Slick Forums

in the zip file there's already a screenshot. i'm currently at work and i'm not allowed to upload files via ftp, so i'll describe where the lines are. they are to the left and the right of the background image and to the left of the one small image below the bar at the top (the image is actually the outer right part of an input field background image).

i will test the current code from svn this evening. thx in advance!

_________________
rastaduck.org

ok, i just tested with the newest slick code. the grey lines at the sides of the background image are still there. the small image below the bar on the top does not have a grey line anymore but it seems that the effect is now "inverse". it looks like a small bit of the image (the outermost left column - 1px wide) disappears after a few seconds.

when i loaded your eclipse project from the svn checkout i had to change the JRE system library to 1.5.0_10 because i had no 1.4.x version installed anymore. after that i built the project using the ant build.xml and copied the resulting slick.jar to the project where i run my test case. not sure if this info is helpful in any way ...

thanks for your effort in resolving this!

_________________
rastaduck.org

What graphics card/driver is this btw, looks like it's a flaw with the driver that we'll have to find a way to work round.

Is it only when you scale and rotate - or is just scale or just rotate?

Kev

Mobile Intel 945GM Express Chipset Family

it doesn't seem to be affected by scaling, only rotation in combination with a larger number of images(!) leads to the effect.

the thing is that i never had such problems when using lwjgl alone. only the combination with slick is causing the issues. isn't slick sitting on top of lwjgl and thus using existing functionality?

anyway, i will transfer the test case to my pc and see if it happens there too.

_________________
rastaduck.org

I think it's got to be an error with the translate(x,y), rotate, translate(-x,-y) - maybe the reversing isn't working but of course I still can't see the effect here. I'll try and my mac later, thats got some intel card in it.

Kev

i just ran the test case on my pc with an Nvidia GeForce 6800 and the result is even more strange. no grey lines this time but the colors of the background image are reduced and i see small errors in some of the images at the top.

here's a screenshot:
http://www.rastaduck.org/tmp/bug.jpg

_________________
rastaduck.org

Looks like a 16bit desktop setting?

Kev

nope. i checked my settings and on the other hand the game runs fine when i only use lwjgl without the slick functions.
you know, i have made this game framework for myself which allows me to run a game using different technologies. awt jframe, applet, android, lwjgl and lwjgl+slick. i never had problems with the other technologies. just the newest technology mix of slick and lwjgl is making problems. chances are good that i'm doing something wrong but i can't find out what.

_________________
rastaduck.org

any news on this?

_________________
rastaduck.org

Not had another chance to look yet. Running short on ideas. Do the artefacts show up on any of the games posted in the games showcase or in the demos? If not, then I'd guess it's a difference in initialisation of GL code.

Kev

the demos run fine on my laptop. i also believe that it's got something to do with how i set the lwjgl flags/parameters in combination with the slick library. if i only would know the openGL stuff better i could problably investigate more.
anyway it doesn't seem we're making progress here. i will try to find other ways to integrate slick into my game framework. thx a lot for your help so far!

_________________
rastaduck.org