Unfortunately these don't make the same decision (integer scale factor
tries to make sure the whole image can fit in the window, while auto
render resolution is glad to step up a level), but at least the user is
not faces with a black screen.
Not sure what's actually the best thing here, possibly this case should
be disallowed somehow.
I failed to notice that when doing the shift to apply the "texel offset"
translating CLUT8 to a CLUT16 lookup, we also need to shift the mask
used to choose color components to read.
The framebuffer does contain the top of the video frame at the present
time though, so things have partially gone well. However the rest of the
frame lies after it in VRAM. What we probably technically should do when
we see this is to assemble an image from both the framebuffer we find
and VRAM contents exceeding the range of that image, but seems kinda
impractical.
This works, but there seems to be some other issue with how the text is
rendered or copied. Could be a color precision issue confusing the game
I suppose.
Fixes#17268, but I feel we could possibly get rid of the bool entirely
since we have maxlevel. Only GL and D3D even looks at this flag.
Oh well, that's for later.
Potentially, a watch or break condition could crash if it was unaligned
between mirrors. This might happen if it's not the condition you wanted,
especially. Play it safe.
Don't manipulate the UVs, manipulate the vertex coordinates instead.
This will simplify the next change that will fix the pixel accuracy
problems in post-processing.
This is mainly useful if you want an authentic pixellated look with 1x
rendering (or software) and nearest display filter. It'll simply round
down the auto-scaled sized to the nearest integer scale factor,
configuring exactly which one isn't that interesting since they all are
gonna look good.
Fixes#17093
Fixes#17182
Not exactly sure what behavior we really want, but I think this one is
OK, and at least more similar to the old one. Now we save
already-replaced textures if the named replacement texture is missing, and there
isn't already a hash-named one in new or the "root".
Henrik Rydgård