Pixel Go 101

Published 28 Oct 2020 by antti (last edited 30 Oct 2020)
tags: pixel-go

Now that the pixel go tournament is approaching, it is time to cover some basic pixel go tactics. Me and Mikkgo and a few other Finnish players have ‘played’ pixel go for over 10 years – fairly rarely, to be sure, but this has still been enough time to come up with some standard patterns.

Let’s start with the basics. If you have been following this blog, you probably know by now that in pixel go you play 2×2 ‘blobs’ of stones which may partially overlap with existing stones or go outside of the board.

Black 1 and white 2 above are examples of valid moves.

It is important to note that pixel go uses area scoring – in other words, neutral intersections count as points. This means that, whenever possible, you should aim to place the full four stones on the board, and that you do not need to be afraid of filling your own territory with your stones.

Let’s look at a sample endgame problem. Where are the biggest moves in the diagram below? You can press ‘Solution’ below the diagram to check the answer.

Solution
The values of the biggest moves are shown here. Note the 2.5-point move – Black’s move there is worth 2 points but White’s is worth 3, making for an average value of 2.5. Both players should go for a full 3-point move before playing here.

As we can see, even late endgame and dame-filling is quite different in pixel go compared to regular go.

Going back to white 2, the ‘shoulder-hit attachment’ which we saw before, this leads to one of the most common patterns in pixel go.

Very often, Black will respond with 3, which can probably be called a ‘pixel hane’, and White will do likewise with 4, resulting in an exchange that just goes on and on, for example until 9. Who is better here?

Since both players are basically forming identical sticks of stones, the shape itself probably cannot be called better for either player. Its position with relation to the whole board is key – and, probably, the fact that White is only forming two lines of territory makes this scenario uninteresting for him.

If the stick of stones were for example three lines to the left, so that White would be forming five lines of territory rather than two, he can probably be a lot happier.

Currently, nobody knows exactly how many lines of territory is a fair trade for the outer influence. This five lines is my guess, as it roughly translates to 2.5 lines in regular go – traditionally, two lines of territory in regular go is not considered much, but three lines of territory is very comfortable, so it makes a kind of sense that the average of these (which only exists in pixel go) should be right.

In any case, the shoulder-hit attachment should be used carefully, lest one end up playing in an uninteresting or small part of the board.

Another interesting response to the white shoulder-hit attachment of 1 here is the counter-shoulder-hit attachment of 2. If you look carefully, you notice that these black stones cannot be separated – they are connected by a miai. (Black has two moves to connect himself, and White cannot prevent both of them at the same time.)

This black 2 is however not good in and of itself, as . . .

. . . White can respond with 3–5 here. If you look at this shape carefully, you notice that it is symmetrical (save for the black stone in the centre) – and furthermore, White was free to choose either direction with 3, meaning that he got to develop the game how he wanted.

One more basic shape in pixel go is this kosumi-jump-keima-hybrid. Let’s call it the ‘pixel keima’. This is a solid, uncuttable shape, and can therefore be used often in games.

However, some care is needed, as . . .

. . . if two such keima are next to each other, like here, White can play a shoulder-hit attachment at 1 and actually get to cut one of them apart. There is no way for Black to defend against White’s double-threat simultaneously.

Attacking moves such as white 1 are very important in pixel go, and when you start thinking about it, there are a myriad different ways to start cutting the opponent’s positions up. This is particularly useful in that, if you can make the opponent busy defending their stones, they will not have time to attack yours – something that also applies to real go.

Continued in Pixel Go 102.


This post has no comments so far. Please log in to comment!

Events

All times are in Helsinki time (eet with summer time).


2nd Corona Cup on November 2 – December 15! Tournament details are here.


Gothenburg Online Open on December 5–6! Tournament details and registration are here.


New pairing for the online league made every 1st and 3rd Saturday of the month.


Public lectures on Twitch every 2nd and 4th Saturday of the month at 1 pm.


Jeff and Mikko stream on Twitch on Fridays at 6 pm.