All posts relating to Maxwell Render 1.x
User avatar
By jo
#141058
Mihai wrote:I think caustics in a mirror seen through this glass is possible.
http://www.maxwellrender.com/files/ags.htm
It seems to be possible
The only strange thing is that they are not changing too much in the mirror.

Ciao, Jo
By DELETED
#141061
DELETED
User avatar
By hyltom
#141062
jo wrote: It seems to be possible
The only strange thing is that they are not changing too much in the mirror.
But there is no glass between the mirror and the ball.
By ajlynn
#141063
hyltom wrote:
Mihai wrote:..and hyltom, I think caustics in a mirror seen through this glass is possible.
I'm not so sure, because if you look the mirror, there is no glass before the ball. One side of the box is open and is the one that is reflected in the mirror.
Anyway, I don't really care about that matter. It already so incredible all the news that you bring during this week. I really want to say, THANKS to all of you (NL and A-TEAM). You make a really good job.
From my understanding of all how all this works, the architectural glass should no longer have the caustics problem. Mirrors still would, when used in combination. E.g., that test scene of Vlado's design that I posted renders of back in January, where a small sphere emitter is in a reflective desk lamp style fixture, most of the light in the scene is caustics and you don't see it in a mirror, still wouldn't render, but the case of an exterior render of a building with glass curtain wall and sun shining on it would be fixed. That's good enough to keep me happy.

Oh, right. A-Teamers: any word on IES files? IES lights would eliminate the need for the reflective light fixture.
By ajlynn
#141067
jo wrote:
Mihai wrote:I think caustics in a mirror seen through this glass is possible.
http://www.maxwellrender.com/files/ags.htm
It seems to be possible
The only strange thing is that they are not changing too much in the mirror.

Ciao, Jo
It's changing quite a bit in the mirror. The bright rectangle is from the sun coming through the glass, and there's the ball's caustics, which might be explained by the AGS image being at a higher sampling level (compare the reflection caustics from the back side of the mirror)... Hey, that's interesting. Is this scene lit by sun? In beta that reflection of the glass ball's caustics wouldn't have worked. Did they make more improvements to caustics besides the new AGS?
User avatar
By misterasset
#141068
ajlynn4 wrote:any word on IES files? IES lights would eliminate the need for the reflective light fixture.
IES lights would be the ultimate surprise this week. My boss is really impressed with what Maxwell can do, but I don't think he'll fully embrace it until it can use IES files. But at least he does let me use it on projects where the client doesn't ask for specific lights in the rendering.
User avatar
By jo
#141070
hyltom wrote:
jo wrote: It seems to be possible
The only strange thing is that they are not changing too much in the mirror.
But there is no glass between the mirror and the ball.
Ops! :oops:
But my question remains about the same refliction of caustic (without glass)

Ciao, Jo
User avatar
By tom
#141073
Hi,

We're glad you like AGS glass, thank you for nice replies. Basically, AGS glass is made of 2 layers, one is ghost, this means no refraction, no reflection, no caustics, simply nothing. The other layer is a low Nd mirror which gives the fresnel look of glass. These 2 layers can be blended normally or additively in any percentage for producing this result which is suitable for simulating thin window glasses. It's not an additional shader, parameter or hack. It's the native result of new powerful maxwell shading system as a material preset.

Best regards,
Tom
Last edited by tom on Thu Apr 20, 2006 6:06 pm, edited 1 time in total.
By ajlynn
#141075
Maximus3D wrote:There's that other preset light/lamp thingy's they added now in 1.0, sorry i forgot what it's called :D

/ Max
I like that feature. Now I don't need to look up the specs of light bulbs all the time. But what I meant was IES files that have information on whole light fixtures - you can get these from manufacturers, they contain the info on how much light the fixture puts out in different directions, so you don't need to calculate it with caustics/GI and a very good model of the fixture..
User avatar
By jo
#141076
tom wrote:one is ghost, this means no refraction, no reflection, no caustics, simply nothing.
So, why is there?
Thank you Tom,
Jo
User avatar
By tom
#141080
Hi Jo,

The ghost is there because it lets the light in when blended with mirror layer. The percentage of ghost determines the amount of light going through and the percentage of mirror determines the light to be reflected back. In addition it's possible to have more layers for numerous effects and different reflection roughnesses or mapped reflections for creating metallic, solid or semi-transparent decals. Simply it's limited with your imagination.

Best regards,
Tom
User avatar
By Mihai
#141084
misterasset wrote: IES lights would be the ultimate surprise this week. My boss is really impressed with what Maxwell can do, but I don't think he'll fully embrace it until it can use IES files. But at least he does let me use it on projects where the client doesn't ask for specific lights in the rendering.
This doesn't make sense to me, I thought IES simple tells how the light spreads, but can you have an IES which specifically gives out say 1300 lumens? And if it does, how do you know the light levels in a scene would be accurate in a renderer like finalrender, or mentalray?

What would be more accurate?

Having an IES file which gives you a nice shape of light, but because of the way the renderer works, the lighting given off by that light will not necessarily be accurate, for example you made the render to look nice with colormapping etc, which will give the impression the light is stronger than in reality.

Or, have the ability to specify exact lumens output of a light (for example, there are more options now for emitters), and model the light fixture?

Ofcourse having IES in Maxwell would be great, so you wouldn't have to model the lights (although many light models are already available for free from light manufacturers), but if your boss is worried about accuracy, then how can he trust a render where you usually have to change the colormapping so parts of the image won't be blown out, or too dark?
Last edited by Mihai on Thu Apr 20, 2006 6:17 pm, edited 1 time in total.
By Romans
#141086
Hello Next Limit,

now the love is endless!!!!! :D

I am sooo happy about this AGS feature!

Really, really GOOD.

Thank you so much for developing AGS!!

Kind Regards,

Roman
  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7

> .\maxwell.exe -benchwell -nowait -priority:[…]

render engines and Maxwell

You could be right about AI, but actually I prefe[…]