Pokechu22

Results 169 comments of Pokechu22

I'm a little confused. Is this patch actually needed for anything nowadays? It doesn't seem to be enabled by default, and I *thought* that whatever Metafortress was detecting was fixed...

This doesn't seem to work correctly with "write to log"; if I create a conditional breakpoint in Super Mario Sunshine at 802f3630 (`J3DGDSetChanCtrl`) with the condition `r3 == 5`, the...

I haven't looked over this in detail, but it feels like a somewhat strange approach. Dolphin already has a system to reformat vertices from the GameCube's format to one that...

>> It might be a bit simpler if there were, say, individual arrays for each component, some of which don't get populated, but I'm not sure if that would work...

I've determined that normalizing distAttn is indeed incorrect, and that Dolphin is currently handling lighting in Mario Tennis incorrectly: Real hardware: > ![image](https://user-images.githubusercontent.com/8334194/155863223-0100df3a-02d8-4f91-a2a2-d7553d8c28f8.png) > ![image](https://user-images.githubusercontent.com/8334194/155863225-e2f26a56-fabf-4343-a94e-fe401c5c3268.png) Dolphin (without this PR): >...

Fifoci diff prior to #10477: https://fifo.ci/version/4ca59f15e4093b85530b1526661accbfbf0b2425/

The diff for `tla-menu` comes from "SW/Rasterizer: Use RAS1_IREF::getTexCoord and getTexMap" - I'm not 100% sure what's causing it (other than that the image is blue because the software renderer...

The FIFO change is caused by "SW/Rasterizer: Use RAS1_IREF::getTexCoord and getTexMap". I've attempted to investigate it further and reproduce the issue on hardware backends, but I haven't been able to....

I did a quick and hacky implementation of the copy filter for `EFBCopyFormat::B8`, which allows seeing what is actually changing with this. Hacky implementation ```diff diff --git a/Source/Core/VideoBackends/Software/TextureEncoder.cpp b/Source/Core/VideoBackends/Software/TextureEncoder.cpp index...

I added one more change to the last commit fixing various XF mistakes: `XFMEM_POSTMATRICES` was calculating the row by subtracting `XFMEM_POSMATRICES` (POS vs POST), resulting in incorrect row numbering.