Forums » Go
Automatic dead stone markup
Replies: 14   Pages: 1   Most recent reply posted by: nefrin (Feb 25, 2014 11:49 AM)
Welcome, Guest
Login
Guest Settings
 Reply to this Topic  Post New Topic  Search Forum
 
  Back to Topic List
 
Replies: 14   Pages: 1  
DEVELOPER

Posts: 513
Registered: Sep, 2004

Automatic dead stone markup Posted: Jul 16, 2013 4:49 PM
Reply
From now on, dead stones will be automatically marked by the server. Hope this will solve the pass/reject abuse.


rutabega52

Posts: 13
Registered: Sep, 2012

Re: Automatic dead stone markup Posted: Jul 16, 2013 5:55 PM
Reply
This is fantastic news! A missing feature long existing on other sites- this will solve a lot of problems and weed out a lot of idiots! Thanks!

TheKarlBad

Posts: 30
Registered: Aug, 2012

Re: Automatic dead stone markup Posted: Jul 16, 2013 7:58 PM
Reply
I love this idea... but there is an error in the seki rule. I played a game and my opponents stones were marked dead... it wa a seki pls fix this, this one mistake can mess up a game

Jiggzaw

Posts: 17
Registered: Jan, 2011

Re: Automatic dead stone markup Posted: Jul 16, 2013 9:28 PM
Reply
About time!
And yes, there's a seki marking problem..

erȷc

Posts: 98
Registered: Mar, 2013

Re: Automatic dead stone markup Posted: Jul 17, 2013 2:17 AM
Reply
I like the intent here, but really this is the wrong approach. This problem is too gnarly. You can't solve it this way. You would need a top level go program and even then you would run into corner-cases where the algo would be wrong. No dedicated Go server does this and there is a good reason for it.

Thanks for trying.

DEVELOPER

Posts: 513
Registered: Sep, 2004

Re: Automatic dead stone markup Posted: Jul 18, 2013 10:50 AM
Reply
Automatic markup will take seki into account now.
Please note, that eyes in seki still count in territory. This will be fixed later. (Previously, you was not able to mark stones alive in seki at all.)

We are aware of the fact that automatic markup won't be 100% accurate. However, we believe that the number of cases where the winner of a match is incorrectly decided due to this inaccuracy will be relatively small to the number of previous pass/reject abuses.

whiteghost3

Posts: 372
Registered: Apr, 2005

Re: Automatic dead stone markup Posted: Jul 19, 2013 12:15 PM
Reply
i like the idea.i played go a few times when i was 18 but havnt played it since apart from a few times on here.i dont know which stones to mark and automatically will make it a lot easier for me.
i understand the game but have very little experience in playing it.

gennan

Posts: 10
Registered: Jun, 2009

Re: Automatic dead stone markup Posted: Aug 11, 2013 7:21 PM
Reply
I think this is a great improvement. Particularly, because is seems to mark correctly (up to now), which is far from trivial in some cases. Good job!

Ras Tafari

Posts: 1
Registered: Sep, 2012

Re: Automatic dead stone markup Posted: Aug 18, 2013 7:42 PM
Reply
Hi, this system is great, but still not flawless (see picture in the link). I can imagine why it is so hard though. Great job !

http://img809.imageshack.us/img809/8586/85gl.jpg

super knicks

Posts: 6
Registered: Apr, 2013

Re: Automatic dead stone markup Posted: Sep 2, 2013 11:09 PM
Reply
There is still a seki problem. I have a screenshot of a game I recently played and as a result of the seki, I lost.

http://flickr.com/gp/101266413@N08/s33ac6

Lion Weiqi

Posts: 1
Registered: Jun, 2013

Re: Automatic dead stone markup Posted: Sep 3, 2013 7:25 PM
Reply
http://senseis.xmp.net/?BentFourInTheCorner

The group was dead indeed, the machine just knew more than you.

smiter34

Posts: 1
Registered: Oct, 2010

Re: Automatic dead stone markup Posted: Sep 6, 2013 1:34 PM
Reply
i just played a game that it gave me a loss because it marked my stones as dead when it was his that were dead, here was the game
(;FF[4]GM[1]AP[FlyOrDie Go:?]
SZ[15]RU[Japanese]TM[1800]
PB[smiter34]BR[42]BL[1365]
PW[iceiggle]WR[26]WL[1389]
DT[2013-09-06]EV[Online game]
PC[http://www.flyordie.com/go/]
RO[1]RE[W+10]
;B[dd];W[lm];B[ld];W[mk];B[cl];W[md];B[lb];W[lc];B[kc];W[mc];B[jd];W[le];B[kd]
;W[lf];B[gd];W[cm];B[bm];W[dm];B[cj];W[bn];B[al];W[dl];B[ck];W[dk];B[hl];W[hm]
;B[im];W[gm];B[ik];W[gl];B[gk];W[fk];B[gj];W[fj];B[gi];W[jl];B[il];W[ij];B[jk]
;W[jj];B[kk];W[kl];B[lj];W[lk];B[kj];W[jh];B[mh];W[lh];B[mi];W[mg];B[ng];W[nf]
;B[oh];W[mj];B[li];W[ki];B[nj];W[nk];B[oj];W[ok];B[ni];W[cc];B[bb];W[bc];B[cb]
;W[dc];B[db];W[ec];B[eb];W[ed];B[de];W[dj];B[di];W[be];B[cg];W[bg];B[bh];W[cf]
;B[df];W[dg];B[ch];W[eg];B[ef];W[ff];B[fg];W[eh];B[ei];W[fi];B[fh];W[fe];B[hf]
;W[fd];B[hg];W[gc];B[hd];W[hc];B[gb];W[fb];B[fa];W[hb];B[fc];W[ce];B[bd];W[bf]
;B[ac];W[mb];B[ka];W[id];B[ie];W[ic];B[je];W[jb];B[kb];W[hj];B[hk];W[ih];B[hh]
;W[hi];B[in];W[jn];B[hn];W[gn];B[jm];W[kn];B[km];W[ll];B[jo];W[ln];B[go];W[fo]
;B[ho];W[en];B[ko];W[lo];B[io];W[nm];B[an];W[cn];B[og];W[of];B[gf];W[ej];B[ge]
;W[jf];B[ig];W[jg];B[ke];W[kf];B[if];W[bo];B[am];W[ao];B[bk];W[ma];B[la];W[];B[ ]
;TR[df:ef][de][dd][id][gc:ic][hb][jb];AE[df:ef][de][dd][id][gc:ic][hb][jb]TB[bl ][oi][gh][nh][gg][he][jc][ab][fb][ib][aa:ea][ga:ja]TW[co:eo][mo:oo][dn][fn][mn:o n][em:fm][mm][om][el:fl][ml:ol][ek][ii:ji][kh][kg:lg][mf][me:oe][nd:od][nc:oc][n b:ob][na:oa]
)


joeb331

Posts: 60
Registered: Nov, 2009

Re: Automatic dead stone markup Posted: Sep 9, 2013 11:50 PM
Reply
smiter34:

As best I can tell the issue was not with the score estimator but with unresolved situations in the game (unfinished ko at f14, white being able to live outright by forcing with c12).

pit39

Posts: 1
Registered: Dec, 2012

Re: Automatic dead stone markup Posted: Feb 23, 2014 4:21 PM
Reply
there is wrong dead stones at my game today and another times too.can't you do batter job?

nefrin

Posts: 21
Registered: Sep, 2010

Re: Automatic dead stone markup Posted: Feb 25, 2014 11:49 AM
Reply
It works fine , one just have to finish the game till the point he is sure it will work fine.

Replies: 14   Pages: 1  
 
  Back to Topic List      Top of the page