On Windows, nvidia don't give us their driver version, so we can't workaround any issues.
As buffer_storage is broken on some drivers, we wanted to disble it for them.
So we can't.
Luckyly only "some" released driver versions are affected as this extension is only available since some months. Let's hope that nobody have to use one of this driver version, else they will get a black screen ...
ARM only at the moment. Could potentially support x86 and MIPS if necessary.
Capable of parsing the manufacturer codes and part IDs of some (but not all part numbers). If anyone knows of part numbers that aren't in the list, please report them.
OSX has their own driver, so performance issues aren't shared with the nvidia driver (unlike the closed source linux and windows nvidia driver). So now they'll also use the MapAndSync backend like all other osx drivers.
fixes issue 6596
I've also cleaned up the if/else block selecting the best backend a bit.
The new chapter title in Paper Mario TTYD had a small graphical bug due to the new code because it read one extra pixel, this fixes it.
I hope this gets everything, I though I had checked most bugs and yet here I am, commit-spamming...
Fixes some remaining bbox related bugs in Mickey's Magical Mirror and a slight graphical glitch in Paper Mario: TTYD when flipping and Vivian as your companion (I've been scratching my head for days to find this one).
Instead of being vertex-based, it is now primitive (point, line or dissected triangle) based, with proper clipping.
Also, screen position is now calculated based on viewport values, instead of "guesstimating".
This fixes many graphical glitches in Paper Mario: TTYD and Super Paper Mario.
Also, the new code allows Mickey's Magical Mirror and Disney's Hide & Sneak to work (mostly) bug-free. I changed their inis to use bbox.
These changes have a slight cost in performance when bbox is being used (rare), mostly due to the new clipping algorithm.
Please check for any regressions or crashes.
The only two devices that do this are Mesa software rasterizer and Intel Ironlake(With a few hacks).
Basically since it doesn't support OpenGL 3.0, it can't grab the version the new way.
So failing that, it sets to GL 2.1, and continues.
Further along, on Ironlake at least, it tries grabbing the extensions the new GL 3.0 way and fails.
So have a fallback that grabs the extensions string the old way, in probably the most elegant way possible.
This branch drops our temporary buffer in VertexLoaderBase.
Instead, every backend now must provide a buffer to convert vertices and indices.
D3D just uses a temporary buffer like before.
OGL maps the gpu based buffer and stream to them directly.
So this will avoid an unneeded memcpy on OGL backend.
The old way was to use big switch/case statements based on a type of buffer.
The new one is to use inheritance.
This change prohibits us to change the buffer type while running, but I doubt we'll ever do so.
Performance should also be a bit better. Also a nice cleanup.
Added some comments about this different kind of buffers.
This is a bit slower on map_and_* because of flushing and _very_ much slower on buffer(sub)?data because of a new memcpy.
But this design allow us to decode directly into a gpu buffer, eg vertexloader will profit :)
gl.h and glext.h provide most of the function pointer typedefs and defines for extensions and core features.
The only one it doesn't provide is GL 1.1 function typedefs, but this is to be expected.
If anything needs defines or typedefs in their header in the future, that's as easy as before.
Prior to this commit it was possible to assign the same keycode to more than one button.
ie. Say I assigned Open with the hotkey Ctrl+O; well, it was possible to also add it to another function as well, which leads to hotkey clashing.
Now, say I assign Open with Ctrl+O, but then assign that same hotkey to Refresh List; it will unbind the hotkey from Open and then assign it to refresh list.
- Spaces -> Tabs | Consistency
- Javadoc everything that was added and not documented.
- Remove duplicated code regarding the adapter that used to reside in DolphinInfoFragment.java. Now it resides in AboutActivity.java without a second duplication of it.
- Properly retrieve all of the contexts in the EGL initialization in EGLHelper.java.
- Remove the attribute EGL_RENDERABLE_TYPE from the pbuffersurface attributes in EGLHelper.java. With this present, the EGL context will always fail to reinitialize if destroyed and attempted to be recreated.
- Break the inner class Limit within GLES2InfoFragment.java, GLES3InfoFragment.java, and GLInfoFragment.java into its own single class. Greatly reduces code duplication.
- Introduce a Type enum into Limit.java (one of the wildly rare cases in Java where an enum is actually an OK solution). Removes duplicated constants from the Java files stated in the previous bullet note.
- Add a copyright comment to the top of EGLHelper.java. Forgot to do this initially, my bad.
- Add some missing override annotations to GLES2InfoFragment.java, GLES3InfoFragment.java, and GLInfoFragment.java.
- Use StringBuilders in the previously mentioned three Java files. This is better than using a String in this instance, as the String object won't have to be recreated multiple times (ala concatenation).
- Fix some constant accessors in the previously mentioned three Java files.
- Added the 'final' modifier to the above three classes and to Limit.java. These classes serve a single purpose only, and are not intended to be inherited.
There is a /lot/ of information in these tabs, we may have to think about changing how the information looks
OpenGL isn't done yet since there are a million limits on desktop GL, may just show a few things and extensions there.
This one was introduced to reduce the glBindTexture and glActiveTexture calls. But it was quite a bit of logic and only an improvment on uploading/creating a texture, which is done rarely.
Buildbot system user