Maxwell Render | Official public forum

Hi Jeremy,

i am just getting a little puzzeled with the angle of sunlight.

I am using the rendered shadow viewport with an 8k shadow map to simulate the exact angle of sun for a montage.

however what i get in the render is different from the viewport.. not much.. but i'd say about 10 min timeshift.

any idea if that is my scene or if that is normal.. (the sun angle is very flat)

cheers,

Daniel

however what i get in the render is different from the viewport.. not much.. but i'd say about 10 min timeshift.

any idea if that is my scene or if that is normal.. (the sun angle is very flat)

How does what you're seeing in Maxwell compare to what you get if you use Rhino render? I don't think there's anything wrong with how the plugin is figuring sun position, so I'd guess any discrepancy would have to be related only to the viewport-shadow code in Rhino.

it's strange .. i compared the them again more carefully.. and found out that maxwell shadow (in my case: adelaide, june 21, 15:44) equals to rhino's at approx. 3:00.

so about 45min.. if it was an hour it would make sense..

Sorry, but I still don't know if you are talking about comparing Maxwell to the viewport or to Rhino render. I did some tests early on, and the viewport was matching up very well, but I think that the accuracy is very dependent on scene size.

i mean the Hardware viewport shadows.

i will test rhino result now...

rhino hardware matches rhino software.. here maxwell hard sun shadows compare like this:

It might be too obvious, but is scene rotation on the location & time page set to zero?

ricardo wrote:It might be too obvious, but is scene rotation on the location & time page set to zero?

nope.. thats not it.

when i forward the clock by approx. 40 min the shadows match.. maybe that is still the problem with the way the sun is responsive to scale of the scene? .. i remember threads about this. like the altitude bug..

My scene is in cm.. ~ 500x500 m in size.

I'll look into it then...I do my own math to figure the sun position, but sometime along the way, functions were added to the Maxwell SDK that allow me to get it from there as well, so I'll try using that instead. Still a little strange to me that they don't match up though...it actually took a long time to get the math right originally, so that they did match up, and they did, so apparently something must've changed.

Could it be related to sun time vs clock time? If one of them is working off the astronomical time that could explain the difference?

Just a thought...

Peder

Found this calculator online:
http://www.go.ednet.ns.ca/~larry/orbits/jsjdetst.html

When I put in the data you provided it (and put Adeleide at 138.5 degrees long looks like the "Siderial time" is 41min 45 seconds different. Or at least that is how I interpret it.

I am a complete noob at this stuff but the difference looked close enough to the symptoms you describe that I thought it might be of interest.

Perhaps Rhino is looking at the stars and Maxwell is looking at the sun. Or was it the other way around?

Here is a Wikipedia article on the subject:
http://en.wikipedia.org/wiki/Sidereal_day

Peder

Some timezone correction on either calculations?

I doubt it - the discrepancy in the comparison looks like alot visually, but percentage-wise, in three dimensions, it's really not very much at all. I'm not a mathematician like the NL guys are, so probably my code is neglecting one small variable or something like this. It may seem simple to look at, but it's by far the most complex calculation I've ever done, that's for sure.

Here's a test using the sdk function...looks better:

nice..