I’m using highp so IEEE 754 compliance is mandatory in the fragment shader. So blend and quantization are the only possible issues. Pathfinder requires that both be precise, so I suspect I would have seen breakage if there were identifiable differences.
-
-
-
D3D requires that the rounding go a certain way: https://docs.microsoft.com/en-us/windows/desktop/direct3d10/d3d10-graphics-programming-guide-resources-data-conversion … So I expect that implementations behave the same way.
End of conversation
New conversation -
-
-
Only for canvas. Besides, OS's are converging on font rendering these days, and I'm not sure how much users care. macOS and Edge removed subpixel AA and nobody really noticed!
-
(NB: The canvas text APIs are terrible to begin with, and serious apps that care about text shouldn't be using them in the first place.)
- 1 more reply
New conversation -
-
-
I haven’t fully tested, but I suspect I would have seen breakage already if the rounding were inconsistent across GPUs, due to the way the antialiasing works.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Heh, except we're no longer doing GPU rasterization on <canvas> :)
-
The software renderers delegate to the OS font rasterizers, which render vectors differently. With Pathfinder I think we should consider doing GPU rasterization of canvas again.
End of conversation
New conversation -
-
-
If you'll ever finish it that is. You know what I mean? :-)
-
I’m working as hard as I can. That’s not appreciated.
- 1 more reply
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.