materials - time calculation to be render

[fario]

I dream ... of a card, listing all material type KeyShot and calculation time to be render​​.

It seems to me that there are matters which require much more  time than others to be render.

Of course I could do it myself, but I'm sure Luxion has these benchs.

?
Thank you

[fario]

It is true that my google  English is very bad .., but I think it's odd that I'm the only one who wanted this information.

Antoine

[fario]

Hello,

I'll try to better express my thoughts:

What do you think of this?

Informative?

These data are not really checked, they should be checked and verified.

I added the Illustrator file for those who want to participate in this evaluation.

Thank you very much.

Antoine

[NDenekamp]

Antoine,

Although I appreciate the idea behind this, I wonder about the practicality. All the ratings in this table only really give a comparison relative to other materials but there is no way to calculate real render time as there are far more variables. And even within the same materials it can vary greatly. Plastic can be opaque shiny (quick) opaque diffuse (little slower) transparent (little slower again) diffuse transparent (you get the idea) add in abbe and heavy refractions and its already a very different material to the opaque shiny plastic you started with.

I think most Keyshotters have a kind of table like this going on in the back of their heads, and you have a fairly good understand of what you could do to speed up your scene / what you shouldn't do to unnecessarily slow it down etc.

One use I could see though, is to set up a standard scene of say a material ball of some sort, and apply a whole range of materials at various settings and save renders at maybe 5 / 10 second intervals. Combine this into a matrix and determine where it reaches acceptable quality levels for each material. Although this will still be completely relative as each machine will be different, it may reveal some material tweaks or material alternatives that may not be obvious. (For example a transparent dielectric material may render fast than a refractive glass at certain settings)

But it may just be too many variables at play, and in truth Keyshot is plenty fast enough for you to quickly run a few tests to get an indication of how long a high quality render will take, so for most it may not be a very high priority concern.

N

[NDenekamp]

Hi Guys,

I had a little go at this, to see if it may be possible to establish how materials perform relative to each other.

For this purpose, I set up a scene with a not too simple / not too complex piece of geometry. And enabled the typical settings I would use when rendering. e.i:

Global Illumination - on
Ground Plane - on with slight reflection
Ray Bounces - 8
Background Colour - on

I then enabled the Head Up display to monitor the frames per second with various materials applied. I used the standard material types with a 50% grey for opaque materials and 100% white for transparent materials. No roughness applied and Refractive Index of 1.5 where possible.

I took a screenshot about 30 second into each material render - when the frame rate has more or less stabilised - to record the materials settings and FPS. (My machine reaches about 49.5 FPS in the Camera benchmark scene)

This info I've put into a table, and because each computer performed different, I normalised the material performance around a baseline of the 'Diffuse' material type (Material FPS / 74.6) to get a relative performance rating for each material type.

If this interests anyone else, I intend to run further tests, with various degrees of roughness, transmission, and more complex settings.

[fario]

wow!

thank you very much!

this is really informative!

Antoine