*
Tsuke [3d\]: but dont ruin it with a move like s15
pookpooi [?\]: omg, convoluted is now my word of the day
gogonuts [3d\]: deep connection issues
Meruem92 [3d\]: i'd play it now
redreoicy [4d\]: very hard ko though
Tsuke [3d\]: b might want to play h17 anwyays
gogonuts [3d\]: it's actually not funny, that zen is so easily mistreating lukas like this
redreoicy [4d\]: after losing ko black isn't alive, and gave white some 15 points
Tsuke [3d\]: if he wants to play h17, then no rush with the ko, can wait until ghave better threats
wurfmaul [3d\]: Epan, how to train nets with more than a few layers? I guess I cannot expect a simple answer
snowmen [1d\]: At least Zen will show how to play a Ko
lofuchi [3k\]: it appears to me that such NN becomes so convoluted, nobody can trace what it is doing anymore :P
pookpooi [?\]: how old is lukas?
ibd2 [?\]: W also about to seal a nice chunk of the center
Meruem92 [3d\]: about 24...maybe?
pookpooi [?\]: younger than fj!
pookpooi [?\]: about the same age as ingo?
Tsuke [3d\]: wurfmaul, multilayer nets isnt any different from 2 layer networks, you just propagete that error backwards, problem is the more layers you have the more data you need for it to avoid overfitting
ibd2 [?\]: Younger than Ingo :)
Christoph [2k\]: a bit
GoIngo [?\]: that is no problem, my son
Tsuke [3d\]: however, using convolution, you share the data in a way, since you have the convolution kernel, allowing you to get away with alot less data without causing overfitting
redreoicy [4d\]: ko is so hard that black cannot play it without local threats
gogonuts [3d\]: does b know he's still connected, what's goin gon?
pookpooi [?\]: the time is stop
gogonuts [3d\]: ah, gone now
wurfmaul [3d\]: thanks Tsuke
lofuchi [3k\]: will Zen get mad?
GoIngo [?\]: i think miao is in Bonnn
Guest2d [1d\]: b has killed a lot of aji's and made itself too heavy
lofuchi [3k\]: :P
GoIngo [?\]: is it hot there right now?
Rodtank [2d\]: miao needs to make zen have weak group to have chance to win
jhtf: miao is testing on the table 
GoIngo [?\]: no
lofuchi [3k\]: zen is learning why b doesn't move :)
ohyou [-\]: could black live with r8 stone somehow? maybe s11?
gogonuts [3d\]: b might have to crawl at j18 first
lofuchi [3k\]: i would play j18 but it looks so bad
Guest2d [1d\]: t16 and j18 miami
Rappelll: t 16 the better of those two moves anyway
lofuchi [3k\]: is miao coming back?
Schachus12 [3k\]: lofuchi, he wrote: miao [6d\]: i have some connection issue will be bac asap
fluislost [?\]: zen is gone too
privatePi [2d?\]: miao is starting his engines
lofuchi [3k\]: yes i saw that, but i now doubt :)
GOuroboros [2d?\]: thunderstorms in Bonn, Cologne predicted for tonight
lofuchi [3k\]: it is more than the time needed for a reboot
Tsuke [3d\]: does w have ponder on?
Tsuke [3d\]: er.. *ponder enabled*
Jazzman [1d\]: .
GoIngo [?\]: it seems we have a problem
GoIngo [?\]: we (the organizers) do not have a phone number of Lukas
gogonuts [3d\]: Houston ...
Tsuke [3d\]: if he does, w will have read quite far by the time miao gets back
ohyou [-\]: is the engine running while the game is paused like that?
lofuchi [3k\]: zen will read the whole truth of go :)
GoIngo [?\]: does someone have a phone number of Lukas - please send me in PM
fluislost [?\]: yellow pages are avaiable on the web.  if someone knows his real name and city
Varelse [-\]: lukas podpera
gosei [2d\]: what abou the game?
trohde [?\]: http://www.unwetterzentrale.de/uwz/index.html … quite heavy thunderstorms already in Germany (it was almost apocalyptic where I live, Northern Germany, 60 km South of Hamburg)
trohde [?\]: so … it may well be that he’s disconnected due to … water or lightning strike, etc.
gosei [2d\]: ok,so didnt resign...
polipor: connection problems he said
gosei [2d\]: thanks
GoIngo [?\]: currently no thunderstorm in Bonn ...
KlausB [?\]: trohde = Bonobo?
GoIngo [?\]: a friend just told me that Lukas seems to have his mobile switched off
martin97 [-\]: which is actually good during a serious game ^^
privatePi [2d?\]: maybe he didn't find the charger ^^
Neuneck [?\]: lol
miao [6d\]: sorry guys
miao [6d\]: im back
gosei [2d\]: yeah:)
gogonuts [3d\]: great :-)
starmie [4d\]: Hurray!
gosei [2d\]: just wonder howlong;)
GoIngo [?\]: I just tried to phone Rainer Schuetze ...
Ootakamoku [-\]: wonder if b should play a ko threat now...
Ootakamoku [-\]: to reset the tree created by w pondering
fluislost [?\]: miao is back
fluislost [?\]: look in cpu go room
martin97 [-\]: he is also back in game ;)
polipor: it's here, yes
]
;B[nf]BL[1701.634]C[fluislost [?\]: zen is missing
fluislost [?\]: who has HIdeki phone number
polipor: lol
Varelse [-\]: zen is here
polipor: no, it isn't
polipor: it's clock is stopped
trohde [?\]: KlausB: ;-)
gghideki [?\]: Zen19X is running normaly.  Sometihng wrong on Intenet connection
polipor: will zen resume the match?
polipor: oh
miao [6d\]: test
gghideki [?\]: I'll restart Zen
GoIngo [?\]: I just mailed Hideki that Lukas is back.
Zen19X [-\]: GTP Engine for Zen19X (white): Zen version 12.0
polipor: you can talk to him here ;-)
fluislost [?\]: zen ist bach
martin97 [-\]: quite some action apart from the board 
trohde [?\]: unexpected thrills :-D
GoIngo [?\]: apocalyptic scenario ?
]
;W[ng]WL[2326.982]C[GOuroboros [2d?\]: so this is natural intelligence vs AI, re-establishing a connection :)
]
;B[rf]BL[1693.136]C[fluislost [?\]: zen had to quit the game, else it would have thought for so long that he would have solved the position
gghideki [?\]: haha
GoIngo [?\]: reestablishing all connections ;-)
]
;W[rg]WL[2292.068]C[polipor: does zen uses previous calculations?
GoIngo [?\]: Puuh, all systems go again
fluislost [?\]: i guess so. kept into RAM
polipor: if so, it remembers now after reboot?
gghideki [?\]: Looks I can't sleep yet :(...  It's 2:10 am
]
;B[mg]BL[1667.308]C[Freund [1k?\]: also them from last year
]
;W[lg]WL[2282.157]
;B[sb]BL[1663.759]C[gosei [2d\]: lukas tries to create ko threats..
fluislost [?\]: hideki write script to auto connect in case of failure from human
fluislost [?\]: then u can sleep
gghideki [?\]: haha, right
]
;W[ra]WL[2234.876]C[GoIngo [?\]: you know the miracly of time zones?
gghideki [?\]: but how to detect disconnection?
]
;B[ma]BL[1653.419]C[GoIngo [?\]: miracle
snowman [-\]: hmm
gosei [2d\]: why nobody talks about the game?
GoIngo [?\]: Once I wrote a paper with a colleague from California
TblKBA [1k\]: f10 
Guest2d [1d\]: now this ko makes no sense for b
GoIngo [?\]: the game?
fluislost [?\]: good question hideki
]
;W[rd]WL[2190.806]C[Varelse [-\]: then something else will hapen therefore you can not sleep
]
;B[sd]BL[1648.907]C[gosei [2d\]: i think black is in trouble...
TblKBA [1k\]: w wins this
]
;W[sc]WL[2169.606]C[fluislost [?\]: with a script that checks the pixels of the clocks of both players.  if both pixels are not changing, then disconnection occured
starmie [4d\]: B is so behind
]
;B[og]BL[1646.984]
;W[re]WL[2154.801]C[gosei [2d\]: the invasion was too succesfull..
martin97 [-\]: you guys can write crazy strong go programs but not detect a deconnect? really? :D
]
;B[nh]BL[1641.93]C[ohyou [-\]: what put back behind? the oversight of the invasion at n17?
snowmen [1d\]: Yes
gosei [2d\]: hoho:)
]
;W[ie]WL[2136.441]
;B[lh]BL[1636.01]C[gosei [2d\]: r4 might be strange
]
;W[kh]WL[2119.44]
;B[li]BL[1634.15]
;W[ki]WL[2118.954]
;B[lj]BL[1631.895]
;W[oj]WL[2089.662]C[fluislost [?\]: how small is b3
hhfjxhfbDE [2d\]: zen19X is the same as zen19K ?
trohde [?\]: no
martin97 [-\]: but why did w exchange s19 n19?
starmie [4d\]: X = experimental
gogonuts [3d\]: s16 is really the only move - w is taking a long tim eto play it
hhfjxhfbDE [2d\]: ok 
snowmen [1d\]: no more aji left for r8
albino [5k\]: this is still an experimental bot at this stage ?
gogonuts [3d\]: sry, i wasn't up to date
GoIngo [?\]: bots are always experimental
Jazzman [1d\]: b c4?
GoIngo [?\]: X in the name stands for experimental
starmie [4d\]: They always experiment
]
;B[rn]BL[1516.986]C[trohde [?\]: well, if we humans are still experimental, how can any bot be anything else? :v
]
;W[rm]WL[2068.944]C[albino [5k\]: got the message
snowmen [1d\]: I wonder what happen after w s5
gogonuts [3d\]: b getting slow-cooked now
]
;B[ro]BL[1447.123]
;W[qm]WL[2054.152]C[polipor: wr?
albino [5k\]: without the question mark
hhfjxhfbDE [2d\]: b doesn't seem that far behind to me
gogonuts [3d\]: but there is also not so much to do - very hard to close the gap
gosei [2d\]: black really need some magic here....
pookpooi [?\]: lukas house is under heavy storm?
Parklane [11k?\]: so true "martin97 [-\]: you guys can write crazy strong go programs but not detect a deconnect? really? :D"
GoIngo [?\]: i do not think so
]
;B[cp]BL[1366.636]
;W[co]WL[2053.648]
;B[bq]BL[1364.279]C[kis [6k\]: g6 is magic?
hhfjxhfbDE [2d\]: y, that's true
miao [6d\]: now it is my time to profit
pookpooi [?\]: I'm afraid he's not concentrating on the match much
GoIngo [?\]: weather in Bonn is hot but no storm
]
;W[el]WL[2038.443]C[wr [?\]: hi polipor
pookpooi [?\]: because of the environment
]
;B[kj]BL[1316.722]C[ohyou [-\]: is anybody counting? looks like black is behind by over 10 points
mafutrct [-\]: To my knowldedge, disconnection detection is a limitation of GTP and/or KGS
]
;W[jj]WL[2004.426]C[pookpooi [?\]: "Yamato recovered his motivation to improve Zen as AlphaGo showed the way."
pookpooi [?\]: is he dismotivated because of alphago
Parklane [11k?\]: (perhaps but i thought the general idea was funny)
gogonuts [3d\]: :-)
pookpooi [?\]: I heard many programmer including darkforest also depress
kis [6k\]: connect d10'
]
;B[ib]BL[1268.399]C[kis [6k\]: ?
GoIngo [?\]: not depressed
GoIngo [?\]: they got new motivation
miao [6d\]: white is comfortable
GoIngo [?\]: at least Remi coulom did, and Zen team too
GoIngo [?\]: and abakus programmer also
pookpooi [?\]: and what is the new motivation, ingo?
GoIngo [?\]: to see that things are achievable
Neuneck [?\]: it's great to see that what you want to achieve is possible, after all
]
;W[bk]WL[1952.591]C[pyramids2 [1k\]: and how they are avhievable, with DCNN
gogonuts [3d\]: dardforrest was a different problem - google depressed his sponsor :-)
GoIngo [?\]: right
starmie [4d\]: Has there been any matches recently between Zen and Remi's Crazy Stone?
GoIngo [?\]: no
]
;B[hd]BL[1233.194]C[GoIngo [?\]: Zen was always ready for fight but CrazyStone did not show up
pookpooi [?\]: the cgi match is like, in april
Varelse [-\]: go is a theoretical problem, disconnection is a physical matter, what kind of thinker are you martin97?
Cam [1k\]: I like blacks move. 
]
;W[he]WL[1914.209]C[asdewq [?\]: 75% for zen
gogonuts [3d\]: b9 is a great problemsolver for w
GoIngo [?\]: is white already in homerun mode?
gogonuts [3d\]: can forget about e6 now
starmie [4d\]: Is 75% Zen's own estimation?
gogonuts [3d\]: i think so, ingo
asdewq [?\]: yea
Parklane [11k?\]: varelse automatically handling what you call "physical matter" is also an abstract problem
GoIngo [?\]: 75 was from Zen 6
starmie [4d\]: Ah
melmak: didnt think w had problems
pookpooi [?\]: can go be solved without exploring every move, like how we know the exact number of legal position
Tsuke [3d\]: its practical coding vs algorithm developmente
gosei [2d\]: there is no chance for black anymore....
pookpooi [?\]: the only breakthrough left for go
gosei [2d\]: zens judgment is really good
albino [5k\]: so white still has a chance ?
]
;B[dc]BL[1112.333]C[GoIngo [?\]: White = ZEn is ahead
gosei [2d\]: white is winning:)
gogonuts [3d\]: this w supersolid style can be infuriating, when you know you're not catching up (enough)
fluislost [?\]: "value network" feature has been added into zen in version 12.1
]
;W[hc]WL[1876.527]C[fluislost [?\]: this zen lacks it
ohyou [-\]: looks like black is catching up really good
fluislost [?\]: hope next month they use a newer zen
snowmen [1d\]: no, this game is over
gogonuts [3d\]: not if really good means actually having more in the end
GoIngo [?\]: Zen is a bit like the truck driver in Spielberg's "Duel"
]
;B[hj]BL[1067.655]C[miao [6d\]: need to take some risk here
gosei [2d\]: yep
GoIngo [?\]: you have to create semeais, Lukas
]
;W[jk]WL[1842.293]C[gogonuts [3d\]: lol, ingo
]
;B[ii]BL[1057.528]C[GoIngo [?\]: yes, hope is the last to die
Tsuke [3d\]: zen has lost some games with very high winrate, to high dans (8d+), iirc
albino [5k\]: g9 
gosei [2d\]: do or die...
Parklane [11k?\]: value network is the most interesting thing for me in alphago...
gogonuts [3d\]: h11 is overplay
gosei [2d\]: he needs it
gogonuts [3d\]: i mean J11
Tsuke [3d\]: good think he didnt play h11
Parklane [11k?\]: so "human" to estimate position
starmie [4d\]: What is the latest version of Zen you can buy?
]
;W[ji]WL[1795.313]C[melmak: dont think its do  or die its just nonsense
]
;B[gk]BL[1047.126]C[fluislost [?\]: zen 11.4
fluislost [?\]: hundreds of elo weaker than this one
GoIngo [?\]: but still enough to evaluate this position correctly
]
;W[fj]WL[1756.364]C[starmie [4d\]: Then I'll wait for the value network version
gogonuts [3d\]: b playing die or die?
]
;B[fk]BL[1031.733]C[fluislost [?\]: wise choice starmie
wareware [2k\]: dead or die!
GoIngo [?\]: life or death
gosei [2d\]: i think the beginning was good for black
]
;W[ek]WL[1731.609]C[albino [5k\]: black might just play j7 
snowmen [1d\]: b looking out for resignation point
pookpooi [?\]: more like next year for zen 7
Guest2d [1d\]: even if b lives w still wins
gosei [2d\]: after the invasion on the top it was completely reversed
agony [4k\]: g10 at f10 would have connected, no?
agony [4k\]: (g9)
Titan99 [2k\]: how strong is zen19x?
]
;B[gm]BL[999.888]C[gogonuts [3d\]: no way w start to finish victory
Parklane [11k?\]: L10 L9 so painful
Cam [1k\]: Agony:wedge
GoIngo [?\]: strong 7dan on KGS
Parklane [11k?\]: er K
]
;W[hi]WL[1706.56]C[agony [4k\]: i see, thanks
Cam [1k\]: Interesting 
snowmen [1d\]: Junsuk
albino [5k\]: how many guessed that right
gosei [2d\]: h11 says easy win?
gogonuts [3d\]: h11, what a killjoy ...
Cam [1k\]: Zen just wants the points... 
snowmen [1d\]: H11 is pro move
gosei [2d\]: not the move
ibd2 [?\]: H11... Interesting
gosei [2d\]: his judgement is pro
gosei [2d\]: the move it self is not special
Tomate: hello. i am new here. does the computer Zen display its evaluation ?
ibd2 [?\]: It does not
ibd2 [?\]: Not to us
pookpooi [?\]: ok just when dcnn is not confused enough, hideki mention drnn, what is that
gogonuts [3d\]: e6 not working
ohyou [-\]: so black is behind by around 10-14 points right now?
Cam [1k\]: No, but we can use commercial Zen to estimate it's evaluation 
GoIngo [?\]: tomate: only to the operator
Cam [1k\]: This one is stronger than commercial Zen 
GoIngo [?\]: it seems so
]
;B[bo]BL[837.155]C[ibd2 [?\]: Might be because of slower time settings mostly
Tomate: unfortunately my skill in go is very low.  So i have to usw my own go programm to receive a more or less  reliable evaluation ?
miao [6d\]: i need to keep attacking here
melmak: ?
gosei [2d\]: you can use lukas comments:)
lofuchi [3k\]: b is great
GoIngo [?\]: tomate: your bot will show your % values
gogonuts [3d\]: bean counting bot can't be unseated anymore
ibd2 [?\]: Tomate, you can use any free or open source program to get your own evaluation
]
;W[gn]WL[1655.062]C[GoIngo [?\]: at this stage of game they should be reliable
snowmen [1d\]: Tomate just use SE
pookpooi [?\]: se said w+102.5
albino [5k\]: lol
Cam [1k\]: Lol
GoIngo [?\]: do not trust SE
ohyou [-\]: ^
gosei [2d\]: se is just ok if the game is finished....
Guest2d [1d\]: well SE shows the potential
ibd2 [?\]: If your go level is low, you might try fuego for evaluation
polipor: not even once game finished
asdewq [?\]: 80% right now
pookpooi [?\]: leela!
gogonuts [3d\]: I think Se's chance of calling the winner is actually below 50% :-)
Tsuke [3d\]: 8d version of zen just started a game with a 9dan
]
;B[fm]BL[762.714]C[polipor: lol gogonuts
ohyou [-\]: wow that's so risky
DoctorNo [5d\]: 9d play another Zen
]
;W[en]WL[1633.882]C[pookpooi [?\]: omg it's too much for kibitz!!!
ibd2 [?\]: Leela is free, but not open source :) I'd trade some strength for open-sourceness
]
;B[im]BL[741.054]C[fluislost [?\]: there is a 9d vs zen 12.1 right now on kgs
gogonuts [3d\]: f7 wasnt risky, but it forced w to play the vital point, so didn't really help
pookpooi [?\]: when the computer go room is quient, it is really quiet,
]
;W[in]WL[1613.552]
;B[jm]BL[733.643]C[GoIngo [?\]: at least today
gogonuts [3d\]: w keeps pushing?
ibd2 [?\]: It has been a nice reduction sequence for black, but now w also gets its turn
]
;W[jn]WL[1572.712]C[asdewq [?\]: j5 or l8 
]
;B[km]BL[722.323]C[Parklane [11k?\]: J6 ?
]
;W[kn]WL[1547.948]C[gogonuts [3d\]: b close to calling it a day, imo
]
;B[lo]BL[696.284]C[ibd2 [?\]: W cuts now?
gosei [2d\]: m5 looks brave
GoIngo [?\]: it is cruel
ibd2 [?\]: M7
gogonuts [3d\]: m5 doesn't care - he must keep the points
snowmen [1d\]: m5 works fine
gogonuts [3d\]: b in "show me" mode
]
;W[lm]WL[1495.326]C[gosei [2d\]: m4 seems better
]
;B[kl]BL[691.506]C[miao [6d\]: black should be fine with this result
ibd2 [?\]: Hehe
ohyou [-\]: white lost 7 points on the left
]
;W[kk]WL[1455.611]
;B[ll]BL[677.681]
;W[lk]WL[1455.147]C[GoIngo [?\]: losing by 0.5 is not so different from losing high
]
;B[mm]BL[674.864]C[GoIngo [?\]: at least for the bot
Parklane [11k?\]: go humans :D
GoIngo [?\]: I mean "winning" of course
gosei [2d\]: im really scared that m5 might be loosing move...
asdewq [?\]: still 80%
ohyou [-\]: doesn't mean you shouldn't count the board to understand what's going on
gogonuts [3d\]: the logic also works for losing :-)
]
;W[mj]WL[1408.008]C[Jazzman [1d\]: ouch
gogonuts [3d\]: b has n9
albino [5k\]: damezumari somewhere
Guest2d [1d\]: now SE is more like to be correct, w+100+
fluislost [?\]: there'S a real 9d game vs a much stronger zen than this one.  
albino [5k\]: so ?
]
;B[mk]BL[625.242]C[fluislost [?\]: time to switch
gogonuts [3d\]: locally no problem - the problem is that b is still losing
KlausB [?\]: fku: do so and let us watch ...
fluislost [?\]: ok
]
;W[nk]WL[1360.859]C[GoIngo [?\]: +1
gosei [2d\]: picknick ko for ehite
]
;B[ml]BL[622.311]C[gosei [2d\]: white
]
;W[ni]WL[1338.311]C[albino [5k\]: ah, thats better
gogonuts [3d\]: ko wasnt for kill
gogonuts [3d\]: so just this
]
;B[mh]BL[599.618]C[starmie [4d\]: I still count it as W+15
ohyou [-\]: if black loses the eye here then h18 is sente, am I right?
miao [6d\]: h18 sente is huge
ohyou [-\]: oh I am
wareware [2k\]: n15 first?
gogonuts [3d\]: is b even alive?
miao [6d\]: but is white alive?
khongtac [7k\]: p12?
]
;W[ij]WL[1292.208]C[Neuneck [?\]: yes, eyes at l18 and k19
ohyou [-\]: yeah previously I think f18 would've worked
GoIngo [?\]: we need semeai !
GoIngo [?\]: multiples !
gosei [2d\]: o11 must be misread?
gogonuts [3d\]: o11 is ok imo
gogonuts [3d\]: b had too many eye options to go for the kill
gogonuts [3d\]: w just cleaning up
snowmen [1d\]: Zen wants to drive home the safe way
starmie [4d\]: W is just removing aji now it seems
gosei [2d\]: well o11 was just aji keshi
asdewq [?\]: zen thinks it\\s more than 10 pts difference
gosei [2d\]: could have played ko...
gosei [2d\]: probably zen thinks
asdewq [?\]: when I set komi at -6.5, w is still winning
gosei [2d\]: he iswinning anyway
gogonuts [3d\]: can't interest a well fed lion with a steak
GoIngo [?\]: thx for ingo, asdewq
ohyou [-\]: why didn't white play h18?
GoIngo [?\]: info
asdewq [?\]: no prob
gosei [2d\]: ahhh
gosei [2d\]: o11...zen might avoid ko?
gogonuts [3d\]: half an eye at p14, and also at h8, if b gets j10
gosei [2d\]: is it true that zen doesnt like kos?
gogonuts [3d\]: no bot likes them
gogonuts [3d\]: not even alpha
polipor: most humans also don't like kos
starmie [4d\]: lol
gosei [2d\]: well then,o11 seems just saying im winning,even if i pass;)
gogonuts [3d\]: no one has taught his bot yet, that he can't win 2 :-)
ohyou [-\]: there is a proverb "if you don't like ko, don't play go"
gogonuts [3d\]: ko's mess up the search tree
GOuroboros [2d?\]: fortunately bots can't understand proverbs so well
gogonuts [3d\]: bot proverb: if you don't like ko, don't play ko
]
;B[bh]BL[263.08]C[pokamon [2k\]: i saw the zen19k version playing perfect ko fights
GoIngo [?\]: ;-)
theolddan [1d\]: has anyone mentioned the h7 ko yet?
GoIngo [?\]: good proverb
GoIngo [?\]: it was implemented in alphago i think
Tsuke [3d\]: perfect ko fight.. from 2k perspective >.<
gogonuts [3d\]: lol, ingo
]
;W[ip]WL[1249.832]C[GoIngo [?\]: i was serious
pokamon [2k\]: maybe:-)
pokamon [2k\]: look for yourself
Tsuke [3d\]: not strong enough to judge
snowmen [1d\]: yes h7 is ko, but Zen does not play this for sure
]
;B[iq]BL[230.313]C[theolddan [1d\]: j4 even fancier way to start that ko
gogonuts [3d\]: great wedge
Tsuke [3d\]: the bot is 8d, if its bad at ko fighting, it might be playing it at 5d level
martin97 [-\]: i dont get b12 instead of b13
]
;W[io]WL[1235.083]C[gosei [2d\]: i think zen like to avoid it
theolddan [1d\]: right
theolddan [1d\]: zen didnt do it
m2m [1k\]: I think zen will win alphago some day
GoIngo [?\]: i hope so
gosei [2d\]: i wonder,if there is 100 point ko for free,and zen is winning,would he start it?;)
gogonuts [3d\]: b must feel slightly being made fun of right now
starmie [4d\]: Would be interesting to see zen vs AlphaGo if they have equivalent hardware
]
;B[gq]BL[152.422]C[pookpooi [?\]: alphago on single machine is very strong
albino [5k\]: or the same picture
gosei [2d\]: especially o11 is like,saying gg in advance...
]
;W[bn]WL[1202.54]
;B[dh]BL[133.706]
;W[eg]WL[1201.994]C[polipor: gosei, no, it won't
gosei [2d\]: cool:)
]
;B[bg]BL[121.111]C[gosei [2d\]: interesting
m2m [1k\]: w getting points here and there
First [1k\]: is this game still going on?
gogonuts [3d\]: you mean b
m2m [1k\]: and i think Zen will invent new go
m2m [1k\]: not alphago
gogonuts [3d\]: but w just paying a safety tax
]
;W[gj]WL[1156.559]C[starmie [4d\]: Good way to see it gogonuts
hummel: before playing d12 black should have asked e16
GoIngo [?\]: just back from a disconnect
hummel: maybe even before b12
]
;B[sm]BL[28.926]
;W[sl]WL[1124.442]
;B[sn]BL[26.401]
;W[rl]WL[1123.942]
;B[ec]BL[7.181]C[YoungPro [6d\]: zen always believe in his counting
GoIngo [?\]: not always
GoIngo [?\]: in multiple semeais not
pookpooi [?\]: zen just won again 9d and now play with him again as 8d vs 8d
]
;W[qo]WL[1088.109]
;B[rp]BL[2.288]C[pookpooi [?\]: interesting
GoIngo [?\]: it was a 9d?
gghideki [?\]: >200 observers
polipor: not a real 8d
GoIngo [?\]: "9d?"
miao [6d\]: maybe i should have played s3
polipor: at least, not know
gogonuts [3d\]: sometimes you can win a hopeless position, when it lets you get close and then it overlooks just one little thing
miao [6d\]: more ko threats but potentially a point more
pookpooi [?\]: smallbird used to be top10 player in kgs, see his expired game records
miao [6d\]: depending on q4 sente or not later
GoIngo [?\]: ok
gosei [2d\]: a ko would not appear;)
]
;W[ln]WL[1041.375]
;B[mn]BL[60]OB[3]C[polipor: nice for smallbird
martin97 [-\]: i think its closer than you keep saying
]
;W[ko]WL[992.499]C[asdewq [?\]: 85% now, 10 pts diff
]
;B[kp]BL[60]OB[3]C[gosei [2d\]: martin is right zen winningby 3,5 maybe
First [1k\]: it is not close 
gogonuts [3d\]: less than 10
gogonuts [3d\]: but where to catch up?
ohyou [-\]: I counted 8 points
gosei [2d\]: not possible against zen in the endgame
GoIngo [?\]: bots go for such wins when they are stable
]
;W[fb]WL[943.755]
;B[ed]BL[60]OB[3]C[miao [6d\]: i didn´t check if i die now
wade [5d\]: what do you mean by stable?
miao [6d\]: but seems unlikely
Guest2d [1d\]: emotionally?
GoIngo [?\]: look at SE screen
GoIngo [?\]: it looks rather stable
]
;W[fd]WL[892.636]C[pokamon [2k\]: cho chikun maybe will kill b
GoIngo [?\]: killing is not necessary
MaLa [2d?\]: there was no kill before F16
gogonuts [3d\]: I think b will have more points on the board
Amenophis [2d?\]: as w i'd give k17 a try
]
;B[hb]BL[60]OB[3]C[GoIngo [?\]: Lukas is in Byoyomi
gosei [2d\]: white wins by 3,5-1,5
]
;W[il]WL[839.412]C[gosei [2d\]: i feel,eventually 1,5
YoungPro [6d\]: zen can make it to 0,5
YoungPro [6d\]: there is no problem
YoungPro [6d\]: its still win
gosei [2d\]: yes
]
;B[hm]BL[60]OB[3]C[asdewq [?\]: monkey jump now
asdewq [?\]: or a5
GoldenBear [1d\]: That is the nice thing when you play against bots. You can continue playing even lost positions as the bot will make it look like the game was close enough all the time :-)
ohyou [-\]: so few aji left for black, how to come back?
First [1k\]: it doesn't matter
First [1k\]: huge victory for zen either way
GoIngo [?\]: yes, they are sort of polite
]
;W[ao]WL[788.93]C[GoIngo [?\]: not only Japanese bots, more or less all of them
Meruem92 [3d\]: chance!
dfan [4k\]: They play teaching games
Meruem92 [3d\]: ko
gosei [2d\]: where?
gosei [2d\]: ahh
gosei [2d\]: might confuse...
First [1k\]: miao got schooled
]
;B[ee]BL[60]OB[3]C[Meruem92 [3d\]: ko for sente
First [1k\]: why not select a strong opponent?
GOuroboros [2d?\]: liek yourself??
GoIngo [?\]: Lukas is the best German player
gosei [2d\]: First.........
readytogo [4k\]: e15  nice yose
wade [5d\]: first: thats uncalled for
]
;W[fe]WL[754.53]C[ohyou [-\]: first in the nickname, last to get brain
First [1k\]: strong in chatting during game maybe
YoungPro [6d\]: zen defeated andy liu 15:1 ... and you guys expect 6d egf to win?
snowmen [1d\]: miao got teached
trohde [?\]: doesn’t B need to play K17?
gogonuts [3d\]: taught
readytogo [4k\]: learnt!
]
;B[ap]BL[60]OB[3]C[Titan99 [2k\]: how far is zen compared to alpha go? xD
gosei [2d\]: andy liu is not stronger than 6d egf...
Guest2d [1d\]: ok learning English
gogonuts [3d\]: is our children learning?
Meruem92 [3d\]: @gosei: andy liu is not stronger than SOME 6d egf
GoIngo [?\]: our English is liquid
Meruem92 [3d\]: there are several 6d egf which cannot compare to miao either
YoungPro [6d\]: andy liu defeated many professionals in japan and had good game with murakawa and was leading in his game with lee sedol but unfortunately lost, no way for 6d egf to do that
]
;W[me]WL[705.119]
;B[ne]BL[60]OB[3]C[Christoph [2k\]: big bad wolf, right?
Meruem92 [3d\]: give them enough chance and they'll prove you wrong YP
Guest2d [1d\]: dead
Meruem92 [3d\]: europe rarely gets to play asian pros
CrazySt0ne [8d\]: P14 ...
YoungPro [6d\]: yeah but andy didn't have many chances,probably if 6d egf get 100 chances they may get some wins against pros
gogonuts [3d\]: n15 pretty strange
OneEyed [4k\]: Would Andy Liu be of similar strength to e.g. Fan Hui?
GoIngo [?\]: what do CS % say here?
CrazySt0ne [8d\]: k17
kis [6k\]: b dead now?
Meruem92 [3d\]: do you even know the european players?
]
;W[bp]WL[652.477]C[evenletter [2d\]: k17 
ibd2 [?\]: K17 p14 Miami?
ibd2 [?\]: Miai***
gogonuts [3d\]: i think n15 was terrible
ibd2 [?\]: Damn keyboard
YoungPro [6d\]: I am just saying that Andy is the strenght of Mark Lee and mark lee is kingsman and hitgiri9
dfan [4k\]: Andy Liu's EGF rank is 7d, FWIW
gogonuts [3d\]: maybe b can win now
pyramids2 [1k\]: n15 was double sente, more so for w, so why is it strange, gogonuts? because it would have been better as ko threat?
gosei [2d\]: im sorry,iwas mistalen andy with albert
gosei [2d\]: mistaken
kis [6k\]: b can play p 14
Meruem92 [3d\]: i'm not saying andy liu is weak, but europe is catching up
gogonuts [3d\]: because a: take ko
gogonuts [3d\]: anb b. b has m14 now
gosei [2d\]: i think europeans are stronger in general...
Meruem92 [3d\]: miao is one of the guys who have a chance to get to a higher level through the pro examinations being held in europe nowadays
YoungPro [6d\]: ofcourse europe is getting stronger(consdering that artem defeated two pros (1 handi and one win with resign)
pyramids2 [1k\]: i see. well the move ordering (ko before threat) is difficult in MCTS, so it's a typical bot mistake, of course
]
;B[lf]BL[60]OB[2]C[OneEyed [4k\]: ah, goratings.org has andy liu a few hundred ranks ahead of fan hui
readytogo [4k\]: yes p 14 
miao [6d\]: not sure about what was best here
pookpooi [?\]: noooo lucas is in sd
snowmen [1d\]: aetem is pro himself
martin97 [-\]: this threat is weird
pookpooi [?\]: oh, not
YoungPro [6d\]: s17 was surely ko threat
readytogo [4k\]: take a few pts then live
gogonuts [3d\]: b must have been very happy to see n15
GoIngo [?\]: still two periods
]
;W[le]WL[617.068]C[OneEyed [4k\]: so that 14-1 is indeed comparable to AlphaGo v13 vs Fan Hui, except for less serious game conditions and unfriendly time settings
asdewq [?\]: e12 was a good threat
]
;B[bo]BL[60]OB[2]C[First [1k\]: I am proud to be european
First [1k\]: but strength wise not as developed
pyramids2 [1k\]: also, gogonuts, would n15 have been better at m14?
gogonuts [3d\]: oh god first
YoungPro [6d\]: well they weren't serious games and quite fast but still 14:1 is impressive for a zen.. too impressive to me..
gogonuts [3d\]: instead of n15 just take the ko
gosei [2d\]: i cantremember any competion,european lost to america...
]
;W[je]WL[564.732]C[readytogo [4k\]: but he also has k17  so its miai
YoungPro [6d\]: back then when zen defeated kingmaker 4:1 and defeated cheater and roln(ilya) ... i was ...
]
;B[an]BL[60]OB[2]C[gogonuts [3d\]: w just gave b a chance either capture 3 stones or break in
pyramids2 [1k\]: yes, i see
YoungPro [6d\]: yeah but this lower was bigger
]
;W[am]WL[527.768]C[YoungPro [6d\]: d15 is big
gosei [2d\]: small success for black
wade [5d\]: the representitive of "the rest of the world" on samsung cup is european
gogonuts [3d\]: i think its very close now
pookpooi [?\]: zen win against smallbird 8d, interesting
kis [6k\]: e18
]
;B[pg]BL[60]OB[2]C[gosei [2d\]: also in a team tournament 2014 europe won against america 3:0
wurfmaul [3d\]: Where can I find information about the 14:1?
gogonuts [3d\]: that was a huge success for b
pookpooi [?\]: see zen19k record
gogonuts [3d\]: because he had already gotten close
gogonuts [3d\]: the game is now up for grabs
pookpooi [?\]: it's short time game anyway don't take it seriously
trohde [?\]: http://www.gokgs.com/gameArchives.jsp?user=zen19k
]
;W[na]WL[476.513]C[Guest2d [1d\]: zen's end game is super conservative 
gosei [2d\]: zens confidence must be huge if he would be human:)
YoungPro [6d\]: but cloudnotes said that it was too late and as you can see he played some strange things in one of the games.. not only in one.. but still.. zen did well..
trohde [?\]: vs. http://www.gokgs.com/gameArchives.jsp?user=zen19x
Meruem92 [3d\]: really gosei? is there a page with the results?
ibd2 [?\]: What just happened??
]
;B[la]BL[60]OB[2]C[pookpooi [?\]: do you want zen's end game to be liberal?
readytogo [4k\]: eye killing ko
miao [6d\]: m19 is a joke
ceylon: lol th
snowmen [1d\]: k17 kolls big
miao [6d\]: p14 is much better
ibd2 [?\]: What was the big score swing, the ko on the right?
ibd2 [?\]: Left**
]
;W[qg]WL[429.397]C[miao [6d\]: bot is cheating my nicely here
]
;B[ph]BL[60]OB[2]C[wurfmaul [3d\]: Thanks pookpooi, trohde. So I should look for cloudnotes or someone else?
gogonuts [3d\]: i would sy b by 0.5 to 1.5
GoIngo [?\]: right - in KGS archive
YoungPro [6d\]: cloudnotes but the quality is not very cool for the last 10 games
]
;W[qh]WL[402.081]C[gogonuts [3d\]: strange time for lukas to moan :-)
]
;B[bm]BL[60]OB[2]C[asdewq [?\]: 55% for zen now, very strange
pookpooi [?\]: cloudnotes, strongly rumored to be andy liu usa 1p
asdewq [?\]: maybe 12.0 sees something else
miao [6d\]: asking some questions
GoIngo [?\]: do we have something?
gogonuts [3d\]: i think b7 is actually a detour
GoIngo [?\]: what are CrazyStone histograms here?
]
;W[al]WL[349.833]C[gogonuts [3d\]: b was already ahead, imo
Meruem92 [3d\]: b7 confuses me
gogonuts [3d\]: oh, but that helped even more ...
Meruem92 [3d\]: timesuji?
YoungPro [6d\]: e18 huge
Meruem92 [3d\]: e18 huge, but w can kill in ko
Meruem92 [3d\]: if b plays it
pookpooi [?\]: b not ahead yet
YoungPro [6d\]: what ko
]
;B[dm]BL[60]OB[2]C[Meruem92 [3d\]: or nvm
wade [5d\]: d7 minus pt
gogonuts [3d\]: oh no
]
;W[em]WL[319.888]C[YoungPro [6d\]: what is he doing :)
gogonuts [3d\]: threw it away again?
gosei [2d\]: nice:)
]
;B[do]BL[60]OB[2]C[gogonuts [3d\]: i can't believe my eyey
gogonuts [3d\]: eyes
martin97 [-\]: nice
pokamon [2k\]: hallu?
gogonuts [3d\]: pardon the pun
miao [6d\]: dangerous for white
polipor: w+77 according to SE XD
CrazySt0ne [8d\]: wtf
YoungPro [6d\]: what is b doing
YoungPro [6d\]: why he is so rude to white
readytogo [4k\]: nice
YoungPro [6d\]: he must be kind
gogonuts [3d\]: b playing for ko
miao [6d\]: no idea about ko threats though
readytogo [4k\]: just a little bite
]
;W[fn]WL[273.471]
;B[dn]BL[60]OB[2]C[gosei [2d\]: ko really confuse zen
]
;W[ao]WL[272.924]C[Meruem92 [3d\]: worry not miao, any ko here is better than no ko
YoungPro [6d\]: s17 is big enough
hummel: can't W play d5 before e7?
YoungPro [6d\]: e12 huge also
hummel: sorry B
YoungPro [6d\]: b win for sure
GoIngo [?\]: ??
asdewq [?\]: zen goes negative
Guest2d [1d\]: e12
Schachus12 [3k\]: did any of you guys count accurately to  know wheter ko was needed for b?
YoungPro [6d\]: w missread so bad
GoIngo [?\]: which %
Meruem92 [3d\]: s17, l15, p5, e12, d9, h6
gosei [2d\]: so if ko appeas, gg easy;)
Meruem92 [3d\]: any other threats?
martin97 [-\]: remember your own words, yp, no way b can win :P
r0n1n [2d\]: YoungPro [6d\]: i dont see any chance for b
r0n1n [2d\]: :D
gogonuts [3d\]: pretty sure it wasn't needed
readytogo [4k\]: too bad s17  donst work anymore
GoIngo [?\]: martin +1
YoungPro [6d\]: yeah its funny how things changes
gogonuts [3d\]: there was no way for b to win
MaLa [2d?\]: so.... the history of codecentric games is always the same: bots go into endgame in confident manner, everyone calls for an easy close botvictory... bots screw up in endgame...
gogonuts [3d\]: but w found a way to lose, it seems :-)
YoungPro [6d\]: i dont know how can zen mess up things?
MaLa [2d?\]: was pretty much the same in the last 2 years too
wade [5d\]: f13 +g12 meruem
YoungPro [6d\]: i mean,he was winning, i thought that when zen is in leading mode,it just..
MaLa [2d?\]: the myth of the invicible endgame of bots falling apart again and again
pookpooi [?\]: b is under sudden death
gogonuts [3d\]: losing on time is another option ...
]
;B[rc]BL[60]OB[1]C[martin97 [-\]: but i think it was less ahead than it thought
Meruem92 [3d\]: oh
Meruem92 [3d\]: yeah
Titan99 [2k\]: zen not enough processor power like alpha go has in end game
Guest2d [1d\]: Like I said, confident doesn't mean to be conservative
Titan99 [2k\]: :)
Meruem92 [3d\]: i meant e12 to be several ko threats
readytogo [4k\]: lol
Schachus12 [3k\]: doesnt w have sone good threats, too?
pyramids2 [1k\]: is that big enough?
gosei [2d\]: no chance for white
gogonuts [3d\]: this is enough for b ...
KiraLNear: definetly enough
YoungPro [6d\]: win for humans ! 
]
;W[sf]WL[232.57]
;B[an]BL[60]OB[1]C[marukome [5d\]: this game is event or something?
Meruem92 [3d\]: and t15's also there
GoIngo [?\]: I want to see bot % here
Guest2d [1d\]: no threat for w
Schachus12 [3k\]: like k17?
Meruem92 [3d\]: it is marukome, yes
gosei [2d\]: might resign now?
marukome [5d\]: thx
Vinicius [2d\]: f8
ohyou [-\]: https://blog.codecentric.de/en/2016/08/codecentric-go-challenge-2016/
nonosse [2k\]: e5 is a threat too
wade [5d\]: p14 k19 2 threats
Titan99 [2k\]: black top dragon not alive also?
]
;W[ia]WL[190.73]C[gghideki [?\]: ?
gogonuts [3d\]: thats a threat
Titan99 [2k\]: it is a ko threat
elsky [1d\]: ignoe
poirette [1k\]: wow
gogonuts [3d\]: cant ignore
poirette [1k\]: funny threat
GoIngo [?\]: hideki what is your baby doing?
gosei [2d\]: j19,-100000 points ko thret;)
gonmf [-\]: huh
pookpooi [?\]: is zen overheat?
Schachus12 [3k\]: why not k17 instead?
Titan99 [2k\]: 1d said ignore? lol
gogonuts [3d\]: threatens k19
ibd2 [?\]: Tiiiiime
Vinicius [2d\]: why not p14
martin97 [-\]: ko is similar in size
]
;B[ha]BL[60]OB[1]C[martin97 [-\]: but probably smaller
snowmen [1d\]: Zen crash
GoIngo [?\]: my nerves ...
asdewq [?\]: c7 and b was wining :)
gosei [2d\]: any move and black is winning;)
Nonooo [2d\]: well b groupwas ko then
]
;W[ao]WL[145.193]C[gogonuts [3d\]: threatened k17, sry
Nonooo [2d\]: oh dead even
YoungPro [6d\]: e5
m1m [1k\]: d9 
]
;B[hn]BL[60]OB[1]C[Meruem92 [3d\]: e5 is a bad threat
Meruem92 [3d\]: d9 better
martin97 [-\]: h6 is surprising
Schachus12 [3k\]: zen ignore?
gogonuts [3d\]: this will do
YoungPro [6d\]: why not something safer :D 
gosei [2d\]: j19 coould be zens move:)
YoungPro [6d\]: but it must be enough anyway
gosei [2d\]: h19 sorry
]
;W[bp]WL[102.692]C[readytogo [4k\]: hmm
Meruem92 [3d\]: err
gogonuts [3d\]: finish it
gosei [2d\]: black very cautious
YoungPro [6d\]: threat now
martin97 [-\]: e5
Meruem92 [3d\]: me is confused
YoungPro [6d\]: e12
]
;B[dk]BL[60]OB[1]C[Meruem92 [3d\]: me gonna start hitting me self
ibd2 [?\]: What is going on
]
;W[dl]WL[84.913]C[pookpooi [?\]: interesting
]
;B[bo]BL[60]OB[1]C[gosei [2d\]: white suddenly felt apart
Meruem92 [3d\]: b's being a bit evil here
]
;W[jc]WL[45.523]C[readytogo [4k\]: huge
rsun [5d\]: hi hideki, have you or yamato tested 12.1 without gpu
]
;B[ke]BL[60]OB[1]C[pookpooi [?\]: rsun you're late!
gogonuts [3d\]: nice L15
rsun [5d\]: I sure am
Schachus12 [3k\]: if w takes the ko he takes 3 stones?
gogonuts [3d\]: better than living right away
pookpooi [?\]: zen19k just won even game against smallbird 9d and 8d
Schachus12 [3k\]: or h5?
gosei [2d\]: well gg,even it was a little lucky,imhappy human won
readytogo [4k\]: b plays very efficiently
gogonuts [3d\]: and if w connects, b gained a threat
]
;W[lf]WL[60]OW[3]C[ibd2 [?\]: Efficiently or effectively?
]
;B[eb]BL[60]OB[1]C[readytogo [4k\]: yes
gogonuts [3d\]: lol
Meruem92 [3d\]: b greedy now
gogonuts [3d\]: b having a good time now
rsun [5d\]: that makes me feel great for being able to win 1 game
Schachus12 [3k\]: ah, even that one
Vinicius [2d\]: E18 good
martin97 [-\]: bot style
Titan99 [2k\]: black has enough threat?
KiraLNear: E18 points but more threats for white
nonosse [2k\]: e18 helps for life
readytogo [4k\]: look at that timing
Vinicius [2d\]: ko small now
agony [4k\]: e18 worth more than h5?
Schachus12 [3k\]: make enough points, so you dont need threats
gogonuts [3d\]: getting e18 for free along the way is great
Titan99 [2k\]: p14?
Schachus12 [3k\]: e18 answers w s threat
]
;W[ea]WL[60]OW[3]C[Meruem92 [3d\]: agony, e18 is to ensure h19's life
]
;B[da]BL[60]OB[1]C[agony [4k\]: ah
readytogo [4k\]: now must p14 
kis [6k\]: if w k14 b dead?
GoIngo [?\]: Zen also in Byo now
kis [6k\]: p14
Vinicius [2d\]: F19 P14 miai
]
;W[ho]WL[60]OW[3]C[gogonuts [3d\]: no because right now b also has connection option
]
;B[an]BL[60]OB[1]C[Schachus12 [3k\]: finally w decided to answer h5..
martin97 [-\]: what was all this mess about now ? :D
gogonuts [3d\]: bs advantage is balloning
kis [6k\]: thx vinicius
Titan99 [2k\]: looks like e18 good end game move for black :)
Schachus12 [3k\]: at least he has threats now
asdewq [?\]: w is losing
ohyou [-\]: what is balloning?
Schachus12 [3k\]: but lost lots of points
ibd2 [?\]: Gggggggg
readytogo [4k\]: good game
GoIngo [?\]: wow
r0n1n [2d\]: gg !
KiraLNear: congrats miao!
trohde [?\]: wow
polipor: wow
CrazySt0ne [8d\]: xD
Neuneck [?\]: gg
rsun [5d\]: strong like king kong
snowman [-\]: gg
gonmf [-\]: gg
martin97 [-\]: gg
ceylon: wow
GoIngo [?\]: congrats, Lukas!
miao [6d\]: thx
gghideki [?\]: thanks
YoungPro [6d\]: gratz
fj [?\]: nice!
trohde [?\]: congratulations to miao!
starmie [4d\]: Nicely done!
miao [6d\]: pffff
agony [4k\]: congrats Lukas!
Christoph [2k\]: grats
gogonuts [3d\]: congrats
miao [6d\]: sorry for disconnect bte
gosei [2d\]: relieved no?
OneEyed [4k\]: Well done!
hummel: yeah seems nobody saw d7 coming
GoIngo [?\]: what a show
miao [6d\]: i am in a chinese restaurant because wifi brike down at my place
gogonuts [3d\]: I thought you were toast for a long time
rsun [5d\]: when is round 2?
miao [6d\]: next sunday7
trohde [?\]: what a thrilling game
agony [4k\]: lol!
gghideki [?\]: 1 week later
GoIngo [?\]: lukas, take photo from restaurant
gogonuts [3d\]: outplayed the bot in the tight endgame, congrats
miao [6d\]: i like whites opening
Amenophis [2d?\]: Thank you for the game.
GoldenBear [1d\]: Congratulations to Lukas! 
GoIngo [?\]: yes, next sunday again same time
miao [6d\]: ok
First [1k\]: how is this possible?
ohyou [-\]: such a good come back
First [1k\]: bot got tricked?
trohde [?\]: thanks to everybody, esp. miao and gghideki!
gosei [2d\]: First,yes
trohde [?\]: and goingo, of course
starmie [4d\]: bot got outplayed
Werfeus [5d\]: nah man, skip that and come to Brno!
gogonuts [3d\]: speciaL THANKS TO HIDEKI
GoIngo [?\]: what a nerve wrecker
GoldenBear [1d\]: And thanks also to Hideki! See both of you next Sunday! Monday (for Hideki :-) )
GoIngo [?\]: yes, thx to Hideki - middle in Japanese night
gogonuts [3d\]: didn't want caps, but fitting :-)
gosei [2d\]: i guess Lukas know now what to do in the next games:)
polipor: gghideki sleeping yet
miao [6d\]: i think the bots opening was like japanese top ama
rsun [5d\]: hideki may be sleeping in his chair
GoldenBear [1d\]: The game will be soon online at go.codecentric.de
gogonuts [3d\]: the opening was pretty crushing
readytogo [4k\]: id like to thank maio zen my mom and god ! !!!!!
readytogo [4k\]: !!
miao [6d\]: not so sure gogo
gghideki [?\]: Yah, it's 3;30 am, I'm going sleep, see you all!
GoIngo [?\]: i have prepared a nice photo too
gogonuts [3d\]: kudos to you for recovering from that
miao [6d\]: just continuation was horrible
])