The first thing I did when I learned python was to write a tetris game in it (using PyGame). So the first thing I did was look at the random piece generator in the go version. It just picks one at random:

    g.piece = 1 + rand.Int()%numTypes

However, the official tetris guidelines say that the 7 types of pieces should all be selected in a random order, as if taken from a bag. After all 7 pieces are picked, you can put all of the pieces back in the bag and select again in a random order. [1]

Of course, you don't HAVE to follow the official guidelines. I just thought it would be fun to compare.

[1] http://tetris.wikia.com/wiki/Random_Generator

Never knew this, but now that I think about it, I don't think I ever ran into a situation where I got 3 of the same pieces in a row in any Tetris game. Makes sense.

I did the same thing, implementing all the guideline rules into my JS version of tetris here: http://simon.lc/tetr.js/

However over the years, even after the guideline was instituted, there have been many different types of randomizers used. Of course, the original having none, and many a long the way having bugs.

Thanks for the tip ;)

Oh, and pull requests are welcome ;)

well, that actually is a great piece of information to be aware of, the whole in-game strategy changes!

What were your thoughts on termbox?

I've used it for a couple of projects now but I can't really get on with it. It feels so easy and straightforward at first but I often keep hitting silly annoyances that slow down debugging time or even causes fatal exceptions of its own (eg resizing a window).

So I'd be interested to hear another developers experiences with termbox

I've built a few things on top of it now: a "rich" text editor, a layout system, and a client to my plain-text board gaming service. The two main things that have caused me trouble are:

* Inability to customise colours on systems that support it. * Weak support for wide characters as termbox works directly on "cells", but can work around it with application logic. * It's quite easy to cause fatals, especially when working directly on the underlying buffer. Need to be quite careful.

The fact that it's native and cross platform is fantastic though, I'm a little worried that early design decisions will make it hard to rectify some of the warts though without large backwards-incompatible changes.

The fatals were my main gripe as well. And to make matters worse, the crash report would be unreadable because it would dump in ncurses mode (which would also break my command prompt too)

My uses were pretty simple, but I enjoyed it.

This was also one of my very first projects when I started writing Go a few years ago.

https://github.com/cespare/go-tetris

Seriously? That's funny.

Be careful calling it Tetris, the company that holds the rights I think pretty consistently goes after people who violate it's trademark and copyright.

There's a Tetris clone for Macintosh called Quinn, and at one point its website read:

"Quinn is an implementation of a popular falling-blocks game which, according to the Tetris Company, must not be named here."

They typically wont bother with small projects like this afaik. I've had a semi popular one going for a while with no contact from The Tetris Company.

I've never tried to write a game before, but that seemed to me to be a surprisingly small amount of code.

Thanks. There are a bunch of other easy games to port: http://www.javascripter.net/games/index.htm

jjinux had been a strong proponent of Python, but happy to see some Go love lately.

I'm a polyglot ;)