I developed a small testcase for parallax etc… On my computer, it ran at 800-1500 FPS, but on my brother’s laptop, it ran at 65. However, he did have Opera running.
I would appreciate it if anyone could test it on their own computers. To properly test it, you must go near the wall you see in front of you when you run, turn to the right until the Object count in the Stats view is 114. Then, tell me the FPS after a few seconds. Thanks in advance to anyone who tests.
1 Like
INFO [com.jme3.renderer.lwjgl.LwjglRenderer]: Caps: [FrameBuffer, FrameBufferMRT, FrameBufferMultisample, TextureMultisample, OpenGL20, OpenGL21, OpenGL30, OpenGL31, OpenGL32, ARBprogram, GLSL100, GLSL110, GLSL120, GLSL130, GLSL140, GLSL150, VertexTextureFetch, TextureArray, TextureBuffer, FloatTexture, FloatColorBuffer, FloatDepthBuffer, PackedFloatTexture, SharedExponentTexture, PackedFloatColorBuffer, TextureCompressionLATC, NonPowerOfTwoTextures, MeshInstancing, VertexBufferArray]
INFO [com.jme3.system.lwjgl.LwjglContext]: Adapter: aticfx64
INFO [com.jme3.system.lwjgl.LwjglContext]: Driver Version: null
INFO [com.jme3.system.lwjgl.LwjglContext]: Vendor: ATI Technologies Inc.
INFO [com.jme3.system.lwjgl.LwjglContext]: OpenGL Version: 4.2.11399 Compatibility Profile Context
INFO [com.jme3.system.lwjgl.LwjglContext]: Renderer: ATI Radeon HD 5700 Series
INFO [com.jme3.system.lwjgl.LwjglContext]: GLSL Ver: 4.20
hope this will help ++
2 Likes
That helps a lot. 947 FPS is quite good. Thanks a lot. Sorry for rating you down, my mistake.
@normen, can you please change it to positive?
EDIT: Thanks normen 
Hello, the scene runs at ~2600fps on my GTX580. Standart settings, windowed 640x480
and at 410fps fullscreen 1920x1080 32bpp and 16x ani
1 Like
Thanks. Never thought fullscreen and resolution made that much of a difference.
no mac version?
This was just a quick upload :3 Sorry about that
± 2250 FPS on GTX 480 @ 1280*1024
± 1352 @ 1920x1080 @ 32bpp @ 59hz Fullscreen
2 Likes
Wow. Such a difference because of resolution. Thanks madjack!
GeForce GTX 560 TI
640x480 24 bpp
Watching between the Walls 3000 fp/s, watching at one of the walls 5000 fp/s.
Fulscreen 1920x1080 32 bpp:
Round about 1150 for both.
The lacking difference in the last point is what astonishes me more than the resolution dependency.
1 Like
@memonick said:
Wow. Such a difference because of resolution. Thanks madjack!
Yes, when lots of work is done per pixel (like in normal and parallax maps) then when you have more pixels it will take more time.
Are you using steep parallax or regular bump mapping?
Steep parallax. Else, it would still seem flat from up close.
@memonick said:
Steep parallax. Else, it would still seem flat from up close.
Not really. Regular bump mapping can have artifacts for dramatic bumps but it will still seem bumpy up close. Steep parallax is WAY more expensive performance-wise, though.
So I just need to un-tick Steep Parallax from the Material Editor and leave the Bump Map in the ParallaxMap?
@memonick said:
So I just need to un-tick Steep Parallax from the Material Editor and leave the Bump Map in the ParallaxMap?
I think so. I don't use JME materials as I forked them a long time ago.
It does occur to me that it would have taken much less time to try it than it did to post the question. ;)
I did try it 
I’m just obsessed with double-checking.
It certainly takes a lot of time tweaking settings and modifying the texture to get a good result, but I think it’s worth-it. How does this affect FPS? Is it because of light? Also, on different resolutions, should it get similar FPS (or more consistency than before)?
Regular parallax does less work per pixel than steep parallax. There is still work done per pixel. The performance differences will be more pronounced the more pixels there are… but regular parallax (and normal mapping, too) will still affect performance as pixels go up.
You can turn them off by setting vertex lighting to true. This turns off per-pixel lighting all together. No more normal maps and no more bump maps. All of the work is done at the vertexes. This will give you an idea of the performance trade-off.
So there will still be a slight difference in FPS when different resolutions are used, but the difference is smaller than if I use Steep Parallax. Is that right?
140FPS 640x480
200FPS 1280x720 (Fullscreen)
1 Like
How did that increase?