Reviewers: dev-osl_imageworks.com, osl-dev_googlegroups.com, Description: I forgot one important case in the get_light_side function - some closures receive light from both sides simultaneously (like

Reviewers: dev-osl_imageworks.com, osl-dev_googlegroups.com, Description: This adds 'spline' to OSL. This op-code is a bit on the heavy side w.r.t. templates and that's due to the variety of ways this

Awesome, Chris. Would you believe we were just talking internally about this topic yesterday? We were considering the amount of waste if there were big gaps of "off" points in the middle. But I think

Reviewers: osl-dev_googlegroups.com, dev-osl_imageworks.com, Description: We had already been executing shader layers lazily (only as params are "pulled" from downstream shaders), but we started by bi

I sent from wrong account. Posting again to avoid bounce from SPI alias... Good point, Chris. An image would be great here. Might I suggest writing the shader so that it interpolates the spline based

Reviewers: , Description: Pointed out by "crunchvfx" in http://code.google.com/p/openshadinglanguage/issues/detail?id=55 Please review this at http://codereview.appspot.com/209049/show Affected files:

Reviewers: , Description: This change, at Chris Kulla's suggestion, is to not fully bind parameters even when it's time to run the shader body, but rather to bind the parameters individually as they a

Reviewers: , Description: We've done extensive analysis showing that for typical shading batches we see in the wild, our method of having an array of true/false runflags is very wasteful, and despite

Reviewers: osl-dev_googlegroups.com, Description: The hair specular as it is now can't be analytically normalized in closed form. But we can ensure that it never integrates to more than one. Its maxim

Reviewers: , Description: I noticed recently that struct field parameters were unnecessarily generating "init ops" to set their default values, rather than relying on the constant initializer values l

Hi, I have read the OSL specification document and I have a few questions about the usage of OSL.... 1 - We have no access to the Viewer/Light position... so, how can we support distance attenuation..

Hi, I contact you in order to ask a few questions about OSL : 1 - In several other renderers they support the PDF, photon scattering etc... how can we support this in OSL and integrate OSL in existing

Hi, In the AshikhminVelvetClosure you have the following code : float cosi2 = cosNO * cosNO; float cosi5 = cosi2 * cosi2 * cosNO; float F = m_R0 + (1 - cosi5) * (1 - m_R0); It compute this : R = Ro +

Here is the fix for the NaN problem (along with a few others). I've verified that everything still passes testsuites in runflag, indices and spans modes. http://codereview.appspot.com/223105/show

Reviewers: , Description: Bug fix: a recent patch wherein I removed unnecessary init ops for structure elements was overly aggressive, and then we weren't generating init ops even for structure elemen

Reviewers: dev-osl_imageworks.com, osl-dev_googlegroups.com, Description: oslc prints a message by default that shows what is being compiled to where. This can be annoying for testsuites or other scri

Reviewers: dev-osl_imageworks.com, osl-dev_googlegroups.com, Description: This patch basically adds some new transform strings, like 'NDC', 'screen', etc... This also moves some of the functions in ex

Reviewers: dev-osl_imageworks.com, osl-dev_googlegroups.com, Description: This patch adds a few new transform strings, like 'NDC', 'screen', and 'raster'. This also moves some functions in exec.cpp so

Reviewers: dev-osl_imageworks.com, osl-dev_googlegroups.com, Description: I forgot that we pass all unknown args over to cpp. On linux this produced a warning that -q was not a recognized option. Plea

Simplify the logic used for binding connections. Also, fix a problem when the source of a connection is an input parameter. Due to lazy param binding, it is possible a parameter is still un-initialize