I’m still working with beta and I’m struggling to get pixel perfect text with DX11, while I got it working perfectly with DX9.. Would anyone have experience with this issue ?
I remember this being a thing, but I just can’t really remember anymore … I also think back then we rendered a lot of text via CEF because of inline-styling.
Just an idea, but don’t get your hopes up: maybe it’s worth calculating the “texelsize” of your output: 1 pixel in normalized space, so for a 2x2 normalized space (or aspect ratio corrected, apply the aspect to 2x2 first) and divide by the output resolution of your renderer. Just in case the text maybe has to be offset slightly so it sits correctly on whole pixels.
Edit: and offset by half a texel in x and/or y (if that’s the problem, most likely both). Does it make any difference in how the text is being rendered?
Thanks a lot ! I tried looking into supersampling and tried every available DX11 text node with precise pixel positioning / res dependent, but it seems to be an issue with non-antialiasing option not available.
This below may be the way forward, but I’m not familiar with creating custom nodes, would this make sense for people who are familiar with csharp ?
It’s kinda makes sense, but I would not be so sure it will work straight away. Is it pointing to the some cs file?