If i use no supersampling on the pimax 8k vs the supersampling on the oculus rift with the traytool,
will the scaler in the pimax 8k do the same thing ?
So maybe get an performance increase because the scaling happening in the pimax 8k?
Or am i wrong?
They do the opposite thing.
Supersampling means you render at a higher resolution than your screen can display, and then downsamples the large rendered bitmap to the size of the screen, “baking in” the extra rendered detail, so to speak.
The 8k’s scaler takes a small bitmap and upsamples it to the larger screen. No detail is added - on the contrary, there is usually a bit of softening of the image - the exact effect depends on the resampling algorithm.
If you are supersampling with the Rift, you are probably going to want to do it with the 8k as well, because they are probably going to prove to have roughly the same resolution (in the detail density sense, not the bitmap size one; In VR context, we often speak of Pixels Per Degree.).
When you use SS on Oculus, you basically instruct your gfx card to render the scene in higher res than what the headset accepts. The overhead is necessary to mitigate the pre-lens transformation which would otherwise lead to the loss of information (and pixelation) of the final output image.
The scaler in Pimax only takes the input video stream (which is already rendered and pre-lens transformed) and upscales it to fit the panel res.
So you would still need some SS even for Pimax. The question is how much, which so far no one from Pimax has been able/willing to answer. There are two theories:
- You would need lesser SS because Pimax screen resolution is much higher than original Oculus/Vive and therefore would not need so agressive SS.
- You would need more because Pimax screens are supposed to show much higher FOV than original Oculus/Vive and to address optical errors at the edge of the FOV you may need even higher SS.
The implications are quite important because they will determine the gfx power needed to render the scene.