Don't draw a game from memory

Author

Thomas Randall
June 30, 2025

Words I like: Don’t solve things you think might be a problem. Solve real problems you observe from players.

Randy’s Dev Dispatch: Don’t draw a game from memory

I've made many mistakes over the years when trying to make a game.

Like a metric shit load...

But this is probably in my top 5. It completely stopped me from making a game that was even playable.

To show you what it is, let me tell you a little story.

The other day, I was trying to draw a grass sprite for my game. So I opened up my editor and started drawing.

"I think this is kinda what grass should look like?"

It turned out looking... pretty bad.

I spent the next couple of hours trying to redraw it over and over again. But no matter what I tried, I couldn't get it looking good.

I then realized that I was trying to draw the grass sprite from memory! I didn't have any reference image to pull from.

So I went in search of a reference that I thought looked good. I found some grass pixel art from a similar game and studied it to draw a new version.

I shit you not, within 3 minutes this new version was done… and looked 10x better!

Finding a high-quality reference saved me hours of more trial and error. I would have kept struggling against it for ages if I had only relied on my memory.

How does this relate to game dev, though?

Well, the same rules apply.

For years, I struggled endlessly with not being able to make a concrete game experience. I’d just work on whatever idea game to mind. Things that I thought would be good.

Nowadays, I first just find a reference. Study the game design. Break it down. And then try and recreate it.

It works way faster, and way better.

Don't draw the game design from memory.

Find a reference.

Draw from that.

If you’d like to have a 1 hour call with me to chat about the biggest problems you’re facing, I’ve just opened up my calendar for a bit.

KEEP CRUSHIN’ OUT THAT GAME DESIGN LADS!

-randy