Can use some critique on my Mental Ray lighting tests.


#1

I’m doing some research with Mental Ray for a current project that I’ve just started. First render is a quick interior daylight only test render. Low quality (hence all the grain). I was focused on the overall feel of the light (Does it “feel” right?)

Second image is the ever popular simulated photoflash type render. I’d call this one an in between medium to high quality setting. Again, focus is on the actual feel of the lighting (not textures, or model)…thoughts?

Obviously with these render times, the configuration will be for still renders only. I doubt I’ll be able to trim the render times enough for animations…but with the satellite nodes now…who knows.

Thanks.


#2

heya… ur still the master! I’d have to say overall the daylight one looks good, but almost too bright… At my school there is a room just like that, and on sunny days it’s never THAT bright in the room. I’d say just tone it down just a little and it might feel more normal. As for the second one, good job… there’s def. a flash in there. It almost looks though, like the windows should be reflecting a bit more or something…? Keep up the good work man! Excellent stuff


#3

Looking good in my humble opinion. Especially considering it’s mental ray (not saying it can’t, just the extra miles you usually have to walk to reach there). :wink:

One thing bothers me with the lightintensity in the first image.
The door down to the right, now that thing glows unaturally imho.
See how it’s almost white, compared to it’s immediate surrouinding walls.
Guessing it’s the angle the light hits at making that.

I’d try to adress that, with either adjusting the angle of the light ever so slightly, or making a separate material for only that geo and dimming it a tiny bit.
Definately possible that it is correct, but it looks funny.

About the flash image, I get the impression a camera flash would have hit the furthest wall as well with a bit more intensity.
It looks like it cuts off from 100% illumination to some 20% in 2~ meters travelled right at the middle of the image.
Could be just me not having my morning coffee yet though. :wink:

Cheers, and very nice work with the mental ray engine overall.

Edit:
Typos and whatnot.


#4

Agreed with Didgeroo about the door in the first. Also, if you look at that beam that’s running across the ceiling, the lighting feels a little strange around the area where that beam meets the wall with the windows on it.

As for the second, you’ve got the falloff of the flash down, but something is still feeling off. I looked at the image before reading what it was supposed to be simulating, and I just couldn’t get a feel for what the lighting was. Maybe the lighting should be exaggerated just a little bit more? Or maybe this is just one of those cases where something that’s fake will appear more real to the viewer than an actual accurate simulation? I don’t know.
Thinking more on it though, I think that maybe what’s confusing me about the lighting in the second shot is the way that the scene has been framed. You get the bright glow on the upper beam and on the wall to the left, and then when you try and follow this to the center of where the beam of light has been focused, you get this dark gap in the scene; the area that the flash wasn’t able to illuminate. In that sense, maybe I was wrong in saying that the falloff is right. But I think it would help the feel of the lighting a bit more to have larger object positioned right about in the middle of those 3 front teapots to help ‘catch’ more of the flash.

But hey, I’m just a student, so take just take my critique with a grain of salt. :stuck_out_tongue:


#5

#1

The area above the window is too bright I think. this area usually doesn’t get that much lighting since the window is a bit reccessed.
Due to the amount of exposure of the Envioremental light (sky) I have a feeling the sunlight can be a little more bright with more effect on the GI

#2

Does the flash have a tint of blue?
I think it decays slower but you probably saw some refs so U know better
and no Envioremental light from the windows. is this intentional?
and again the glass should reflect something. it’s like its a peice of mat black film.

that said. it’s great effort. and you probably know better in these issues than I do.


#6

Very nice!

#1

  • The floor looks to dark.

  • The door. Most likely it’s the contrast between the floor and the door. I think it will look better with a higher quality setting because then you can see better that the light part is the floor under the door.

    #2

    • The floor again, it really has a ‘bitmap’ feel to it.
  • The teapots look unnatural. They don’t have any reflection that I can see (you would expect this from a teapot like that) and the saturated blue doesn’t help.

  • I’m not sure about the windows. You could try a Fresnel reflection for the windows and the teapots. With an IOR of around 1.5 and 5% white for the front and 95%-99% white for the sides. If you haven’t already that is. :slight_smile:

    Maybe try a concrete floor? That should look better.

    Good luck!

    • Rens

#7

Hi,

Great work as always jeff. Im just going to comment on the second one - It’s missing shadows, a flash that strong would produce pretty harsh shadows…that could be why it doesnt look so right.


#8

Very helpful input guys! I agree with all of it. I’m not sure why the glass doesn’t look correct. I’m using the glass physics phen material…so there’s not much I can change other than the IOR (which is correct at 1.54). But I agree it just doesn’t “feel” right when you look for the reflections…especially on the glass at the far end…it should look really reflective due to the fresnel effect. I’ll try some of my other glass solutions and see what they look like for reference.

The flash is using a single photometric light…so again, not much I can do about the settings there (specific to the falloff issue). The only thing I can think of is that the room’s scale is off. Since this is a test I’ll just leave that one alone for now. The model I’m making for this project will be to feet/inches scale so hopefully that will fix the light/dark falloff problem.

I think the harsh shadows are there in the flash render, but my light source is directly behind the camera. I’ll try offsetting that and see if I can bring those shadows out a bit.

I’ll take care of the floor problem on #1 lighting…I’m not sure why it came out so dark either. The bright door in #1 was also a texture problem (it had a reflective material on it).

I used that orange wood floor (and white(ish) walls) on purpose (color bleeding). I wanted to see if I could effectively control the bleed to achieve a normal look. I might even switch that out for a red painted floor or tile, etc. Just to further test some bleeding solutions I’m working on.

The only comments I’m not sure about are the overbright areas on number one. You guys might be right since the windows are recessed pretty deep here though. But I have many reference pictures of lighting situations and large windows in a white room = super bright in most cases. (even blown out white areas)…but again, I’ll double check. Maybe my environment light was too bright.

Again, thanks for the critiques…I’ll make use of this info and post some updates. :thumbsup:


#9

Hey Jeff, nice tests!
I think I agree with you about the brightness near the windows, I have no issue with that. The door is the only thing- but you have that figured out :slight_smile:

Flash render looks good, that is 1 trend I don’t get though LOL Whenever I see those things I wonder why someone would want to mimic a photo with poor lighting conditions, you did a pretty good job with it, just an issue with that type of rendering in general haha

Fun stuff, thanks for posting them!

Todd


#10

Yeah, for sure, the whole flash photography thing is weird, fad fad fad

Anway, enough griping

Some ideas… Could you maybe bake most of the lighting that you can see in this scene? Maybe break down your textures just so that their diffuse component is evident and then render that out with the texture baking engine. It works on a renderform these days I think, so you could batch process the scene that way. Then bring in those maps and use them as colour self illumination maps and you have baked GI? Obviously this wont bring the rendertimes down too low, but it will mean the main render hit is the ray-traced glass and a couple of other things, but it should render like the wind.

Worth a try? And if so do share. Great lighting BTW :wink:


#11

LOL, yeah I had second thoughts on posting that flash render…but someone HAD to try it with mental ray…might as well be me…lol.

Chris, good idea, I didn’t think of baking the solution. I’ll have to look into that. Never tried texture baking before so I don’t know how much luck I’ll have with it…but if it saves time…it’ll be worth a shot.

I’m thinking I’ll be ok with the render times once I start using a saved FG map…but of course that first render to make the FG map will be the beast.


#12

I did a test a few years back on the FG map in XSI. Its a bit sneaky, but from what I can tell, what it is is a render of the FG from the camera. Then at rendertime Mental Ray just loads in that image and uses it for that element of the render comp. Its not really usefull for animation as it simply HAS to re-render the FG map each time the camera moves. But yeah, for stills it is a time saver.

So it looks like you’d have to use render to texture util to get a lighting solution you can move within. I tried it out the other day with the amb occ shader in MR, and it worked just fine. I think you have to setup your materials to only have the element you want baked i.e. diffuse + shadows, but other than that it works a treat. A cool way to have two shaders on one object BTW is the Shell shader. I use these with scripts (to switch between the two options) and their great for switching shader sets. What I mean by this is one shader set for your texture baking, where the textures have no reflectivity or spec etc, and a second material for the actual final render, where the amb maps produced by the first shader are loaded in to each material to provide their baked ambient component.

Cheers


#13
That doesn't sound right. The photon map is a [i]3d[/i] map of where all the FG photons have been calculated to be in the scene, so it shouldn't be dependent at all on the camera view. It should only have to recalculate if anything in the scene changes. Maybe XSI automatically has the FG recalculate when the scene changes, and it wrongfully interprets the camera movement as a change in the physical scene and forces an unneeded recalculation(Max is stubborn in this way - it won't recalculate even if you move objects in your scene or change the perspective, you have to tell it when to recalculate yourself)?

The attachment shows what I mean. I unchecked the ‘rebuild’ option, and set a green sphere up against a wall to get a lot of color bleed. The FG calculates for this first time. I then drastically changed the perspective, and moved the sphere. You can clearly see that the FG didn’t recalculate.

EDIT: Or were you just referring to the way that XSI handles FG?


#14

I know you haven’t done textures and props yet, but wouldn’t having those in there dramatically change the lighting? when doing lighting test wouldn’t it be better to have all the textures in there first? especialy when using GI lighting?


#15

well, i think the pictures are great and for me they feel very natural. however, finished textures will make them just better.
the only think that feels unnatural is the sunlight on the floor in the daylight render, i think it’s a bit overexposed. it may be caused by the texture (maybe it’s washed out in dark areas like jpegs tend to do - i don’t know…) but that’s the same thing i often encountered in my own lighting tests and never managed to get rid of.


#16

True, Photon Mapping is not based on the camera, however, I am pretty sure gathering is (Final Gather). So while the photon map may not need to be redone each render (assuming camera animation), the Final Gathering would.


#17

But still, isn’t the generation of the photon map the greatest part of the final gather calculation process? I’ve done some tests with reusing the solution, and reusing the map has dropped the final gather time by at least 75% in all cases. The time it takes for MR to process the photon maps seems insignificant next to the amount of time it takes to bounce thousands of FG photons around a scene.


#18

Vormav, your getting your terms mixed (for a good reason). Let me state it as I understand it?

Traditionaly Mental Ray can produce both diffuse and specular (caustic) GI using photon mapping. The only issue with this is that the photon only solution is usually patchy and a little coarse looking (sooty) and also prone to fluctuating from frame to frame if you had moving objects or lights. So again traditionally in MR GI was photon mapping based yeah? And the photon map was based on world space, not camera space, so you could move your camera and the photons would still be valid which is cool but you could move objects and not re-render the photons and that would leave ghost objects (as seen in your corner sphere test image).

Now… to arrive at a smoother solution a final sampling system was added to sample some final GI values which could be used to reach a much smoother/pleasing result. This was called FINAL GATHER (i.e. final as in a final stage to acheive good quality). So, the Final Gather option was something added to help out with photon mapped GI. And the Final Gather result WAS camera dependant as it took what it already saw in the render and refined it. Move the camera and the Final Gather solution would have to be re-calculated. And this solution was a mixture of Photon based light and Final Gathers re-working of that info.

Ok, so this is where it gets messy. Some people noted that you actually got a pretty good skylight like GI effect if you didn’t bother with the photon mapping and simply just rendered with Final Gather turned on (with a self luminous sphere or an env map to act as a sky colour). I say skylight “like” though as Final Gather also calculated bounced light also, with colour bleeding, and imho because of this I’ve found it un-suitable for the role of a pure skylight (its gives crazy colour bleeding effects in places and also overbright light bouncing in others.

So after a while when people refered to GI in Mental Ray they often called it “Final Gather” where as in fact the Final Gather system was originally designed as a buff for its photon based GI.

So, there we are. If your talking about caching the “Final Gather” solution (i.e. no photons involved) then it IS camera dependant. If your rendering using ONLY the photon solution then they are already cached and that is not camera dependant. Problem is, photon only GI in MR tends to be pretty cack, so lets forget this option. Third scenario is rendering photons with final gather turned on to smooth out the photon sootiness effect. This scenario again IS camera dependant because Final Gather is involved, and it is view dependant…

So, there we go, hope that makes sense. So MR’s caching of the final gather result is actually it saving out the GI only component of the current render to re-use in the next render. That can be Photons + FG, or FG only. If you move the camera that cache file is useless. Believe me, tested it (3 years ago now) and its how it works (the cache file if you can find it on disk can actiually be loaded in photoshop and it looks like a GI only render).

Cheers


#19

Problem: Black spots appear in Mental Ray renderings.

Reason: Final Gather does not work with Multi/sub materials with Raytrace reflections that are “FACETED”.

Solution: Turn “FACETED” OFF.

See 3DS MAX files attached.

http://www.3darchitecture.net/samples/Mental_Ray_BLACK_SPOTS_SOLUTION.zip

Kenneth Loyd

http://www.3darchitecture.net/samples/Mental_Ray_Simple_Test_(BLACK_SPOTS).jpg

http://www.3darchitecture.net/samples/Mental_Ray_Simple_Test_(NO_BLACK_SPOTS).jpg

Kenneth Loyd
3DArchitecture


#20

Really? You mean the *.fgm or the *.pmap? or is there something else?
how can photoshop go about opening this? Is there an import/export plugin in?

Thanks