Another milestone in computer Go. Zen19 is the first bot to hold a 3d rating on KGS for at least 20 games. Zen19 was also the first bot to hold 1d and 2d ratings.

You can see some more historical bot ratings here:
http://senseis.xmp.net/?KGSBotRatings

It's really too bad that I can't buy a copy of Zen19 and play it at home.
I only speak English.

It's really too bad that I can't buy a copy of Zen19 and play it at home.

I only speak English.

-

and where to buy the copy please? thanks

You can (or could) buy zen at http://soft.mycom.co.jp/pcigo/tencho/index.html

But don't ask me what to press and so.

Did email Yamato (the creator of Zen) when we can expect an English version.

(or when can we buy the program at an english website)

But even then do you only buy the program on CD rom or can youi also download updates?

Isn't that a link to the first version of the software? You can find the second version here:
http://www.amazon.co.jp/毎日コミュニケーションズ-天頂の囲碁2/dp/B003XFYTTC

sorry i don't know I cannot speak or read Japanese

Ps the link doesn't work but that is maybe becouse the japanese characters

Another milestone in computer Go. Zen19 is the first bot to hold a 3d rating on KGS for at least 20 games.

Just out of curiosity, I have watched some of the computer programs playing on KGS yesterday.
I did not see Zend, but I have seen ManyFaces, ranked at 1d.

I was very surprised at the level it played. It seemed to be making plenty of completely retarded moves - and I mean REALLY RETARDED! Like pushing the losing side of a 3-move ladder which even a 15k should have seen as not working. And stuff like that. Repeatedly! And that should be the "strength" of the program, no? These few-move forced sequences which can be easily calculated. It made some moves which were not so bad, but the overall level was rather low, certainly not a 1d stuff. Not even a 5k stuff, most of the time, I'd say.

I had the impression that only the fast time controls (10 sec per move) allowed it to beat an opponent every now and then. I watched quite a few games and it only managed to win one - and that by what I assumed was a time-induced blunder by the opponent.

I watched another program which was 2k and two that were 4k - same story, pretty much.

Which makes me wonder... are these programs really playing at the advertised level?
If I buy MFoG and set it up at home, try to play a game against it on slower time controls, will it really play like a 1d? Or is that just a marketing gimmick? Is the rating skewed by the game timing or other factors (like playing a lot among each other, or something)? I do not wish to make a negative publicity for people writing Go programs here, but this is what I have observed on KGS. So I wonder...

I had the impression that only the fast time controls (10 sec per move) allowed it to beat an opponent every now and then. I watched quite a few games and it only managed to win one - and that by what I assumed was a time-induced blunder by the opponent.

It's really too bad that I can't buy a copy of Zen19 and play it at home.

I only speak English.
-

Was it behind when it played like that? That's when the really retarded moves surface.

... are these programs really playing at the advertised level?
... 1d? Or is that just a marketing gimmick?
...

The playing style of Monte-Carlo bots (Many Faces is one)
is completely different from human playing style. Therefore,
it is very normal that sometimes

(A) humans think that such a bot plays very poor moves from time to time

This is tricky.

Time controls -- as far as I know we have only one example of a single bot on the same hardware playing at under two different time controls. Manyfaces is playing at 30 minutes for all the moves and Manyfaces1 at 10 sec per move. Note that these aren't actually very different average times per move. If you look at the stats, MFOG is stronger against a human opponent under the 10 sec time control. Why? Probably because the human player sometimes needs longer to read out a position, using time saved by making other moves faster.

Bad moves, "not 1 dan moves", and similar comments. I think you need to go by the stats. If you have a large number of games between a bot and human 1 dan players (and stronger or weaker human players with the appropriate handicaps) and the bot wins half of them then it must be playing at 1 dan.

Well... it was more than just 'thinking' - these WERE retarded moves.

... it would stand to reason that a 1d programs making retarded moves so much, would also have to make a fair number of really good moves which would balance out all the blunders.
...
It might be as you say, the program views the game differently than humans, but then the question is - what is its value as a sparring and teaching partner?