I have even gone as far as replacing loops with either a makelist or makemat (double loop) even if I'm not generating lists or matrices. Simply create a procedural function that returns garbage data and don't store the resulting list or matrix. Have the procedural function do what you would normally do in a loop.
mainprogram()
begin
  makemat(prg(I,J),25,25);
end;

prg(x,y)
begin
  // code that loops x from 1 to 25 and y from 1 to 25 goes here
  return(0);
end;
I don't remember if the speed difference is still significant ever since the most recent firmware (never retested).

Can't wait! (It'll be interesting to see how you'll handle the bullet sprites.)

I have come up with several techniques; the real question will be which one is the fastest. I will probably put a soft limit on the total number of movable objects on the screen and see how the machine handles everything. So far, I'm pretty impressed with what this machine can do. The current main loop even requires a delay in between frames in order to slow it down since no delays means the scrolling goes by too fast (even if it's single pixel scrolling). We'll see what happens when more stuff is added.

Zera,

Thank you for sharing! I'll hold off on peeking too much into your code for now. So far I have only looked at the sprites. I would like to see if I can get this project off the ground from scratch. Your mockup certainly looks cool, though!

I have attached it to this reply.

Woah, with a little bit of search I managed to find the original assembly source code for columns! Not sure if it is worth my time to learn a new set of assembly instructions just to learn a scoring system lol... but anything is probably better than what I already implemented.

My suggestions were merely so that users do not end up resetting stored data. Cheaters will always find ways to cheat in games. I'm all for learning by viewing other people's code. Likewise with the archive idea. It is way to easy to delete the contents of a program at the moment, not to mention the entire list of programs in the catalog. At least with the archive feature it prevents accidents like this. The archive feature does not prevent users from viewing code -- they just have to take a few extra steps to do it so that they don't end up with a null-content program.

3 (or more) in a row tetris style game

That much I already knew. But how are points given when you make more than 3? For example, you could theoretically get 5 in a column or row or diagonal, or even a combination of all three for up to 19 blocks in a single combo if you set up your columns just right. So for each additional block, is there some multiplier?

Wow Han, that is impressive! Do you think complex graphing would be possible, too? Since this would map a complex number to a complex number (C→C) it requires 4 dimensions, so typically 3D+color is used. That would be pretty cool for complex analysis:)

I had given it some thought once someone reported some inconsistencies with complex mode on. If there is enough demand, I may add it in a future update.