Is there a reason you did this on a thread that the last post was 6 years ago? Why not just make a new thread? I did vote though. On one of them because I only remember my games against one of them even though I have played more.
Don’t be. Some will like it, others won’t. That doesn’t change no matter if it is a new topic or a new post on a thread. It will all be fine in the end. Though I think it deserves its own thread👍
I play against Kata now and again, and sometimes I would play against Leela Zero back in the day.
I played against Kugutsu because I found giving it free placement handicap interesting (I managed eventually to give it five stones.)
These days I only play against a bot if I want to get beaten very smoothly and then be surprised about it. At my level / style I find KataGo to be a very instructive influence-cum-moyo player.
I remember when the main OGS bot was RoyalLeela, with a side of DarkGo :3
I just started going down the list and seeing what options I get… Noticed quite a few bots can’t even play bigger boards, even limited to 7x7’s… After the amybotddk the next I could choose that played a 19x19 and correspondence is GNUv3.8.
I was bored so I decided to play a bot. I don’t play bots often but I was inspired from this thread, and me not wanting to worry about playing in my 19x19 or lost correspondence games, that I will not timeout in because I am careful in that perspective, but I was going to play a bot. I originally wanted to play amybot beginner because that is the only bot I remembered, but I couldn’t play a ranked game because it was a 25k and I was a 13k. I then moved on to amybot ddk. I went 0-4 and then won the last game by resignation that I thought I had lost until they played somewhere. I moved down to a 16k and then back up to a 15k. Is this because I
A. Suck at playing people my rank
B. Had an overinflated rank
C. Playing bots is harder than people
I can personally see all of them being true. A because I am in many tournaments and I can beat people sometimes but I still lose most of the games in tournaments. This can be many factors including the fact that I mainly play 9x9, which shouldn’t matter here unless this bot is very good at 9x9, or that tournaments have a wide rank gap meaning that sometimes I will play some 1d or other times I could play someone closer in rank to me. B could also be true because I beat someone I should’ve lost to today and I have recently received some timeout wins in tournaments because things come up or they forget about the game and timeout. At one point I went up one or two ranks after two timeout wins in a tournament. That just leaves option C. Is playing bots harder than playing humans? I don’t know an easy way to find out so I will just give some numbers. In the month of February, I played 200 ranked games, all against humans. I had 88 wins and 112 losses(for the sake of this math, I am rounding this to 80 wins, 120 losses). That would give me a 2:3 ratio or for every 5 games played I had two wins and three losses. My record against bots starting on February 4th(it turns out I played ranked games against bots in Feb) is 2-7. For every 9 games played against bots I get two wins. Is there a reason for this, or is it just that playing bots is harder and different then playing against people and I am just not good at playing them.
Yeah it’s harder. They just compute moves faster, ya know? They can go through a dozen or more variations by the time I go through one. I just started beating the ddk amybot and wanted to work my way down. GNUGO3.8 is def a step up from amy. I doubt I’ll win my first game with it.
At this point I’m checking on my computer so often that I can keep up with the games no prob and figured I’d try to annoy myself with bots that take constant advantage of my lowly mistakes.
GnuGo was the next homie that accepted a correspondence game with the rank restriction. GnuGo is definitely making me face palm and pull my hair out, but in a good way.