General Questions: Math for Radiance; Estimated time to proficiency

Hello Radiance Community,

I'm from the ladybug/honeybee community. I'm not very familiar with the
technicalities behind ray-tracing software so I was just wondering about
what Mathematics/Physics I'd have to study in order to have a firm
understanding of how Radiance works. The only thing that comes to mind is
Optics and maybe Linear Algebra but I'm sure there's more to that.

For example, I studied Thermodynamics in order to understand EnergyPlus,
and Multivariable Calculus w/ Fluid dynamics to understand CFD software.

Also, I was just wondering about the learning curve of Radiance. Although
I'm not an expert, I'm quite confident in my Python skills. How long might
it take me to get a good grasp of Radiance (Intermediate - Proficient) if
study it for 2-3 hours a day?

Thank you very much!

Glassner’s *Principles of Digital Image Synthesis*. (2 vols. San
Francisco, CA: Morgann Kaufmann Publishers, Inc., 1995) is the
reference I used, back when; anyone know if there has been a
subsequent book which covers the same ground? The physics, I would
say, is basically simple, *but* presents computational problems which
remain difficult, even with modern computing hardware.

For Radiance specifically, the mathematical specifics of the basic
system are covered in *Rendering with Radiance* part III. Newer
components are covered in various Radiance Workshop presentations.

Randolph M. Fritz

Hello,

I started just like you a few years ago... I also have a Thermodynamics
background that ad to be complemented with Radiance (lighting simulation)
knowledge.

In my experience (from learning and teaching), the first really frustrating
challenge people face is understanding that Radiance is not "a program" as
they are used to, but a "suite of program" that have to be used in "that
ugly, old-school and unpleasant terminal". Getting to understand the Unix
way might be hard.

However, if you already know Python, you are, probably, already with those
concepts. After understanding that, I found the tutorials available in the
Radiance-online website very useful for general matters. Also, for
understanding better the different programs and learning how to organize
your files, I found the tutorials on the Rendering With Radiance book very
good.

A third step, I would say, is to understand how Ray-tracing works. There
are a few good lectures on rendering and computer graphics on YouTube that
you can watch. They will guide you on some core concepts, I think. After
that, it may be useful to read a bit of how Radiance work itself. This can
be done through the Rendering With Radiance book and/or journal and
conference articles by Greg Ward.

Finally, to understand climate based daylight simulation methods (all first
3 steps were focused on how radiance did things), I would recommend to read
the advanced tutorials on the website (3-phase method, 5-phase method,
etc.) and maybe some journal articles on those topics. John Mardaljevic PhD
thesis is also a good asset, as well as some papers by Christoph Reinhart,
both developed methodologies for performing annual simulations using
Radiance. In this regard, I made a literature review on lighting and
thermal simulation methods for my M.Sc. thesis a while ago... after a while
I noticed there were things missing (i.e. John's 4-component method being
the most remarkable one), but it might help you.

I hope that helps... I learned using these sources, as well as this list,
which is full of very nice people who may help you a lot

Germán

···

2017-08-28 0:10 GMT-03:00 Randolph M. Fritz <[email protected]>:

Glassner’s *Principles of Digital Image Synthesis*. (2 vols. San
Francisco, CA: Morgann Kaufmann Publishers, Inc., 1995) is the
reference I used, back when; anyone know if there has been a
subsequent book which covers the same ground? The physics, I would
say, is basically simple, *but* presents computational problems which
remain difficult, even with modern computing hardware.

For Radiance specifically, the mathematical specifics of the basic
system are covered in *Rendering with Radiance* part III. Newer
components are covered in various Radiance Workshop presentations.

Randolph M. Fritz

_______________________________________________
Radiance-general mailing list
[email protected]
https://www.radiance-online.org/mailman/listinfo/radiance-general

I second the recommendation for Glassner's comprehensive 2-volume reference, and I consider most of it still relevant, though it of course doesn't cover the newer sampling algorithms like Vertex Connection & Merging and Markov Chain Monte Carlo (Metropolis and derivatives), but these are absent in RADIANCE anyway.

I'd also like to point out Philip Dutré's excellent Global Illumination Compendium (https://people.cs.kuleuven.be/~philip.dutre/GI/). This is a highly condensed formulaic summary of a lot of what goes in RADIANCE (and other GI software). It does assume a strong background in mathematics, however. It's an absolute must-have for developers, and was always on my desk when I developed the initial photon mapping release. If you ever wondered how radiance handles hemispherical geometry in its sampling routines, for instance, this document will explain it in probably more detail than you care to know. ;^)

The "Advanced Global Illumination" book referred to in the above link is worth a look too; it's a more general overview of GI techniques and more recent than Glassner's reference, though I personally find it lacking in detail compared to the latter. As an overview, it does the job, tho.

"Physically Based Rendering" by Pharr, Jakob, and Humphreys (http://www.pbrt.org/) is the most up-to-date of these references, and describes advanced raytracing algorithms as implemented in PBRT, an open-source raytracer with functionality similar to RADIANCE, albeit for more general rendering using newer algorithms. This is a hands-on book that places less emphasis on the maths and more on the implementation.

Hope this helps; best regards,

--Roland

···

On Mon, 28 Aug 2017 05:10:19 +0200, Randolph M. Fritz <[email protected]> wrote:

Glassner’s *Principles of Digital Image Synthesis*. (2 vols. San
Francisco, CA: Morgann Kaufmann Publishers, Inc., 1995) is the
reference I used, back when; anyone know if there has been a
subsequent book which covers the same ground? The physics, I would
say, is basically simple, *but* presents computational problems which
remain difficult, even with modern computing hardware.

--
Lucerne University of Applied Sciences and Arts
School of Engineering and Architecture

Institute of Civil Engineering
CC Building Envelopes
Dr. Roland Schregle
Senior Research Associate

T direct: +41 41 349 36 26
[email protected]

Technikumstrasse 21, CH-6048 Horw
T +41 41 349 33 11, F +41 41 349 39 60
www.hslu.ch/ccease