@RustyCrab @binkle I'm not sure why they chose this terminology, and I see the difference the same way, but my point is that without supersampling and sample shading fragment output corresponds to exactly one framebuffer pixel (before blending)