I am currently discovering the world of isometrics, now I found out there are two approaches to creating the tilemap;

  1. Just create 2:1 ratio tile-images and draw those.

  2. Creating squares and transforming them to the 2:1 ratio. What is the general approach on developing an isometric game?

Now I was wondering a few things;

  • How do more known games like AOE1/2 do this?
  • What are the pros/cons of both methods?
  • Which method is preferred to be used in this day and age?

Edit added more general question

  • \$\begingroup\$ Why does it matter how AOE1/2 does this? \$\endgroup\$ Nov 10 '12 at 23:39
  • \$\begingroup\$ Because I think AOE is a perfect example of how a game using isometrics should look/work/feel. \$\endgroup\$
    – Thizzer
    Nov 12 '12 at 8:52
  • 2
    \$\begingroup\$ AOE supported a maximum screen resolution of 1024x768. And didn't support 16:9 resolutions at all. My feeling is that good technical choices 15 years ago do not necessarily translate to good technical choices today, so there's only a limited amount of practical wisdom to be gleaned by examining the techniques used in implementing the games from that era. (unlike their designs, which are often still quite relevant) \$\endgroup\$ Nov 12 '12 at 10:50
  • \$\begingroup\$ True, I added another question to the list.. \$\endgroup\$
    – Thizzer
    Nov 12 '12 at 11:04
  • \$\begingroup\$ Did you even read my answer? This site requires you to give feedback - please do so by voting/accepting or commenting, or both. Thanks. \$\endgroup\$
    – opatut
    Nov 12 '12 at 20:39

I never did this, so I might be wrong with all of this. But I have a pretty good understanding of graphics programming, so here are my thoughts:

  • When AOE 1/2 were written, memory and disk space were valuable. A 2:1 tile-image is bigger than a square which is transformed.
  • If you draw squares and transform them, you do not have pixel-perfect control over you final image.
  • ... but it is easier to create tilable 2D sprites as squares.
  • transforming a square to an isometric tile is not necessarily trivial

I for myself would do the first approach (2:1 tiles), because it is easier to implement...

  • \$\begingroup\$ Of course I read your answer (forgot to +1 it), I understand 2:1 tiles are easier to implement and because of that it would be your method of choice but I was wondering what the general approach to isometrics is and why. There are a lot of isometric games out there but I find it hard to find information about how and especially WHY that way. (and I understand how stackexchange sites work, been around on sites like stackoverflow for a while now) \$\endgroup\$
    – Thizzer
    Nov 13 '12 at 9:08
  • \$\begingroup\$ Well then, I guess, there is no real answer to your question. I think there is no such thing as "the general approach", people have always found different solutions with different pros and cons. As I said, I would go for 2:1, simply because it is easier. If your framework can transform you graphics tiles, you can just as well use that (e.g. use OpenGL with an isometric camera transform matrix). \$\endgroup\$
    – opatut
    Nov 14 '12 at 6:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.