Seamless tile creator

Seeing as there's been a lot of talk about this sort of thing I thought I'd have a go at one myself.

It's doesn't do anything clever but it seems to work quite well.

If anyone feels they want to improve upon this then please do, there's a lack on contrast on the blendy edges so I might have a go at improving that at some point...

Anyway... It's in the code archives, have a play, rip the code apart, improve it etc.

http://www.blitzbasic.com/codearcs/codearcs.php?code=1325

Hmmmm... Fart in a spaceship popularity obviously... Oh well...

I used to have a fart in a spaceship, but accidentally opened the docking bay... Typically I hadn't done a backup.

I tried downloading Drekinn's Fart in a Lift, but the featureset was a bit under-developed (only one degree of freedom.) If you decide to develop your Fart in a Spaceship idea feel free to put me down on the beta-tester list.

Every time I see some of your code I smile at the 'code to monkey with' line - there must be a game concept in there !!

Now why would anybody want this for free when they can get sswift's one for..... erm.....

I'll take TWO! :D

What I dont get is that someone decides to make a program and sell it. Then because that person sells it other people decide to code one and release it for free. Now with all the posts on why sswift is charging over 40 bucks for the software and now there appears to be free code that does it for free, it leaves me thinking and saying to myself "What was the point in buying it?"

The same thing happened when I bought A.L.E. Two weeks after buying it there were no less than three free landscape editors. All of them doing the same job as the one I bought.

I think its time for me to have a rethink on whether I will buy anything produced by a blitzer again.

This isn't as good or as clever as swift's tiler jobby. Swifts one doesn't need overhanging texture I believe and uses very clever maths and way beyond anything I would even bother with. I'm far to lazy for that kind of stuff.

There are plenty of flaws in it so please improve it.

To answer your question Amon of why do people write code for free when other people are selling it.
It's because the concept seems like it might be a fun challenge to write something that does what they're selling. That's the only reason I did this, and probably the only reason why other people have done the terrain things for example. No malice, just curious to see if they can do it too.

"The same thing happened when I bought A.L.E. Two weeks after buying it there were no less than three free landscape editors. All of them doing the same job as the one I bought."


Amon, Rob's tool does NOT do the same thing as the one you just bought. That's crazy talk! :-)

His tool doesn't correct lighting across the image, or allow you to remove perspective from the image. When blending, it only does a simple alpha blend with noise. The maximum size of the source image is limited by your screen resolution. There is no load dialog, no save dialog. No ability to modify the size of the output texture. No abiltiy to change the radius of the area blended.

No offense intended to Rob here. There's a reason it took me six weeks to write my program!

(I spent two weeks on the algorithms alone, and the interface has taken me a firther four weeks so far.)

The idea that you wasted your money on my tool because Rob's tool now exists is silly! It's not like Rob here has developed the first free seamless texture tool. My tool exists, not because no other tools like it exist, but because it does a better job than the other tools. Including Rob's here.

That's why you should be happy you bought it... Because your textures will tend to come out looking great, even if you're not a good at making art.

Yes you can. Look at the size and Qsize variables.

Anyway, I'm not trying to compete with Sswift here, as I said, it just looked like an interesting thing to have a go at. As Swift says the thing I put in the code archives is nowhere near polished or clever, however, there's no reason why it has to stay that way if people want to add to it.

"Yes you can. Look at the size and Qsize variables."

I meant without recompiling it. :-)


Oh, and I think your method is clever. I liked the idea of using noise to influence the alpha map to make it look better, even if that wouldn't work well in all instances, and I like the idea of blending areas outside the texture into the other side of the texture. It seems more intuitive than the "stretch" methods of blending which do something similar.