TheKitchenAbode

As per your signature "Smokin!!" Yes I know, your rig is right up there with the best of them. I was doing ok with X9, never any real issue but with this new PBR feature in X10 my system is definitely showing it's weaknesses. I really hate the thought that I'm going to need a new system just for this one feature.

Pleased you found this helpful. Just thought it might be useful as it was my first time using the PBR feature as an integral part of my workflow on a real project versus my previous explorations which have just been on an experimentation basis. It definitely revealed the need for top notch hardware if one desires to do some serious PBR'ing. Although Ray Tracing is also demanding it had the advantage of running as a background service which allowed one to continue working in CA. PBR'ing does not currently permit this and as such multi-tasking within CA is not very viable. I currently have a project with 8 tabbed windows, 6 being PBR cameras; switching views(window tabs) and making changes is very frustrating and time consuming. Hopefully CA will be able to make some improvements in this as the feature is attractive, and once you play with it for a while you definitely want to use it.

Use the material selector(Rainbow) click on the pic to open the Define Material properties. In the texture tab you will see Scale, sizing options, select stretch to fit. This will fill the frame with just one iteration but you will need to resize the frame to the proper height to width ratio as your pic.

That's interesting. I'm running one now that has 60 active lights and the PBR takes about 18 seconds to complete. I don't mind the time per say, it's just that I can't do anything else in CA until this process is over.

I'm looking to upgrade the video card but the prices right now are killing me. Are you actually using 100 lights or is that the setting you have in the DBX. If it is just in the DBX then it has no impact, all it's doing is telling the PBR to include all lights up to 100 lights if you have that many in your plan that are turned on and assigned to that camera.

I suspect it is the video card, it maxes out at 100% while the CPU is only about 25%. Also, it will depend on how many active lights you are using, the more lights the longer it will take.

Here is my experience with PBR on my first real project. This is an existing home that required modeling and the generation of a conceptual main floor. Everything is CA including all objects and materials. All scenes are generated using the new PBR camera feature. Here are the resulting scenes. Work Flow Working with multiple PBR camera view windows was extremely frustrating and very time consuming. The main issue here is that every time a window is activated the PBR regeneration occurs and this results in a 15-20 second delay each and every time. Also, when making any changes in an active PBR window it automatically forces a regeneration, another 15-20 second delay. The big problem is that this PBR regeneration is initiated for each and every change, there is no way that I know of to perform multiple changes and then manually initiate the PBR regeneration. I would estimate that between the 7 scenes I easily made more than 100 changes to objects, materials and lighting, which at say 15 seconds per change results in at least 25 minutes of regeneration time, this does not take into account the number of times I flipped back and forth between windows. What makes this even more frustrating is the fact that while the PBR regeneration is taking place CA is in full lock-down mode, no other work can be done during this process. Lighting Lighting was very challenging and required considerable tweaking to get things under control. As with Ray Tracing, lighting is extremely crucial to how materials will appear and shadow strength. The sun intensity was a major issue as I was generating both interior and exterior scenes. An intensity that looked good on the interior resulted in an exterior that was way over lit and as the sun intensity settings are global there is no way to have specific sun settings assigned to a particular camera. Material Properties Extremely challenging. Unfortunately, I found the little tea pot a very poor indicator of how a material will look when making changes in the material properties DBX. As such, the scene needed to be regenerated for each change for me to really see the change(s) affect. Also, as the materials properties appear to be highly sensitive to camera/light angles there were many times when I needed to flip between window views or zoom/pan to check out the effect, anther 15-20 second delay each and every time. Computer Resource Usage Very demanding on everything CPU, GPU and RAM. I actively monitored this throughout the process, as the number of PBR windows increased and the lighting complexity grew so did the computer resource usage and the time to regenerate a PBR. RAM usage was a really big surprise, never a problem before but when PBR'ing my meager 8GB was easily overwhelmed and the swapfile kicked in all the time causing further slowdowns. Every PBR regeneration maxed out my GPU which again results in slowness. The other issue with this high resource usage is that it also affected other programs such as my browser which became sluggish during PBR regeneration, likely due to my GPU being maxed out. Conclusion Despite all of these frustrations, I was reasonably pleased with the end result and would say they are as good as what I would have expect from a Ray Trace. Did I save any time over Ray Tracing, not sure and it is likely to early to say, would need more time to learn the lighting and material properties as they relate to PBR'ing. Recommendations 1.) Make PBR regeneration manually activated so multiple changes can be made before regeneration. 2.) Allow windows/tabs to be changed/flipped without forcing a PBR regeneration. 3.) Improve the tea pot display reference to provide a better indication of a material property change. 4.) Restrict or allow us to allocate PBR resource usage, similar to core assignment in Ray Trace. 5.) Allow each camera to have individual settings for all camera related elements including the sun.

It certainly seems that in PBR there is a high degree of sensitivity concerning the cameras position. In a PBR you can see some drastic changes in light intensity, and material properties as the camera is being moved around. I believe this is due to PBR's heavy reliance on sun light entering the room as both directional and indirect light. As such, it can be a bit challenging if you are shooting the same room from several differing camera angles. Also, when making material changes the effect you see will depend on the particular camera you are using at the time, switch to another camera and the material can look completely different, especial with those polished and reflective ones.

Rene - Absolutely Stunning Work! That's worth more than $6,000. I hope the client appreciated the skill/expertise and time you put into that.

Here's another example of some strange thing. Check just where the two recessed lights cast onto the wall, jagged shadow/shading. This comes and goes as the camera is moved around.

Developing a pricing strategy involves a number of factors. 1.) Competitors. 2.) Your target market. 3.) Your level of expertise/skill. 4.) Your rendering level of quality. 5.) Your level of productivity, time spent. There are many possible combinations in the above that will affect/determine what you can charge, or maybe better said what a potential client would be willing to pay. The client you worked for indicated that your desired pricing was too high and when you yourself were considering an outside rendering service you also came to the same conclusion about their cost. Did you discuss with your client as to why they thought your price was too high? Is that customer representative of the type of customer you are intending to target? Was your pricing estimation just based on what you believe others are charging or did you derive it based on the value of your time, skill and profit margin needs? There are two distinctly different approaches. 1.) This is what clients are willing to pay and as such how do I deliver a similar service at that price and make money. Or 2.) This is the service I offer and this is the price I need to make money and as such how and who do I sell this to.

Here is an example of an exterior PBR with two additional point lights added to overcome those strong shadows. PBR with sun only. PBR with sun and two exterior point lights, one in front and one on the right side. The two exterior point lights have their shadows turned off, are placed about 25' high and about 25' away from the structure. The shadow depth can be altered by adjusting the intensity of the two point lights.

The only method I have found to soften those PBR shadows on exterior shots is to place a spot light or point light outside and positioned to throw light at those shadows. I usually have their shadows turned off so they do not create conflicting shadows.

Not sure that there is more "consistency" between PBR and Ray Tracing in respect to material properties unless this is in the context that they work in both. What I have found is that the degree to which they render varies greatly between the two rendering methods; polished, glass, reflectivity(specular) and bump maps are prime examples. If their properties are adjusted to look reasonable in PBR then they do not look good in Ray Trace and vice versa. Lighting is also another area where there seems to be a disconnect between the two rendering methods. Again, settings for one do not often result in similar results in the other.

Reflections are still challenging, their visibility seems to be highly dependent upon the cameras angle and the reflective light source. Yes, glass is a big big problem, I usually change the material to General and play with the colour, add a slight texture and adjust the transparency and specular settings. This tends to work fairly well if the glass has some color in it such as the bluish vase on the floor in my kitchen. Clear glass, no matter what I do, just doesn't PBR properly.

Island color variances solved. Have had this happen before, for some reason even though the materials are all the same they PBR differently, some components will look brighter or duller than others. To correct this I just use the paint dropper, select the material and then reapply.

The light splotches on the beam are from recessed light fixtures on that side. Yes, there is a bit of a lighting or material property issue on the island components, I'm checking into it.

Thanks Mick. Here is an example, this PBR has the following light intensity settings: Exterior Sun - 8.0 Lux Recessed Lights - 15 Lumens Under Cabinet Lights - 0.5 Lumens Island Pendants - 15 Lumens 3D Area Lights - 4 Lumens As can be seen, even with the sun intensity at only 8.0 Lux the light cast on the floor through the exterior door system is fine as are the under cabinet lights at only 0.5 Lumen. If the overall scene is a bit lacking in brightness then just boost the brightness in the PBR DBX. Also, the fact that I'm using a very low sun intensity results in amore realistic exterior light levels, the backdrop is not way over exposed and if I shoot the exterior it also shows a more correct level of lighting. There may be no relevancy but when using lower settings I have not yet encountered the situation where the scene completely over exposes. Just wondering if that white-out situation might be related to a condition where one or more lights are just way too intense and the camera position is such that it is looking directly into this high intensity light source.

Just an observation related to light intensity settings. When PBRing it appears that the intensity of lights are based on some relative relationship versus absolute values. In other words, the actual intensity in lumens does not seem to matter all that much but the relative difference between the lumens of one fixture to another does. For example, if there are two light fixtures and one is 1,000 lumens and the other is 500 lumens then the most important aspect is that the second fixture is half as intense as the first. As such, if the fixtures lumens are 100 lumens and 50 lumens or 10 lumens and 5 lumens the results will be similar as the relationship still remains. This also appears to be tied in some manner to the sun's intensity setting. Just guessing at this time but I believe this is related to some form of auto adjustment feature that assumes that the highest lumen setting will represent the maximum luminosity value in the scene and all other luminosity values are determined relative to this maximum based on the lumen intensity relationships. If this is true, then one needs to be careful to not place a light that has an extremely high lumen value compared to the lumen value of other lights as it's effect on the relationship would effectively diminish the other lights contribution to the scene, if it is high enough the other lights could be reduced to the point where they have no effect at all.

Which recessed can did you use? I used the 4" one and it was fine. Try reducing the cut-off angle, the 4" and 6" ones are at 175 degrees, see if say 90 degrees make a difference.

Thanks Greg - The rings I believe are related to X10's modified rendering method, I've seen this happen in some circumstance. Seems to vary depending upon the lights. For the disappearing crown you need to have some level of ambient occlusion, for a standard 3D camera view I usually have the slider set between 40% - 60%.

That may be so, but your preferences and default settings will be different from ours. Posting your plan with the problem would allow for some level of comparison. Also, the rings could be related to the video card. The issue with the crown is most likely related to the light settings, camera settings and the 3D view settings. There are more options in X10 than X9 so some tweaking may be required.

Just be aware that unchecking these will disengage certain rendering features. The new PBR Camera will not function correctly without Global Illumination being active. Those weird rings must be due to some other anomaly. Suggest posting that plan to see if we can spot what's going on.

Yeah ! the default sun intensity setting seems crazy, it completely blows out the background. Depending upon the scene and the backdrop I use between 100 lux up to 10,000 lux. I also adjust the Camera Exposure between 0.2 and 0.3, Brightness about +5 and Saturation +10. In this one the sun intensity is 4,000LUX.

Here are the finals after a few more lighting and material adjustments and a minor amount of colour balancing was done in a photo editor. They are at 3690 X 2000 should you wish to take a closer look at the result.