View Full Version : FG Reflection bounce capping FG Diffuse bounce?
06-01-2007, 09:51 AM
Final gather secondary diffuse bounces is being capped by the number of FG reflection bounces in render settings window.
If I have 100 diffuse bounces (just for test) and only 1 FG reflection bounce it limits my diffuse bounce to 1, the only way to fix this is to put FG reflection also up to 100.
Can anyone else confirm this? seems like a bug to me.
MacOSX, Maya 8.5.
06-03-2007, 03:12 PM
here's the reply from Alias support:
I am able to reproduce this but I believe this behavior is expected. The diffuse secondary bounces are either (diffusely) reflected or refracted. One can reduce the number diffuse bounces but one can not increase the number of diffuse bounces past the reflection limit.
Is this what everyone would expect?
06-04-2007, 03:28 PM
...Is this what everyone would expect?Ha, ha, LOL,
Me as a user with some Mental ray experience, I am already “corrupted” – meaning I always expect a lot more than a single action to set things with Mental ray. In fact in this case I feel pleasantly surprised that the Reflections in the Raytracing sections doesn’t put yet another cap too and feel lucky that that it is only two places needed to be checked.
Yeah, I agree, it is pain using Mental ray in Maya, it seems to me that the people who integrated it were thinking that the biggest thrill is to click and drag things as may times as possible. Why on earth increasing reflections or refractions should be capped somewhere else? Apparently when you increase it, you want it this way, and any cap should be automatically increased to reflect your last changes. If later on you go and lower the settings globally from a cap, then fine, respect that. But the philosophy should be based on what the user wants last not what was set previously. And what the hell is Max Trace depth for? As you could just type your reflection and refractions and then think “oh may be its too much lets reduce it in another entry field” as if one cannot reduce it in the respective fields that are right next to it and also clearly show how much the reduction will affect reflections and refractions.
Sorry for the rant but it just grew naturally while typing my reply and I couldn’t resit.
Yeah, I agree, it is pain using Mental ray in Maya, it seems to me that the people who integrated it were thinking that the biggest thrill is to click and drag things as may times as possible. Why on earth increasing reflections or refractions should be capped somewhere else?
This behavior is not caused by Maya. See here:
06-06-2007, 12:47 AM
Thanks for your reply Chris.
Yes, I know that it all comes from Mental ray, but here we are using its Maya integration. Although I’m not very familiar with Mental ray standalone, from what I gather it is not something for the average user. It is just a common sense that if it was, everybody would simply go and buy it, and nobody would ever bother integrating it with other 3d programs and trying to make it more usable. Isn’t it? I think it is up to Autodesk to make the user experience including the learning curve more efficient.
Anyway, I also would like to thank you for the link you provided. I checked it out and thought that I was learning a few new things from it, but then I actually got more confused after I tried a few things including reproducing your experimental scene just to check if the things work as explained there.
Since the linked thread is already closed, and I can’t post there, I’ll try to continue it here. Just to clear up any confusion, with respect to the FG trace depths...
The FG diffuse depth, or number of 'multibounces', is the number of diffuse interactions, whether they are reflection or refraction.
The FG reflection depth, number of reflections, is the number of diffuse, glossy, or specular reflections spawned by an FG ray.
The FG refraction depth, number of refractions, is the number of diffuse, glossy, or specular refractions spawned by an FG ray.
The FG trace depth total is just an overall limit for the total number of interactions spawned by an FG ray.
So, if you have only multiple diffuse reflections, as in the case of objects using only the default Lambert shader, then the trace depth will be the lesser of the reflection and the diffuse depth numbers. I don’t know if it’s me, the explanation, or the concept, but at the end I got totally confused from this explanation.
Let’s say I want 2 FG reflection bounces and 4 FG refraction bounces. So I have to type 6 in the 'multibounces' box, then specify in the other 2 boxes what is what from this total? So I type 2 for reflections, 4 for refractions, and then one more time type in the FG trace depth the total of 6 again.
Is this correct?
If it is and that’s all there is to it, I don’t see any reason as to why the 'multibounces' and FG trace depth controls are needed at all. Can’t we just simply type how much reflections and refractions bounces we need and that’s it?
Now assuming that there is more to it, here’s what’s confusing me is the number of diffuse, glossy, or specular Does this mean that if let’s say I have:
5 for 'multibounces', 7 for reflections, 9 for refractions, 16 for FG trace depth, then Mental Ray will make the total of 5 diffuse bounces reflecting and/or refracting depending on what kind comes first on its way only from diffused surfaces if it hits any? Then 2 more reflection rays (7 – 5 = 2) will be traced for glossy or specular surfaces if any. And then again 4 more refraction rays (9 – 5 = 4) will be traced for glossy or specular surfaces if any. Is this correct?
If it is, what happens if the FG trace depth limit is less than 16, then what kind of rays will be traced and in what amount and priority?
Now let’s take a look at this: In the Maya UI, there is only a checkbox for diffuse interactions (multibounce). When it is translated, it comes out as 100, which is usually much higher than any other of the numbers. This makes it effectively an enable for diffuse reflection, not just specular and glossy, and then you only have to pay attention to your FG reflection trace depth number. First of all does this mean clicking the checkbox for Secondary Diffuse Bounces in the Render settings, will set the FG Diffuse Bounces to 100 somewhere internally but be displayed as 1 in the FG Diffuse Bounces box in the miDefaultOptions tab in the Attribute Editor? Meaning the displayed 1 is actually 100. Then what happens if we type 2? Are we reducing it to 2 or making it 200? (oh I’m “falling in love” with the Mental ray integration, sorry for the outburst)
I spend some time actually trying this and it does seem to work along these lines and I really appreciate this piece of information because so far I used to dig for and go to the miDefaultOptions tab and type the number of FG Diffuse Bounces there, which apparently wasn’t necessary.
But I am confronted by yet another confusion. I recreated and rendered the experimental scene from Chris and when I tried it with just Secondary Diffuse Bounces checked in the Render Setting and all other values set to 1, it still worked as shown on the image below. This experiment completely defeats the bgawboy’s explanation about the meaning of the values as quoted at the beginning of my message. And it also defeats the conclusion from wizlon (the original poster) that if you have only 1 FG reflection bounce it limits the diffuse bounce to 1. As you can see in the image and the scene’s settings below the lighting comes through the corridor and lights the object, which should happen only with more than one bounce. Is this correct?
Does anybody have a clear idea of what all this means and how all this works? I would really appreciate any attempt to explain this. More people trying to explain it in their own words, the better.
06-06-2007, 02:38 AM
Well, as for your scene its not correct, because as you could notice, the side of the cube and the wall close to the camera are really black, there is where secondary diffuse bounces comes to help.If you increase the FG Diffuse Bounces to 2, you will notice that the parts i mention will get bright.As for the FG Refeltions/Refractions and FG trace Depth, i always thought that it was related to how much the FG points could pass or reflect through a glass surface, and that the FG Trace Depth was just a limit value that you could set to make the FG rays not to go above that number. But as i saw the post from Chris, changing the Reflection amount also changes the Diffuse bounce amount.Theres where im lost too :)
06-06-2007, 06:15 AM
Well, as for your scene its not correct, because as you could notice, the side of the cube and the wall close to the camera are really black, there is where secondary diffuse bounces comes to help... I'm not surprised that the side of the cube and the wall are black. In fact that is what I would expect, but what is actually surprising and the whole point of posting this is to find an answer of the question as to why there is light coming to this place and objects in the scene. With all values set to 1 there should be no light at all and the scene should render complete black. If you look at the top view of the scene you will see that the light casting object cannot be reached with a straight line through the corridors from where the illuminated objects and place are in front of the camera. This means that a ray has to get bounced more than once to get from the light to this place. That’s the mystery I’m trying to figure out in order to understand how FG works. Because the map visualizer doesn’t really show much about how the bounced rays are traveling I’m using this experimental scene to found out more about it.
06-06-2007, 01:03 PM
Does 1 really mean 100? I think this is incorrect.
Emil3d see the thread here - http://forum.lamrug.org/viewtopic.php?t=1114 - and post your explaination, if possible.
06-06-2007, 09:09 PM
Wizlon, definitely 1 = 100 bounces is incorrect. In a moment I thought it could be, being confused by the results from that test scene and bgawboy’s explanation but most likely he meant something else that is beyond me. Also after proceeding with my regular way of setting things and paying attention to how these values affect each other it is obvious that in order to set multiple FG diffuse bounces one has to type/ensure this in 4 places:
Secondary Defuse Bounces option checked,
Max Trace Depth value,
and FG diffuse bounces value in the miDefaultOptions node.
The lowest of these values equals the number of the possible FG diffuse bounces.
I’m still somehow lost about how the glossy and specular bounces come into play with all these numbers like in the example I posted earlier. It would be very nice if someone can make an explanation that can make sense for people like me:).
And thank you for the lamrug link and the cross pointer there to this thread Wizlon, I’ll keep an eye on it, and may participate if necessary.
06-07-2007, 12:46 PM
Wizlon, definitely 1 = 100 bounces is incorrect. In a moment I thought it could be, being confused by the results from that test scene and bgawboy’s explanation but most likely he meant something else that is beyond me.
I think he just said that in Maya GUI, way of implementation of secundary diffuse bounces is that checkbox by Default enables 100 bounces, and then it gets capped by FG Reflection and FG Refraction settings.
What is so confusing there?
06-07-2007, 03:44 PM
MaNdRaK18, thanks for replying.
It is totally confusing. He said that after enabling 100 bounces “… you only have to pay attention to your FG reflection trace depth number”
It beats me what “FG reflection trace depth number” is. Is it the Reflections attribute or Max Trace Depth attribute in the Final Gathering Options section of the Metal Ray tab in the Render Settings? In any case, putting both or all values there - Reflections, Refractions, and Max Trace Depth, to any number, will be capped at the end from the number entered in the FG diffuse bounces value in the miDefaultOptions node. And the FG diffuse bounces value in the miDefaultOptions node becomes 1 not 100 when you check the Secondary Diffuse Bounces.
06-07-2007, 04:20 PM
Two things are being confused here, in part due to the time span of the discussion.
1. Maya 8.5 translates differently than previous versions. Some of my quotes regarding the 100 bounces refer to the older versions. In Maya 8.5, enabling the Secondary Diffuse Bounces limits it to 1, unless you go into the miDefaultOptions node (the one shown on the right above), to change it explicitly.
2. FG trace depth is for secondary FG rays. Even if all are set to 0, you will see results from the primary FG rays.
Lately, the easiest way to teach these concepts has been to use a 'counter' metaphor. These numbers are all limits. Each time an interaction happens with a surface, one or more counters are incremented. When any of the counters hits their limit, no more secondary FG rays are sent out.
06-07-2007, 05:36 PM
In my miDefaultOptions node, though I enabled Secondary diffuse bounces and set FG Refl/Refr and Depth to 2-2-4, the FG diffuse bounces remains 100 (!) but in the output window there's that error message that says "warn 082069: finalgather diffuse depth 4 too high set to reflection 2 + refraction 2".
Am I missing something? I really can't understand.
06-07-2007, 05:57 PM
Then, I assume you are in a Maya version less than 8.5, right?
The 100 number will just be limited to 4, as it is _trying_ to report.
What do the numbers in the options report, using at least Info level verbosity?
06-07-2007, 06:45 PM
Thanks for stepping in and helping Bgawboy, This cleared most of the mysteries for me. I wasn’t aware that a value of 1 means total of 2 rays (one primary + one secondary). That’s way the test scene received illumination – when I extruded another corridor and moved the lighting object at the end of, the illumination come to the objects only with values of 2, meaning a total of 3 rays for the 3 corridors. They should have named the whole section with those values as “Secondary FG Rays” to make this more obvious.
It is still a little bit hard for me to imagine what numbers will limit what kind of interactions but I hope I will figure it out if I strain my brain harder. :scream:LOL Meanwhile, like before I’ll keep them high enough to ensure that at least what I want is there (at the expenses of performance I guess)
I have one more question though. I know it is not your responsibility but do you know by any chance as to why they changed the 100 to 1 in the miDefaultOptions node. For me this just adds a bunch of additional clicks to ensure one more attribute. Are you aware of any scenario where this change makes things easier or better than before?
Again thank you for your contribution here. I really appreciate it.
06-07-2007, 07:06 PM
It would be pure speculation on my part, but it is possible that it would discourage use of more than one diffuse secondary FG ray. Often at numbers greater than 2, it is speedier to use FG+GI.
Although the ease of setup for FG in a scene is often worth the tradeoff in overall rendering speed.
06-07-2007, 07:26 PM
Thank you for your reply. I was already aware of that situation form different sourses and personal experience but idefenitely it wasn't the change that made me aware of this and I'm not sure if that's the intention it will work with others, but anyway, I asked you this question to make sure I'm not missing something significant associated with this change.
06-08-2007, 05:08 PM
For what its worth, in the artist track mr training class, 102 Using mental ray - FG, we show how to use multiple levels of secondary diffuse FG interactions.
However, we recommend a method where higher levels are used as an illumination reference (even up to 10!), and that you use FG scale to scale up a 1 or 2 level FG diffuse bounce to match the illumination level of the reference. Then, you get the same illumination level at the faster speed. The illumination will be slightly different, but unless it is an extremely windy corridor, it usually doesn't affect the relative distribution of the light much.
06-08-2007, 05:08 PM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.