On Wed, Apr 2, 2008 at 3:49 PM, Kamilche <[EMAIL PROTECTED]> wrote:
> Brian Fisher wrote:
> The resulting tile can be any shape, I don't care. I may have coded it up
> wrong, but it appeared Pete's code resulted in something that didn't tile.
>
Pete's code wasn't complete, but after fixes for rota
Brian Fisher wrote:
So why do you need a rectangular tile to tile? why not tile a rotated
image on a rotated grid? (i.e. like with Pete's code)
The resulting tile can be any shape, I don't care. I may have coded it
up wrong, but it appeared Pete's code resulted in something that didn't
tile.
So why do you need a rectangular tile to tile? why not tile a rotated
image on a rotated grid? (i.e. like with Pete's code)
"Aaron Maupin" <[EMAIL PROTECTED]> wrote:
> Other angles will repeat, too, but like Brian says, they'll be
> disgustingly large. As in many, many times the size of the original tile.
Some of the new tile sizes will be large, but they don't need to be "many,
many times" the size of the original t
Brian Fisher wrote:
but if you want to make a rectangular tile that tiles on a rectangular
grid from your rotated image, then that is more complicated (unless
the rotation is 45 and the original image square - which is what Aaron
was talking about), but I would guess it can be done, but may resu
Well, that was interesting, but also didn't work. It's similar to my
attempt of tiling then rotating then cutting a piece out of the middle.
I've modified the code to show what happens when that tile is tiled.
Interesting, tho!
--Kamilche
import pygame
import math
filename=r"G:\Incarnation\
there were 3 problems with the thing above
---
the major problem was that sin & cos work in a space where +y is up,
pygame in a space where +y is down
a minor problem was that the four corners will never be visible at all
(think about the surface rotate produces - the bounds go through the
four cor
Pete Shinners wrote:
I would expect you could do this. You'll need to rotate the surface, and
then blit it back onto itself 8 times with offsets. The offsets will be
the rotated corners of the original image rect.
Yeah, it didn't work. It seems like it should, but run the following
code (y
Kamilche wrote:
Hi all. I'm trying to create a rotated tilable bitmap out of a tilable
bitmap, but it's not working. My first attempt was to tile the bitmap
across a larger picture, rotate the larger picture, and then cut a tile
out of the middle, but that didn't work - the resulting picture wa
Well, basically, I want to take a picture, rotate it 15 degrees, and use
some algorithm to jam all the pixels lying outside the bounding box back
inside the box, and have the result be tilable.
I thought it should be possible, but it's a thornier problem than I
realized.
--Kamilche
On Tue, Apr 1, 2008 at 8:32 PM, Aaron Maupin <[EMAIL PROTECTED]> wrote:
> That will only work if you rotate the bitmap 45 degrees. (Or some
> multiple of 45 degrees ;)
>
You may be right about particular rotation implementations, but in
general taking a tile and rotating it and expecting it tile
Kamilche wrote:
My first attempt was to tile the bitmap
across a larger picture, rotate the larger picture, and then cut a tile
out of the middle, but that didn't work - the resulting picture wasn't
tilable.
That will only work if you rotate the bitmap 45 degrees. (Or some
multiple of 45 de
when you say the resulting picture is not tileable - what exactly do
you mean? (screenshot?)
do you mean that you see seams (dark or partially transparent lines
through the joints between tiles) when you try to draw the tiles next
to each other? If you do, it's probably an alpha problem (using
rot
Hi all. I'm trying to create a rotated tilable bitmap out of a tilable
bitmap, but it's not working. My first attempt was to tile the bitmap
across a larger picture, rotate the larger picture, and then cut a tile
out of the middle, but that didn't work - the resulting picture wasn't
tilable.
14 matches
Mail list logo
