Life In 19x19

First of all I'll agree with those who say turn-based Go is not good for total beginners. When you are 30k you just need to play a lot to build up experience and get a feeling for what works and what doesn't. With turn-based Go the feedback loop for learning between doing things and finding out if they make you win is too long. But once you are maybe 20k+, I think it can be useful.

Let me illustrate with an example: me. I started playing Go around September 2005, playing a few games a day on KGS. Around 15k I slowed down my play to 1 hour main time. As well as playing; reviews from the kind folks in the KGS Teaching Ladder, shygost's lectures, Sensei's Library, real-life club, and books helped me improve. I got to 1 dan around 1.5 years later. By then my progress was leveling out, gone were the joyous DDK days of gaining a rank every few weeks. I went on a Go trip to China for 2 months in summer 2007 (I skipped my university graduation!) and went up to 3d (my rank graph is a pretty good advert: http://homepage.ntlworld.com/andrew.j.s ... graph2.png).

After that I got a job and no longer felt like putting in the time/effort for a whole serious game in the evening after a hard day at work, so switched most of my playing to the turn-based OGS. In the 4 years since I've gone up to 6d OGS and the top player there (though a few strong players have joined recently). However, I'm still only 4d KGS and 3d EGF.

So, what conclusions to make? My switch from shooting up through the ranks to slow improvement correlates near exactly with my switch from real-time to turn-based play. So is this a damning indictment of turn-based play? It may seem like it, but I don't think so. If I had continued to play on KGS, I don't think I would be 9d by now; as you get stronger it gets harder to get stronger. There are a few lucky people, such as kghin or artem92, who kept on rocketing through the dan ranks, but I don't think I'm as talented as them. I feel like low dan is about the level of my innate talent at this game, and to get stronger than that requires a lot more work than getting here (is this realism or a defeatist attitude that holds me back?).

However, in the years since I switched to OGS, I do feel like I have got better at Go: my Go has matured and my direction of play and strategy have improved. However, I don't think my reading has improved commensurately. This is because on OGS I get lazy and instead of reading, I play out variations on the computer. Another thing I've not gotten much better at is winning real-life tournament games. I play slowly, and am prone to playing blunders in overtime. OGS doesn't help me eliminate these.

So my skill on OGS is higher than my real-life play. As an example, there is one player, an AGA 5d (which is about the same as EGF 3d), against whom I lost my first game, but I've won all 12 since. This doesn't mean I'm lots stronger than him and if we played on the board I could lose to him, but on OGS where I can spend a long time analysing (sometimes I probably end up spending as much time on OGS as it would take to play a whole serious game on KGS!), I can control the game and win. There is another player, an AGA 7d that would probably beat me quite badly in real-time, but on OGS I can hold my own against him. But I do think my real-life play has improved too, just not as much (e.g. I have beaten Matthew Macfadyen 6d, 25-time British Champion, which I don't think I would have been able to do immediately after going to China).

I also find playing on OGS a useful driver of study: I try out new pro openings, see what pros did in similar shapes etc.

As one final point, I think my playing turn-based Go not only helps me improve, but could help others improve too as it makes me a better teacher. As I spend a lot of time on my OGS games I can remember various shapes from many of them and thus am building up a mental database of hundreds of my games. When I am giving lessons I am able to bring up example games invloving the various ideas being discussed to better explain them in a real-game context.

I think turn based games can be very useful and a great way to squeeze in a lot more go into a busy life. This is especially true if you have a smart phone. Both DGS and OGS have android apps (not sure about ios). These apps let me squeeze in go as I wait for my girlfriend to get ready, go #2, am in the backseat of a car, on the train, etc... I think more go can only mean more improvement. I also get really invested in the turn-based games and find them highly entertaining.

So there seems to be two schools of thought here: play a few games very seriously and analyse them thoroughly or play a lot of games concurrently to maximise exposure to patterns, shapes and so on. Would anyone care to expand on these two conflicting ideas? Couldn't one just do both with a bunch of fast games going and a couple of slow games and take a hybrid approach?

In language learning (which I don't think is that much unlike Go), it's usually taken as true that more exposure of whatever caliber is better, up to a point. I think that's still definitely true at my (low DDK) level--I'm improving just by playing and reviewing games. Most of the quibbles are about what is optimal, but this seems mostly beside the point in real life.

I've been playing turn-based games over the past couple of months, but am going to stop once my current batch of games runs out. It doesn't really work for me for a couple of reasons: Giving me days to think doesn't really help me--my thoughts aren't really deep enough it turns out, nor do I have the patience to exhaustively analyse open positions or every possible variation--I am sort of unhappy with myself every time I give up and move, bc I still could've taken more time; and also that I find it difficult to turn on my thinking about Go when a move comes in, reorient myself with the game, make my move (suspecting I didn't think about it hard enough), and then to just turn off my interest in it for however long it takes my opponent to move.

For me, it seems that right now it works much better for me that I get a certain finite amount of time to play the whole game, do the best I can, review it, and put it in the past, but it's more a matter of temperment than anything else, I suspect.

Boidhre wrote:Couldn't one just do both with a bunch of fast games going and a couple of slow games and take a hybrid approach?

Yes

I'm a Go newbie and only played a few games against the computer. Usually I cannot afford the time for realtime games. So I registered at DGS, which I love... I play a few concurrent games now and I'm sure I will get more experience from this instead of not playing at all.

Hmm, I seem to be improving from turn based, or at least my regular sdk opponents on KGS and at the club tell me so. I won't claim to have gained 2 stones on KGS yet as I haven't played enough games for me to consider the ranks solid but people have been noting an improvement in my live play with them. The improvement hasn't been massive, and maybe I'd have improved faster if those games had been real time games but there seems to be something going on.

I play a lot of games though turn based though, I can fit them in pretty easily to my day. Real time games on the other hand are a much trickier proposition.

For me it's a false dichotomy.

I play DGS on the phone during lunch breaks, commutes, etc. I wouldn't play a kgs game at those moments.

I play KGS when I have 1h+ of consecutive free time (never). I wouldn't dedicate an hour to DGS, among other things because I'd have to be playing fifty simultaneous games to have something to do for an hour.

I don't care which one is better for improvement because I wouldn't be able to replace one for the other even if I wanted to.

I really play on turn based servers to improve and there is a simple reason: I tend to play to fast in tournament games. There are only 2 other players in my city and from them I claim to be the strongest by a few stones. Therefore a long time my only "serious" games were on KGS and tournaments. The result was that I was playing far to fast in tournaments and nearly never used more than half a hour while my opponents were in byo-yomi. With playing more on OGS I learned to think more about my moves. You can see that improvement on OGS: Long time I was 8k-7k there, now I'm around 4k and have the feeling I finally can brake through to the real (tournament) life 1d.