Talk:Path tracing

Wrong BRDF calculation in pseudocode?

Why is there a multiplication by 2 when calculating the BRDF in the pseudocode? According to the [Lambertian reflectance] article, shouldn't it be a simple reflectance * cos_theta?
— Preceding unsigned comment added by Noname030919 (talk • contribs) 7 July 2017

The pseudocode has been replaced. KaiaVintr (talk) 20:11, 27 May 2026 (UTC)

Ray tracing VS Path Tracing features

It's not at all clear from the article what the difference is (or if there is one) One line says

>"Path tracing naturally simulates many effects that have to be specifically added to other methods (conventional ray tracing or scanline rendering), such as soft shadows, depth of field, motion blur, caustics, ambient occlusion, and indirect lighting."<

The wiki page for ray tracing touts most of these features as a seller for ray tracing. Is it just harder to implement in ray tracing? There are no citations for this 167.32.2.2 (talk) 13:21, 2 October 2024 (UTC)

I think this is addressed now in the lead, History, and Description sections. KaiaVintr (talk) 19:57, 27 May 2026 (UTC)

v Ray tracing section

This section is needed. Essentially, simply put PT is RT (: "The equations for modeling the behavior of light are the same, the use of data structures to accelerate the searching for ray-triangle interactions are the same, too, and modern GPUs use the same units to accelerate the process." ), but PT is less computationally intensive than RT because it uses approximations. 37.205.110.124 (talk) 14:59, 3 February 2025 (UTC)

Just in case there is any uncertainty about whether PT is "a kind of RT"...

From Kajiya's 1986 paper (first citation for this article) page 148:

"This diagram also points out an alternative algorithm for conventional distributed ray tracing. Rather than shooting a branching tree, just shoot a path with the rays chosen probabilistically. For scenes with much reflection and refraction, this cuts down vastly on the number of ray object intersections to be computed for a given plxel and performs a remarkable speed up of ray tracing for very little programming work. However, for this new fast form of ray tracing--called path tracing--we have found that it is very important to maintain the correct proportion of reflection, refraction, and shadow ray types contributing to each pixel."

I think it's more obvious that path tracing is a form of Distributed ray tracing. For someone who is only familiar with Whitted-style recursive ray tracing, path tracing might seem like a very different algorithm, hence the confusion (and occasional arguments online, it seems).

Part of the problem is that "ray tracing" has been used as a marketing term by Nvidia and Microsoft (and others). Games are doing ad-hoc combinations of ray casting, recursive ray tracing, distributed ray tracing and (occasionally) proper path tracing, and it's all called RT for marketing. KaiaVintr (talk) 20:27, 3 February 2025 (UTC)

I think this is addressed now in the lead, History, and Description sections. KaiaVintr (talk) 19:57, 27 May 2026 (UTC)

"Components of a modern path tracer" section and subsections

There are a lot of components to talk about, each needing its own subheading (9 originally, and there is one more now). I broke them up into three categories using another level of headings: "Sampling", "Spatial data structures", and "Turning samples into images". This looks good in the table of contents, but when reading the article it might look weird (I couldn't find any manual of style articles addressing this issue).

The grouping sections like "Sampling" could have some brief introductory text, but I wanted to avoid editorializing and adding bloat/clutter (which could be distracting or confusing). I think it's safest to just go straight into the subsections, but if people object strongly to sections that have subsections but no text of their own, then something could be added. KaiaVintr (talk) 14:02, 29 June 2026 (UTC)