Re: [Computer-go] Time policy

I think there are some short papers about it out there. But I would suggest looking at the source code of existing bots like michi or pachi. What I use in my bot is really simple. I use the following formula: time for next move = remaining time / (C * max(vacant points, M)) Where C is some

Re: [Computer-go] Congratulations to AyaMC!

Thanks for the writeup Nick! It seems that I shouldn’t use pre-release versions of my bot for a tournament. ;) That was one horrible performance and I will have to spend some time figuring out where (and when) I messed up. Urban On Tue, Oct 11, 2016 at 12:14 AM, Nick Wedd

Re: [Computer-go] October KGS bot tournament

for the Formal division. > > Nick > >> On 3 October 2016 at 11:13, Urban Hafner <cont...@urbanhafner.com> wrote: >> How do you define “beating GnuGo”? My bot is a standard MCTS bot that is >> considerably stronger on 9x9 than it is on 19x19 and it has no problem >&

Re: [Computer-go] October KGS bot tournament

How do you define “beating GnuGo”? My bot is a standard MCTS bot that is considerably stronger on 9x9 than it is on 19x19 and it has no problem beating GnuGo on 9x9 but is much weaker than GnuGo on 19x19. Which division should I apply for? Urban On Sat, Oct 1, 2016 at 4:26 PM, Nick Wedd

Re: [Computer-go] Bug with resigned games in gogui-twogtp

Another bug that only happens with -alternate: -debugtocomment only saves the output for even games or similar. On Thu, Sep 8, 2016 at 12:41 AM, Gonçalo Mendes Ferreira wrote: > This happens in 1.4.9 too. The game result is written relative to the > original colors in the SGF and

Re: [Computer-go] Old versions of Gnu Go

You can find some of them here: http://ftp.gnu.org/gnu/gnugo/ And according to this page (https://www.gnu.org/software/gnugo/devel.html) all 3.x versions are in the git repository. Urban On Wed, Jul 13, 2016 at 5:45 PM, Andreas Persson wrote: > Anyone have or know where I

Re: [Computer-go] 25x25 experiment

On Mon, Apr 27, 2015 at 2:24 PM, Detlef Schmicker wrote: > Sorry, it is not, > > > but I offer my trained DCNN http://physik.de/net.tgz > > it has about 44% prediction rate and uses only the position > > it a a caffe file which is quite easy to use: > > Sorry to reply to a

Re: [Computer-go] Congratulations to Zen!

Hey Nick, thanks for organising this tournament each month! It’s always a lot of fun. Cheers, Urban On Sun, Jun 19, 2016 at 6:50 PM, Nick Wedd wrote: > Congratulations to Zen19X, winner of today's KGS 9x9 bot tournament with > 20 wins from 20 games! > > My report (which

Re: [Computer-go] June KGS bot tournament: 9x9

I'll participate with my rather weak bot in the open division tomorrow. I encourage other developers to participate as well. It's great fun especially when there are many participants. Urban > Am 10.06.2016 um 14:11 schrieb Nick Wedd : > > Hi Ingo, > > Thank you for

Re: [Computer-go] DarkForest is open-source now.

Brilliant! Another engine to have a look at and/or copy from. Such a great learning opportunity! Oh and some day my engine will have NN support as well. Maybe I'll be able to beat GnuGo then. ;) > Am 10.06.2016 um 20:31 schrieb Yuandong Tian : > > Hi all, > >

Re: [Computer-go] Leela 0.6.2, OpenCL support, including GTP engines

Thank you so much Gian-Carlo for the GTP only version! Should my bot ever outgrow GnuGo (it can barely beat GnuGo on 13x13 right now) it will be great to have another opponent available. Urban On Sun, Jun 5, 2016 at 1:20 AM, Gian-Carlo Pascutto wrote: > Hi all, > > I've done a

Re: [Computer-go] new bot friendly go server.

It's always great to have more options. I will give it a try at some point. One question that's not totally related to your server. How is scoring generally handled when it comes to the clock? On your server the clock seems to run even when running final_status_list and final_score when if I

Re: [Computer-go] Commonly used neural network architectures

s to work well, and pooling is not used (for obvious > reasons). > > > > In my experiments it is essential to have the first layer filters larger > than 3x3. In higher layers 3x3 seems to work fine. > > > > Hope this helps. > > > > David > > > > *From:* C

[Computer-go] Commonly used neural network architectures

Hey there, just like everyone else I’m currently looking into neural networks for my go program. ;) Apart from the AlphaGo paper where I can I find information about network architecture? There’s the network from April 2015 from Detlef

Re: [Computer-go] KGS bot tournaments: structure

That sounds like a good compromise to me. And the lower devision would hopefully encourage other participants with weaker bots so that Iomrascálaí and matilda don’t feel so alone. ;) Urban On Tue, May 10, 2016 at 9:15 AM, Nick Wedd wrote: > At present, KGS bot tournaments

Re: [Computer-go] Congratulations to Zen!

cumbersome. Urban On Tue, May 10, 2016 at 9:32 AM, Nick Wedd <mapr...@gmail.com> wrote: > > > On 10 May 2016 at 07:57, Urban Hafner <cont...@urbanhafner.com> wrote: > >> My impression was that you can only get the status of a ranked bot if >> it’s somehow “stable”.

Re: [Computer-go] Congratulations to Zen!

My impression was that you can only get the status of a ranked bot if it’s somehow “stable”. So that would mean that I would have to have a new account for each new version of the bot, wouldn’t it? Especially with early stage bots like mine the jumps in strength can be quite large. But on the

Re: [Computer-go] Congratulations to Zen!

Thanks for organising this Nick! Even though Iomrascálaí came second to last it was very useful as I could figure out how to make it play on KGS. :) And like Ingo already said, it played against abakus in the second round. Also, you give me too much credit. I’m not the primary author of HouseBot,

Re: [Computer-go] Problems with the kgsGtp client

Never mind! I just had to remove the kgs-chat command and it worked. On Sun, May 8, 2016 at 6:37 PM, Urban Hafner <cont...@urbanhafner.com> wrote: > Hey everyone, > > I’m currently participating in my first KGS computer go tournament but the > client crashes on me. I use versi

Re: [Computer-go] May KGS bot tournament

May 2016, at 10:09 AM, Nick Wedd <mapr...@gmail.com> wrote: >> >> I would happily run more than one division (which the KGS scheduler will >> regard as separate events). Is there enough interest to justify a second >> division? >> >> Nick >> &g

Re: [Computer-go] May KGS bot tournament

r they get to play against GnuGo or Urban's program. > > By the way - I haven't yet received any entries for this Sunday's tournament. > I hope my message reached the list. > > Nick > >> On 4 May 2016 at 10:01, Urban Hafner <cont...@urbanhafner.com> wrote: >> I’m cons

Re: [Computer-go] May KGS bot tournament

of (i.e. have met the entry > >>> requirement of "consistently beats GnuGo version X.Y) might be a nice > >>> filter as the number of those desiring to participate (with weaker > bots) > >>> rises. > >>> > >>> On Wed, May 4, 2016 at 4:0

Re: [Computer-go] May KGS bot tournament

I’m considering entering with my bot but it’s rather weak (a lot weaker than GnuGo on 19x19) so I don’t know if it makes sense. Unless of course other weaker bots were willing to enter as well. If no one is interested in this (or if it’s even discouraged by Nick) then I would refrain from entering

[Computer-go] GoGui analyse commands vs live graphics

I’m at the point where it would be nice to see what the engine is “thinking” while playing. If you just play against the engine on GoGui then you can use the analyse commands facility and trigger one of your custom commands after each move played. Once you move away from GoGui you seem to loose

Re: [Computer-go] CGOS again

On Fri, Nov 20, 2015 at 3:41 PM, Joshua Shriver wrote: > Thank you so much for running CGOS! Looks like it's running very > smoothly and you are doing a great job. > Yes! Thanks again. It’s great to have 9x9 and 19x19 versions again with some strong bots playing. Great to

Re: [Computer-go] CGOS again

Very cool, thank you! I will see if I can start running my bot on there soon. Urban On Tue, Nov 10, 2015 at 2:11 PM, Hiroshi Yamashita wrote: > Hi, > > I have started CGOS on my VPS(Virtual Private Server). > 19x19 and 9x9 are running. > In 9x9, komi is 7.0. Draw is 0.5 win

Re: [Computer-go] alarming UCT behavior

This is expected. Search for the horizon effect related to computer go (e.g. in Petr Baudis thesis http://pasky.or.cz/go/prace.pdf) On Fri, Nov 6, 2015 at 8:48 PM, Gonçalo Mendes Ferreira wrote: > That's certainly possible, but assuming the tree is being limited on > number of

Re: [Computer-go] AMAF/RAVE + heavy playouts - is it save?

To make matters more difficult I assume that this also depends on the exact node evaluation you’re using. There’s UCT + RAVE, then there’s just RAVE (as used by Michi). And then you can add other things in there as well like criticality (like Pachi, and at least at one point CrazyStone). I

[Computer-go] Understanding statistics for benchmarking

So, I’m currently running 200 games against GnuGo to see if a change to my program made a difference. But I now wonder if that’s enough games as I ran the same benchmark with the same code (but a different compiler version) and received different results: 85.5% wins (171 games of 200) the first

Re: [Computer-go] Understanding statistics for benchmarking

nteresting statistical tool: > https://chessprogramming.wikispaces.com/LOS+Table > > For more advanced tools for deciding when to stop testing, there is SPRT: > http://www.open-chess.org/viewtopic.php?f=5=2477 > https://en.wikipedia.org/wiki/Sequential_probability_ratio_test > > Rémi > > &

Re: [Computer-go] Understanding statistics for benchmarking

On Tue, Nov 3, 2015 at 2:22 PM, Petr Baudis wrote: > (The situation is a bit dire if you have limited computing resources. > I admit that sometimes I didn't follow the above myself in less formal > exploratory experiments, but at least I tried to look only > "infrequently", e.g.

Re: [Computer-go] Understanding statistics for benchmarking

Yes, I noticed that too. But luckily that's the one thing I didn't even consider doing. Running the same number of games feels like the most natural thing to do anyway. Von meinem iPhone gesendet > Am 03.11.2015 um 14:22 schrieb Petr Baudis : > >> On Tue, Nov 03, 2015 at

Re: [Computer-go] Codecentric Go Challenge is over! FJD won!

The IT world in Germany is so small. Just last week I was at the Munich Codecentric offices for the first Munich Rust meetup. And I’m writing a Go AI in Rust. ;) Urban On Sun, Nov 1, 2015 at 8:55 PM, Tobias Pfeiffer wrote: > Hi everyone, > > haven't seen this on the mailing

Re: [Computer-go] Codecentric Go Challenge is over! FJD won!

On Mon, Nov 2, 2015 at 10:07 AM, "Ingo Althöfer" <3-hirn-ver...@gmx.de> wrote: > Hi Urban, > > that is indeed a nice "small world" example. > > Did you meet Dr. Georg Snatzke in the "rust meetup"? He is working for > codecentric and is the connection man to the go scene (Georg is 3-dan >

Re: [Computer-go] Understanding and implementing RAVE

On Tue, Oct 27, 2015 at 6:57 AM, David Fotland wrote: > Many Faces uses 2200 for RAVE_EQUIV. I found that anything between 2000 > and 3000 was about the same, and CLOP recommended 2200. 1000 was a little > worse, and 500 was much worse. In discussions with other

Re: [Computer-go] Understanding and implementing RAVE

On Mon, Oct 26, 2015 at 5:51 PM, Gonçalo Mendes Ferreira wrote: Well I took the liberty to steal and very crudely modify a MCTS diagram > from wikipedia: > > http://pwp.net.ipl.pt/alunos.isel/35393/mcts.png > > Maybe with images it is clearer. You seem to be using an acyclic graph

[Computer-go] Understanding and implementing RAVE

With the help of Michi (thank you Petr!) I’m currently working on adding RAVE to my UCT tree search. Before I get too deep into it I’d like to make sure I actually understand it correctly. It would be great if you could have a quick look at my pseudo code (mostly

Re: [Computer-go] Understanding and implementing RAVE

On Mon, Oct 26, 2015 at 1:20 PM, Gonçalo Mendes Ferreira wrote: > I'm not sure about "if color of that particular (child) node was the first > to play on that intersection in the playout". What I think the authors > meant was to only increment the AMAF statistic once because of ko

Re: [Computer-go] Understanding and implementing RAVE

On Mon, Oct 26, 2015 at 1:12 PM, Petr Baudis wrote: This RAVE formula has been derived by David Silver as follows > (miraculously hosted by Hiroshi-san): > > http://www.yss-aya.com/rave.pdf > > Also note that RAVE_EQUIV (q_{ur} * (1-q_{ur}) / b_r^2) varies widely > among

Re: [Computer-go] Understanding and implementing RAVE

On Mon, Oct 26, 2015 at 5:03 PM, Gonçalo Mendes Ferreira wrote: > I think we mean the same. So when black plays at a1 first we count it for >> black everywhere but never for white. Correct? >> > The question is, if both black and white play a1 - because there were > captures in

Re: [Computer-go] Playout speed... again

Very nice speed up Goncalo! This prompted me to check my playout speed and with the recent addition of 3x3 patterns I'm now down to 6pps on 19x19. Whoops! ;) I may need to do some optimisations ... Urban On Fri, Oct 16, 2015 at 1:09 AM, Gonçalo Mendes Ferreira wrote: > Thanks

Re: [Computer-go] Javascript go programs

Hey there, there are none that I'm aware of. But if you're looking for ideas I'd suggest having a look at https://github.com/pasky/michi. It's written in Python but should be fairly easy to read as the code base is so small. I'm still in the process of implementing/stealing stuff from Michi for

Re: [Computer-go] CGT endgame solver

On Tue, Jul 14, 2015 at 9:22 AM, Ingo Althöfer 3-hirn-ver...@gmx.de wrote: * Question to Igor: You are with us now for some time already. Do you have a bot in the meantime? Would you be willing to let it participate in the EGC bot tournament? As Igor is living in

Re: [Computer-go] How about a 39x39 tournament?

That sounds fun. But we should figure out how KGS implements the coordinates for such large boards. On Mon, Apr 27, 2015 at 10:40 AM, Nick Wedd mapr...@gmail.com wrote: I find that I can create KGS tournaments on sizes of board up to 38x38, but not 39x39 and larger. I will gladly do so if

Re: [Computer-go] CGOS future

On Thu, Apr 9, 2015 at 4:14 PM, Detlef Schmicker d...@physik.de wrote: Hi, I just started playing on 13x13 again, very busy :) It is you that's running all the strong bots? If so, thank you! My second copy of GnuGo had reached an ELO rating of almost 1900 these last days, but now it's back

Re: [Computer-go] What's a good playout speed?

On Sat, Mar 28, 2015 at 11:51 AM, Petr Baudis pa...@ucw.cz wrote: On Sat, Mar 28, 2015 at 11:45:48AM +0100, Urban Hafner wrote: Good to know Petr! Where does the strength come from? Sophisticated playouts or a search algorithm or both? Frankly, I don't know for sure! The downside

Re: [Computer-go] What's a good playout speed?

On Tue, Apr 7, 2015 at 12:56 PM, Petr Baudis pa...@ucw.cz wrote: Hi! On Tue, Apr 07, 2015 at 12:03:12PM +0200, Urban Hafner wrote: Now that I have a bit of time again, what would be a good starting point to improve upon UCT and light playouts? RAVE definitely comes to mind, as well

Re: [Computer-go] What's a good playout speed?

On Tue, Apr 7, 2015 at 12:20 PM, Hideki Kato hideki_ka...@ybb.ne.jp wrote: For prior values in the tree, almost(?) all strong programs use Remi's method these days. http://remi.coulom.free.fr/Amsterdam2007/MMGoPatterns.pdf Thank you! I will put that one on my reading list! Urban -- Blog:

Re: [Computer-go] What's a good playout speed?

On Tue, Apr 7, 2015 at 4:04 PM, Christoph Birk b...@obs.carnegiescience.edu wrote: On Apr 7, 2015, at 4:34 AM, Urban Hafner cont...@urbanhafner.com wrote: I suspected you'd say something like this. ;) It is definitely on my list of things to steal a few things from Michi. But maybe I'll

Re: [Computer-go] What's a good playout speed?

On Tue, Apr 7, 2015 at 4:44 PM, Christoph Birk b...@obs.carnegiescience.edu wrote: thanks, I agree 1400 is about as far as simple UCT will get you. My simple UCT implementation (myCtest-xxk-UCT) gets about 1200, but it does not do any adjustments to the number of playouts per move depending

Re: [Computer-go] CGOS future

Joshua, I think you should just pick the language(s) that keep YOU motivated. The only hard requirement seems to be that you need the clients to work on Windows, Linux, and Mac OS X. A web framework seems to exist in every language these days so that doesn't really narrow down the selection. I

Re: [Computer-go] April KGS bot tournament, 13x13

On Wed, Apr 1, 2015 at 1:29 PM, folkert folk...@vanheusden.com wrote: I would like to as well but when I asked (somewhere in 2012 iirc) I was told that it should first become a bit better :-) Yes, I talked to Nick about it and the general rule seems to be that the bot should have a reasonable

Re: [Computer-go] April KGS bot tournament, 13x13

Thank you for organising this for so many years, Nick! I think the email is a bit wrong though. Here's the link to the actual tournament: http://www.gokgs.com/tournInfo.jsp?id=955 It's also not 19x19, but (at least according to the KGS page) it's 13x13. :) Now on to my actual question: Are any

Re: [Computer-go] April KGS bot tournament, 13x13

status for a stable version of it, I can encourage the admin who can assign this status. Best wishes, Nick On 31 March 2015 at 13:00, Urban Hafner cont...@urbanhafner.com wrote: Thank you for organising this for so many years, Nick! I think the email is a bit wrong though. Here's

Re: [Computer-go] Learning from CGOS

Awesome Joshua! I agree with the others. Start open sourcing it right away. That's what I did with my Go bot that I started writing in a language I didn't know. And people (well, two ;)) just decided to help out. As for features. Well, I'd be happy if you just reimplemented CGOS and it were

Re: [Computer-go] fast + good RNG

29.03.2015 um 12:29 schrieb Álvaro Begué alvaro.be...@gmail.com: If your PRNG is consuming 40% of your CPU time, your playouts are too light. That's what I was going to say, too. My program isn't the fastest (5k pps on 9x9) and the RNG never even appeared in the profiling output. Urban

Re: [Computer-go] What's a good playout speed?

But my guess based on ad hoc tests during the development is that the contribution of basic playout heuristics and RAVE+priors may be about 1:1 (with large pattern priors giving further extra boost). I'll have to start reading papers on those concepts then, I guess. :) Thanks Petr!

Re: [Computer-go] What's a good playout speed?

if the computer could learn the heuristics itself, by self-play. Which is why my bot is still stuck on 25k rating :-D At least, that's my excuse :-P On Sat, Mar 28, 2015 at 7:18 PM, Urban Hafner cont...@urbanhafner.com wrote: But my guess based on ad hoc tests during the development

Re: [Computer-go] monte carlo search; all valid moves?

Exactly you don't check super ko in the playouts and you generally stop playouts at 3 * boardsize * boardsize. But once you start adding more logic into the playouts (like not playing in single point eyes) they tend not to hit this limit. Von meinem iPhone gesendet Am 28.03.2015 um 11:53

Re: [Computer-go] What's a good playout speed?

On Sat, Mar 28, 2015 at 2:34 PM, Hugh Perkins hughperk...@gmail.com wrote: Well, what I did was connect my bot to kgs a few times, and watch as it go repeatedly beaten by anything much better than 25k :-) Normally, there's a few 'randombots' there to start with. Once you can beat those

Re: [Computer-go] What's a good playout speed?

On Sat, Mar 28, 2015 at 6:14 PM, Urban Hafner cont...@urbanhafner.com wrote: On Sat, Mar 28, 2015 at 2:34 PM, Hugh Perkins hughperk...@gmail.com wrote: Well, what I did was connect my bot to kgs a few times, and watch as it go repeatedly beaten by anything much better than 25k :-) Normally

Re: [Computer-go] monte carlo search; all valid moves?

20 playouts is a bit slow Folkert. But I've been there, too! Now my bot is at around 1000 19x19 pps and 5000 on 9x9 (but I also check to not play into single point eyes). That's still quite slow and further optimisation will be required. But OTOH on a 4 core machine without UCT (only the AMAF)

Re: [Computer-go] What's a good playout speed?

/computer-go/ On Wed, Jan 14, 2015 at 09:14:24AM +0100, Urban Hafner wrote: ..snip.. Right now I'm at ~2000pps on 9x9 and ~1000pps on 19x19. This is for playouts with simple ko and suicide checks and no concurrency. Now I wonder if this is fast enough to even start thinking about

[Computer-go] How to correctly use GoGui live graphics?

Hey there, I'd like to use the live graphics feature of GoGui to see what the engine is thinking during a game. I played around with it a bit, but I think I'm doing it wrong. I always run into the problem that the output I want to display gets cleared right away by GoGui. I understand why that's

Re: [Computer-go] [ANN] Michi - 15x15 ~6k KGS in 540 lines of Python code

On Wed, Mar 25, 2015 at 4:57 PM, Petr Baudis pa...@ucw.cz wrote: I'm afraid you won't get a good inspiration about properly implementing a board structure. That would make the code a lot more complicated and much longer! And, uh, the program actually does not implement UCT. ;-) It uses RAVE

Re: [Computer-go] [ANN] Michi - 15x15 ~6k KGS in 540 lines of Python code

Awesome, Petr. I haven't programmed in Python for a long time (I like Ruby better), but I think I should be able to understand it without a problem. It seems like a good starting point to see how UCT is implemented (I still haven't gotten around to it for my bot https://github.com/ujh/iomrascalai)

Re: [Computer-go] Weak bots to run on CGOS

Thanks Christoph! On Fri, Mar 20, 2015 at 4:17 PM, Christoph Birk b...@obs.carnegiescience.edu wrote: On Mar 20, 2015, at 5:11 AM, Urban Hafner cont...@urbanhafner.com wrote: So, I now have a new version of my bot running on CGOS ( http://cgos.boardspace.net/13x13/cross/Imrscl-016

Re: [Computer-go] Weak bots to run on CGOS

On Tue, Mar 10, 2015 at 4:29 PM, folkert folk...@vanheusden.com wrote: I have a small cluster of raspberry pi-s (7) that normally run a distributed chess engine. Since they're mostly idle, I can a couple of Go engines on them if you like. They're the 512MB rpi1 models. There's +/- 256MB of

Re: [Computer-go] Observing a game in GoGui that's running on KGS/CGOS

as the engine for the CGOS or KGS client and then run gogui-client locally to observe? Urban On Mon, Mar 9, 2015 at 8:53 AM, Urban Hafner cont...@urbanhafner.com wrote: Yes, that it! Thank you, Rémi. On Sun, Mar 8, 2015 at 6:12 PM, Rémi Coulom remi.cou...@free.fr wrote: gogui-display: http

Re: [Computer-go] implementing simple ko

You only need to keep track of one point. If the last move was a single stone capture, then that point is the Ko point for the next move. A play by the opponent on that point is disallowed unless it's a capturing move. Urban On Thu, Mar 12, 2015 at 6:07 AM, Ray Tayek rta...@ca.rr.com wrote: i

Re: [Computer-go] Tromp Taylor rules http://senseis.xmp.net/?LogicalRules

As Alvaro said, you should just implement Tromp Taylor and if you want to play on CGOS or KGS (chinese rules only) you just need to make suicide illegal and you're good to go. Urban On Wed, Mar 11, 2015 at 12:44 PM, Álvaro Begué alvaro.be...@gmail.com wrote: Ko is not missing: It is a

Re: [Computer-go] Weak bots to run on CGOS

On Mon, Mar 9, 2015 at 3:50 PM, Christoph Birk b...@obs.carnegiescience.edu wrote: myCtest-xxk is a pure random player. myCtest-xxk-UCT adds a tree, nothing else. Thank you Christoph. It's just that sometimes your bots aren't running so I'd like to run some other bots myself, too. Urban --

Re: [Computer-go] Weak bots to run on CGOS

On Mon, Mar 9, 2015 at 7:48 PM, folkert folk...@vanheusden.com wrote: `Stop' is definately between brown and maybe a tiny bit better than that. http://www.vanheusden.com/stop/ I'll see if I can connecct it to the 13x13 server. Thank you Folkert! -- Blog: http://bettong.net/ Twitter:

Re: [Computer-go] Computer-go Digest, Vol 62, Issue 8

Thanks, I'll check it out! Urban On Mon, Mar 9, 2015 at 11:31 AM, Pawel Koziol pp.koz...@gmail.com wrote: On Mon, Mar 09, 2015 at 10:08:04AM +0100, Urban Hafner wrote: I'm currently running Brown (random bot) and GnuGo on CGOS 13x13. Mainly to get a feel for the strength of my own bot

Re: [Computer-go] Weak bots to run on CGOS

, Urban Hafner wrote: I'm currently running Brown (random bot) and GnuGo on CGOS 13x13. Mainly to get a feel for the strength of my own bot. And my bot is really bad. ;) So bad that it looses all games against GnuGo, but wins all games against Brown. So, the rating is a bit useless I

Re: [Computer-go] Observing a game in GoGui that's running on KGS/CGOS

Yes, that it! Thank you, Rémi. On Sun, Mar 8, 2015 at 6:12 PM, Rémi Coulom remi.cou...@free.fr wrote: gogui-display: http://gogui.sourceforge.net/doc/reference-display.html On 03/08/2015 10:04 AM, Urban Hafner wrote: Hey everyone, I'd like to display some statistics from my bot

[Computer-go] Weak bots to run on CGOS

Hey everyone, I'm currently running Brown (random bot) and GnuGo on CGOS 13x13. Mainly to get a feel for the strength of my own bot. And my bot is really bad. ;) So bad that it looses all games against GnuGo, but wins all games against Brown. So, the rating is a bit useless I assume as there are

Re: [Computer-go] Weak bots to run on CGOS

Thanks Erik! Yes, that will work. Urban On Mon, Mar 9, 2015 at 10:35 AM, Erik van der Werf erikvanderw...@gmail.com wrote: Perhaps AmiGo http://amigogtp.sourceforge.net/ On Mon, Mar 9, 2015 at 10:08 AM, Urban Hafner cont...@urbanhafner.com wrote: Hey everyone, I'm currently running

[Computer-go] Observing a game in GoGui that's running on KGS/CGOS

Hey everyone, I'd like to display some statistics from my bot as an overlay for the current board. Using live graphics on GoGui is probably the easiest way to do it. I just need to run the bot in GoGui and emit the proper commands on stderr. Now my question is if there's a way to do that for

[Computer-go] Chinese vs Tromp-Taylor rules

Hey everyone, I'm still at the early stages of development of my bot https://github.com/ujh/iomrascalai. Right now it only supports Tromp-Taylor rules as this is what CGOS uses (well, almost). This is a bit problematic, as for one my bot and GnuGo seldomly agree on a score (even when running

Re: [Computer-go] Chinese vs Tromp-Taylor rules

, 2015 at 10:25 AM, Urban Hafner cont...@urbanhafner.com wrote: Hey everyone, I'm still at the early stages of development of my bot https://github.com/ujh/iomrascalai. Right now it only supports Tromp-Taylor rules as this is what CGOS uses (well, almost). This is a bit problematic, as for one

Re: [Computer-go] Chinese vs Tromp-Taylor rules

Hey Petr, thanks for all the tips! There's still so much to do, which is quite a lot of fun! ;) Urban On Fri, Mar 6, 2015 at 8:12 PM, Petr Baudis pa...@ucw.cz wrote: On Fri, Mar 06, 2015 at 10:25:14AM +0100, Urban Hafner wrote: Now my question is what exactly do I have to do to implement

[Computer-go] Failures when playing on CGOS (connection timed out)

Hey everyone, I'm currently running my bot on CGOS, and recently I get the following failure after a while: error reading sock7: connection timed out while executing gets $sock s (while body line 22) invoked from within while {1} { set err [catch {set sock [socket $cgos_server

Re: [Computer-go] cgos.computergo.org down?

On Tue, Mar 3, 2015 at 9:57 PM, Joshua Shriver jshri...@gmail.com wrote: Thanks for the heads up *sigh* I really am contemplating re-writing CGOS from scratch. TCL is just quirky as hell. Plus would make updates easier and attract other developers or contributions. If you could find the

Re: [Computer-go] cgos.computergo.org down?

On Mon, Mar 2, 2015 at 4:58 PM, Joshua Shriver jshri...@gmail.com wrote: I recycled them, hopefully the 13x13 will be back up at least. Unfortunately it seems that it didn't work. It still shows games in progress from 2015-02-07. Urban -- Blog: http://bettong.net/ Twitter:

Re: [Computer-go] cgos.computergo.org down?

Yes, it did move back there. Unfortunately even when your bots can connect there they won't play any games. The 13x13 used to run until the beginning of the month, but sadly that one has stopped working, too. I wonder if there's anything we as the community can do to help? Urban On Sun, Mar 1,

Re: [Computer-go] Best resources to get started with UCT?

On Mon, Jan 19, 2015 at 12:37 PM, hughperkins2 hughperki...@gmail.com wrote: (by the way, i was getting 84000 playouts a second on an amazon c3.8xlarge, in case you want to compare, which got me around 25-20kyo... ) Thanks! That's a great data point as it should be easy to run my own bot on

[Computer-go] Best resources to get started with UCT?

Hey everyone, the playouts of my bot are finally (somewhat) fast enough to start implementing UCT and I wonder what the best resources are. There's the Senseis page http://senseis.xmp.net/?UCT as well as the original paper, but I wonder if there are other papers out there that are newer and maybe

Re: [Computer-go] Best resources to get started with UCT?

Hey there, thanks, I'll check out Petr's thesis. That seems like a good overview of the whole topic. And as far as what to implement. Yes, UCT can only be a first step, but as I do this in my free time I have to do it is small chunks. So first UCT, then maybe RAVE, Mogo style patterns, etc.

Re: [Computer-go] CGOS back online

On Sat, Jan 17, 2015 at 10:38 AM, Detlef Schmicker d...@physik.de wrote: Seems a good idea to me. It is a quasi standard in publishing, so why not set Gnugo-3.7.10 at level 10 to 1800ELO on every board size?! Why 3.7.10 and not 3.8? IIRC all Gnu projects use the odd numbered point releases as

Re: [Computer-go] CGOS back online

On Fri, Jan 16, 2015 at 8:44 AM, Urban Hafner cont...@urbanhafner.com wrote: On Fri, Jan 16, 2015 at 12:13 AM, valky...@phmp.se wrote: I just connected Valkyria on 13x13! And last night I connected Brown to just see how it works. :) It seems however that something is wrong again

Re: [Computer-go] CGOS back online

On Fri, Jan 16, 2015 at 5:42 PM, Detlef Schmicker d...@physik.de wrote: I also set up a 13x13 client. Seems to work fine, but rating is off I think. I will let it up for a while, hopefully some anchors coming up:) Yes, without at least one anchor the whole rating will be problematic. How

Re: [Computer-go] CGOS back online

1800 was for 9x9 as far as I remember, I do not remember what on 13x13. On 19x19 it was GnuGo 3.7 level 10 I think and from some historic results I guess 13x13 gnugo 3.7 level 0 was set to ELO1800 Am 16.01.2015 um 18:31 schrieb Urban Hafner: On Fri, Jan 16, 2015 at 5:42 PM, Detlef Schmicker d

Re: [Computer-go] CGOS back online

On Fri, Jan 16, 2015 at 7:45 PM, Urban Hafner cont...@urbanhafner.com wrote: I see. I wonder what the correct options for GnuGo are. gnugo --mode gtp --level 0 --chinese-rules --positional-superko --capture-all-dead seems like a good start, right? I will give that a try now ... It seems

Re: [Computer-go] CGOS back online

On Fri, Jan 16, 2015 at 12:13 AM, valky...@phmp.se wrote: I just connected Valkyria on 13x13! And last night I connected Brown to just see how it works. :) It seems however that something is wrong again this morning. Urban -- Blog: http://bettong.net/ Twitter: https://twitter.com/ujh

Re: [Computer-go] CGOS back online

On Thu, Jan 15, 2015 at 6:45 PM, Christoph Birk b...@obs.carnegiescience.edu wrote: On Jan 15, 2015, at 9:35 AM, Joshua Shriver jshri...@gmail.com wrote: Aye I'm still tinkering with it, and trying to get anchors on. Still having issues. :( The 13x13 server is up and running, Indeed

Re: [Computer-go] CGOS back online

On Wed, Jan 14, 2015 at 5:45 PM, Christoph Birk b...@obs.carnegiescience.edu wrote: http://cgos.boardspace.net/9x9/standings.html was updated last about 2 years ago. I noticed that, too. Also, it seems like there are still games in progress from 2012 which is rather unlikely. So it seems

Re: [Computer-go] CGOS back online

Great! Thank you Joshua. For me personally that will be a great help as testing if my new bot works is so much easier on CGOS than during a KGS tournament. :) Urban On Wed, Jan 14, 2015 at 10:14 AM, Joshua Shriver jshri...@gmail.com wrote: CGOS is back on boardspace.net and running. Please

Re: [Computer-go] CGOS back online

BTW, which GTP commands are required for a bot to play on CGOS? The basic commands are probably not quite enough, right? At least time_left and time_settings should be implemented if I remember it correctly. Urban On Wed, Jan 14, 2015 at 10:43 AM, Urban Hafner cont...@urbanhafner.com wrote

Re: [Computer-go] What's a good playout speed?

On Wed, Jan 14, 2015 at 1:32 PM, Petr Baudis pa...@ucw.cz wrote: Note that the mistake I did with Pachi is that I worked hard to optimize the Go board (playout) code with minimal functionality to perform a playout correct by the rules, only then discovering how much other information I need

  1   2   >