"How does the Shadow of The Colossus™ engine perform its seamless long distance rendering?" SS from http://www.businessinsider.com/shadow-of-the-colossus-remake-playstation-4-video-2017-9

  • $\begingroup$ Perhaps if you posted an image from a different game where you feel there's a discontinuity it might make things clearer. Also, it's a bit difficult to tell from a single frame. $\endgroup$
    – Simon F
    Feb 14, 2018 at 9:08

1 Answer 1


Taken from The Making Of "Shadow Of The Colossus":

The rough algorithm is outlined below:

  1. Rendering the distant view, and save it.
  2. Foreground rendering.
  3. Composite the distant view and foreground together. It uses the contents of the Z-buffer like a mask pattern (see image 2).
  4. Again using the Z-buffer, generate a separate image of the mask pattern.
  5. Reduce this to 64x64 pixels, using a bilinear filter. (This makes it ultra-low resolution.) This gets reused in step (8).
  6. Blend in the previous frame (from step 7) using a percentage specified in the scene box.
  7. Store the image for use in step (6). It is used for drawing the next frame.
  8. Expand step (5) to a suitable resolution, using the bilinear filter. Thus, it becomes the material for the bloom effect where the distant view starts bleeding out from the shape of the foreground.
  9. Blend (7) and (8) using a ratio specified in the scene box. enter image description here

Landscape rendering is divided into 3 stages, the furthest background being either a rendered image or a texture, which is stuck on a distant polygon. The internal development team nicknamed this "Super Low".

Consequently, most of the medium range which goes from nearby to just past the furthest background, is rendered using a low-resolution landscape. This is managed in units of 600x600m, and as the player approaches it, once he gets nearer than a certain fixed distance, it changes to the high-res landscape model for the foreground. This switch is designed to blend well so it not obvious.

In addition before being combined down into one picture for the Super Low, most of the distant scenery being managed in 600x600m chunks has been reduced to around 1/30th - 1/100th of the polygons. Because there is little memory, we usually try and spread it around the landmark which represents that area. The combined image for the most distance scenery is done using this low-res model. With this, the low-resolution model appears seamlessly out of the Super Low image. By connecting the foreground and distant background well, we are able to produce a landscape rendering system which seems very natural. enter image description here


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.