Jump to content

The SSS Diaries


Recommended Posts

A question for anyone who's been testing out these SSS ops...

It occurs to me that this whole business with the automatic determination of the pointcloud density (in the SOP HDA), while a nice idea in theory, seems to be turning out to be not so useful in practice. Is anyone using it, or does everyone go "manual" with it?

14567[/snapback]

I've only used manual so far, just cos on the models I've worked with I just end up with too many points in my clouds and the whole thing chokes.

Link to comment
Share on other sites

Hi Simon,

Any chance of a hint about where to find this stuff? Is it mentioned in the paper?

14615[/snapback]

The paper includes a few materials, and the actual measurements for these can be found here.

And then there are a few independent efforts to publish BRDF measurements from different institutions. The main one is the CUReT database. There are others that use their own protocol for the measurements, like the one that NIST puts out, or the one by MIT. And there are other sources of BRDFs for large ecosystems, like the one here.

However; these are raw measurements that you'd need to convert to the homomorphic format first; if that's the representation you choose to go with (but as you can see, there are alternative representations out there ;) ).

I've only used manual so far, just cos on the models I've worked with I just end up with too many points in my clouds and the whole thing chokes.

14630[/snapback]

Yup; I'm finding the same thing. It was a nice idea, but... oh well..

I think I've made up my mind to remove the automatic thing now.

Thanks Simon! :)

Link to comment
Share on other sites

I must say that I've never set it to Manual. What I do is set the Samples per Disk to something low, like 3, and then increase that until I get the density I like. It makes more sense in my head to think of Samples Per Disk than total samples.

Maybe the SOP needs a "AutoCook" toggle that could make the generation of the points optonally happen at a button-stroke instead? This way you could give some feedback to the user about how many points are going to be scattered before you actually incur that cost?

Does anyone know where it slows down the most? Is it large areas, or large numbers of polys or large numbers of points that slows it down the most?

Link to comment
Share on other sites

I must say that I've never set it to Manual. What I do is set the Samples per Disk to something low, like 3, and then increase that until I get the density I like. It makes more sense in my head to think of Samples Per Disk than total samples.

Maybe the SOP needs a "AutoCook" toggle that could make the generation of the points optonally happen at a button-stroke instead? This way you could give some feedback to the user about how many points are going to be scattered before you actually incur that cost?

14668[/snapback]

Huh... OK.

Maybe an "estimate" tab then -- shows you what the estimated density would be for the given scattering distance, but doesn't actually create the points.... Alright. I'll try that. Thanks!

Does anyone know where it slows down the most? Is it large areas, or large numbers of polys or large numbers of points that slows it down the most?

14668[/snapback]

I'd say large number of points, but perhaps more importantly, large number of Points to Filter more than anything else.... :unsure:

Link to comment
Share on other sites

Hi all,

Another update. Wooohooo! :P

Things that have changed:

*) The single vop is now split into two separate vops -- one for multiple scattering, and another for single scattering.

*) Reworked the auto-generate portion of the SOP HDA, so that now the estimate is given as just that: an estimate. It is then up to you to use it or not by toggling the wonderfully-named "Use Density Estimate" toggle.

*) Added help cards for all the OPs.

*) Changed the style sheet in the packager so that it matches the orange look of the new help cards in v7.... very important, that! :lol:

*) Fixed a small bug in the single scattering code where the samples counter for the chromatic loop was being updated in the wrong place.

*) Initial commit to Subversion (migrating from good'ol CVS.... very scary :cry2:)

Uhmmmm...... I think that's it...

Take it for a spin, and let me know what you think. Thanks!

SSSfull4.zip

Cheers!

Link to comment
Share on other sites

  • 2 weeks later...
OK; I cleaned up a couple of small things, and uploaded it to the Exchange.

Cheers!

14873[/snapback]

Hey Mario,

This is great of course!

It seems this thread lends itself to a great OdWiki discourse too. It'd save us pinning the topic forever (which is just what I'm going to do right now, lest it ever drop off to Page 2 of this forum;))

Link to comment
Share on other sites

Arrrrgh.... how typical!

I was trying out some ray-jittering code for single scattering, and I accidentally uploaded that version to Exchange. Needless to say, that version was buggy. I've now updated it with the correct one.

Sorry everyone!

I always screw up these postings...<sigh>

14882[/snapback]

What a cheap trick to push your Number Of Downloads!

^_^

Link to comment
Share on other sites

Hi Simon,

The paper includes a few materials, and the actual measurements for these can be found here.

And then there are a few independent efforts to publish BRDF measurements from different institutions. The main one is the CUReT database. There are others that use their own protocol for the measurements, like the one that NIST puts out, or the one by MIT. And there are other sources of BRDFs for large ecosystems, like the one here.

However; these are raw measurements that you'd need to convert to the homomorphic format first; if that's the representation you choose to go with (but as you can see, there are alternative representations out there ;) ).

14667[/snapback]

Thanks Mario, I only just found this reply. Good stuff :D

Link to comment
Share on other sites

  • 1 month later...
  • 3 months later...
  • 1 month later...

Hi all,

I've been asked by a couple of people if I could post the slides from a talk I gave on this topic recently ("Toronto Technical Evening"). I'm not sure how useful they are since they are... well... slides... but maybe they help clarify some of the stuff talked about in this thread... plus it's got some more purty diagrams :)

So I'll post them here (as a pdf file) until I have a chance to flesh it out some more and maybe migrate it to the Wiki...

SSS4.pdf

P.S: It just deals with multiple scattering (none of the single scattering stuff there).

Cheers!

Link to comment
Share on other sites

I was knocking about Wikipedia and came across these links:

18252[/snapback]

Thanks for the references Jason. Very interesting stuff.

I was familiar with Rayleigh and Mie scattering from using them in atmosphere/volume shaders, but the coherent-light (laser) effects are pretty mind-boggling (not that I can follow all the dense math in there, just the main descriptions).

As you say... "fun read", thanks :)

Cheers!

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...