Originally posted by arpaagent I can pleasantly report now that I can definitely see the jaggies, and the image *snaps* into this processed state
Hmmmh,
I checked with a roof texture. The image on the rear LCD updates after about 0.5s into a higher resolution version. Both versions (before and after the "update") show about the same amount of Moir pattern on the roof texture - only that the Moir pattern changes with the "update".
So, the downsampling routine is such that it doesn't prevent Moir. Bilinear resampling would prevent it. However, I cannot say what algorithm they use to downsample. Nearest neighbor would preserve most of the Moir pattern.
I also checked if one can eventually see individual pixels on the rear LCD. Doesn't seem so. In upscaling, the pixels seem to be still interpolated.