- Apparently, MSVC has a more strict implementation of IEEE-754 single-precision floats (with 23-bit significands) than Clang and GCC, and so we going to drop 2 LSBs during the calculation so that we're multiplying z by a 22-bit significand. Coincidentally, this now matches what we're doing with the OpenGL renderer, so this tends to better code consistency.
- New Behavior: Due to the rarity of needing to emulate 'Depth-LEqual polygon facing' and its guaranteed reduction in performance in all games, this accuracy feature is now OFF by default.
- Expose these new settings in the Cocoa port UI.
- SoftRasterizer may now drop at most one LSB, down from dropping 9 LSBs.
- OpenGL will now drop only 2 LSBs, down from 9 LSBs. In this case, dropping 2 LSBs was specifically chosen to ensure that the Dragon Quest IV overworld map continues to work.
- If this change makes the depth inaccuracy too much worse than before, then we may have to make these particular depth calculations optional in the future. This will need additional testing.
- Now, the only two methods for changing any firmware setting is by modifying CommonSettings.fwConfig or by loading an external NDS firmware binary file.
- All methods for changing the firmware MAC address through the WifiHandler class have been removed.
- The FirmwareConfig struct can now handle the WFC User ID.
- Clients can now retrieve the current MAC address and WFC User ID using NDS_GetCurrentWFCUserID(). It is also possible to retrieve the WFC User ID from CommonSettings.fwConfig.
- Setting up the firmware in NDS_Reset() should now be more consistent. However, this does change some of the loading/unpacking order previously set by NDS_FakeBoot(). This will need additional testing.
- Do a whole bunch of code refactoring and cleanup.
- This change shouldn't actually change any functionality in practice... probably. This change is there to silence a compiler warning more than anything else... hopefully.
- This change obsoletes "CommonSettings.wifi.mode", which now does nothing. Ports that make use of this setting should remove it.
- Also do a bunch of code refactoring and cleanup.
- This fix properly emulates the less-than-or-equal depth test rendering for front-facing polygons drawn on top of opaque back-facing fragments, but only if the front-facing polygon is opaque. Translucent front-facing polygons are not supported at this time due to requiring extensive changes to the rendering logic and shaders in order to emulate this extremely rare and niche NDS feature. (If you require the proper rendering of translucent front-facing polygons on top of back-facing fragments, then you must use SoftRasterizer.)
- The new behavior for the Multisample Antialiasing checkbox: Checked - GFX3D_Renderer_MultisampleSize = 4, Unchecked - GFX3D_Renderer_MultisampleSize = 0. (If someone else wants to make some UI so that GFX3D_Renderer_MultisampleSize can be set to other sizes, then have at it.)
- Add a unique sequence number to fetched frames to ensure that older frames are not drawn after newer frames.
- After much research, finally settle on a method for fetching the NDS framebuffers -- using a MTLBlitCommandEncoder to blit a MTLBuffer to a MTLTexture. It is faster than uploading a texture using [id<MTLTexture> replaceRegion:mipmapLevel:withBytes:bytesPerRow:], and also faster than using a pinned-memory backed linear texture. This method will be the way going forward for fetching framebuffers in Metal.
rogerman