Show Us Your Bryce Renders! Part 4
This discussion has been closed.
Daz 3D is part of 
Connect
DAZ Productions, Inc.
224 S 200 W, Suite #250
Salt Lake City, UT 84101
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2024 Daz Productions Inc. All Rights Reserved.
Comments
@Dave: Thank you for the information. I'll try using it and see what happens. I'm familiar with what MRD does when used in creating abstracts, but not with an object like I'm working with. Set too high and the intensity of light increases. Set to low and things don't look very good. But I sense it works similar.
I know of TIR but not what it does. Judging from its name it increases the number of reflections within an object. If this is so will it give an object a better transparent look? I'm interested to see what affect increasing this has on the renders I'm experimenting with.
@TLBKlaus: Those abstracts look fabulous on a shirt. Congrats. That glass image looks pretty darn good.
@TLBKlaus - T-shirts look cool. Glad to see that it worked out for you.
@GussNemo - have a look at those 6 years old renders of jelly by David where TIR was changed from the first to the second. First: Staplers and Jelly, second: Staplers and Jelly with TIR6.
TLBKlaus- Cool T-Shirts Congrats. The glass is looking good.
Yes, from what I gather it is a sub set of MRD that increases the amount of times reflections are bounced around inside transparent object, potentially enhancing the look of more complex objects. In my example it made no difference as there isn't much to reflect around inside.
Interesting observations Dave thanks for sharing
@DocSavage64
Interesting collection. The result of nearly no difference between 6 MRD and 12 MRD can lead to a wrong conclusion, I think. With perfect spheres there are really hardly any differences, but you should repeat the examples with more complex objects. Once I used a glass with a lot of little bumps and rendered it with 6 MRD, 12 MRD and even 24 MRD and there were recognizable differences. I am afraid, I cannot show it just now, but maybe later.
Interesting results, indeed. However, simple objects do not show everything. Even though volume colour is said to have only an effect if inside such an object, the glass as seen from the outside gets a more saturated colour and the light shining through it gets coloured, as the example nicely shows. For my dragon, transparent and volume colors were both set to the same colour, not white.
@Dave: Those are very nice examples, I'm drooling while I write this. :-)
I did use the information you gave and discovered two things. 1) High MRD + high TIR = out of memory, the first time in the entire time I've been using Bryce that this has happened. 2) If all selections in the color section of the Mat Lab are set to white, increasing MRD or TIR will not produce a white object if transparency is set to 100%. With the atmosphere off, that color has to be white in order for the object to be white(ish). This I discovered in one of my "I wonder what happens if I do this" moments. The first image is the result of MRD 24, TIR 2, Transparency 100%, Atmosphere off and set to white, and diffuse for the sun set a bit lower. RPP was at 32. I then added a bit of color to Volume and Transparent to get the second image, which has RPP at 64; I thought it a bit much to wait 2 hours with RPP set to 144. I have to laugh at myself because after 40 iterations, all I would have had to do was set the atmosphere to white and I would have been where I first started. I know one sure thing, this lesson will not soon be forgotten. And thank you very much for the help.
@Horo: My untrained eyes can't see much difference between the two images. I perceive a difference, but not sure what it is.
@mermaid: Thanks.
I explained to Dave, above, how the first image was obtained. It really was a duh moment. The second image has a bit of orange set in Volume and Transparent, but rendered at RPP 64.
Good idea in such cases is to open both images in two separate browser tabs. Then click on the tabs to swap from one image to the other. This is some sort of a "blink-comperator" and helps noticing small differences.
Thanks Electro-Elvis and Horo. I wasn't being very clear when I wrote my post above and you are of course both correct.
My point was supposed to be that if there is no discernible difference, higher than necessary MRD is pointless and in some cases can still lead to much higher render times because the rays are bouncing more and yet the result isn't really worth the extra effort.
Of course, more complex objects will benefit in some cases from higher MRD and TIR. Then there will be other cases where overly complex objects may need less detail showing to make them sit better in the scene. It is like most things a judgement call depending on the scene and the intent of the artist (and as always to some extent a measure of the patience of the artist).
In this example I used more complex models to how MRD can pick up extra detail above MRD6.
The first one is set to MRD6, looks good enough and rendered in just over an hour.
The second one is set to MRD12 and has more detail in the glass objects, but took just under 9 hours to render.
For my money in these two examples, the extra rendering time hasn't resulted in a 'better' render, just a more detailed one. That extra level of detail can distract from the otherwise simplicity of a scene and maybe the artist is trying to get across a feeling of relaxation and not intensity. As such, I then made a totally postworked version to show how less detail can be effective depending on what 'mood' is being portrayed. :)
Setting the orange in the glass has done the trick here Guss. If you want to try it, maybe try keeping the orange in the glass and changing the ground plane colour to blue. You glass will go back to looking a bit smoky again.
I think another thing to consider and one of the things that originally cropped up with glass looking a bit like gummy bears (not something to concern you in your latest renders) is that well polished glass should have really well defined reflections. Specular can't achieve anywhere near the sharpness required so the only way to do it is using HDRI reflections (or having objects out of frame that reflect in the surface of the glass, but that can be problematic in it's own ways). Setting an HDRI in the IBL tab and turning the HDRI Effect down to '0' and tweaking the Intensity control up until you get bright sharp reflection won't increase render times too much. Though again the choice of which HDRI background to choose is one of personal choice, I would usually go for a very simple one with bright sharp highlights in it so as not to overly complicate the glass surface with too much reflection detail.
The quality of light you've got passing through the object and hitting the ground looks extremely natural too. :)
A little contribution from me on this glass topic. Bryce "Nuts and Bolts" - Embed specular highlights in a materal - by David Brinnen
Edit. Here in TA rendering, a test, using this approach and some ambiance to add light inside the glass material. Even still you can see a few black fireflies.
Edit edit. And with regular diffuse - but again, no "real" reflection.
Two landscapes, but can't decide which one is the best.
Opinions please.:-)
I like the haze density better in the lower image, but not the colour, which you could modify using colour perspective.
My reflection mapping experiments have lead me off in another direction.
I like the top one better; while the haze is fine for the distance, it seems too thick between the viewer and the building which isn't that far away. Unless it is intended to be smoke?
The building seems a bit distorted by the camera though in each, at least to me.
@wilmap - work on the second one. Haze gives a clue to distance but Sean is right, it is too dense in the foreground. And yes, camera FOV is probably a bit on the wide side considering how the building is distorted.
@Horo: This is something I may have to try.
@Dave: Love the look of those first two images. The third one actually looks like a modernist painting. That image has really nice effects.
I started all this wondering if good looking glass could be made without using an HDRI, after watching David's blue glass dragon. I have since learned glass can be made but not good looking glass. In my latest images above, even I can see reflections are missing(?) in some of the facets or are dull(?). I did find out specularity only produces a brighter object and not reflections, so I abandoned that track after awhile. It was your latest information which helped me discover why I kept getting a black object if only transparency was used. And now David's latest video, another very good one, has given more information I'm going to try with my sphere. BTW, I did think about changing the color of the infinite plane but felt I should get one thing looking as I wanted before trying something else. Now I'm curious to see how a blue plane affects the orange glass; or any other color for that mater.
@David: Another really good video, chucked full of information that just happens to be needed. This one I'm going to try with my current project, hopefully the dull looking facets will pop. Love that first dragon, and the effects. The second one looks like something drawn in a comic book, it has that style.
@Wilmap: To me, there's a focal point in your first image just to the right of the center peak. That bright spot of haze draws my eyes away from the rest of the image. I think the building, the background mountains, the foreground and color look really good. It's just that bright spot of haze that bothers me. In your second image I think you might have the haze too thick, which obscures the foreground colors. Adjusting the haze thickness will pull the haze back towards the mountains. As close as the viewer is, it might be a good idea to also adjust the intensity so there's haze but not a grass fire haze. As nice as that building is you might want to make sure its details can be seen clearly.
GussNemo, I think your comment was directed to Wilmap, not to me?
Aftrer having downloaded the free 3D trees from Xfrog, set the foliage transparent and put everything neatly into the objects library, I finally gave it a test. As far as the branches and twigs are concerned, you can zoom in as near as you wish. This cannot be done with Bryce trees because some thin twigs are not connected to the limbs, and things get angular. However, the foliage is a bit of a disappointment. -- at least for this model: they are flat. So there's a limit to how close you can get. Nevertheless, the plants are really very high quality. Two EU55 trees, a terrain and an IBL backdrop. The HDRI is a sky dome type without a sun.to provide ambient. The main light comes from the Bryce sun.
Not really into cars apart from rebuilding engines but thought I'd have a go at rendering a couple.
Credit to the model makers whomever they were. Downloaded them a long time ago and can't remember. The scene is actually set inside an arched glass roofed building, although it looks like it's outside. Reflections on the cars are from an HDRI.
The trees look very cool, and so do the cars... nice work.
Here's another one, having some fun w/intricacy this time... :D
@Sean: My profoundest apologies. I'm blind in one eye and can't see out of the other.
@Horo: I like the look of those closeup leaves. The background tree really looks good.
@Stuart: Nice job. Got the name of your detailer?
@TLBKlaus: That image is way neat. Appears someone dismantled several clocks.
@StuartB4 - great camera angle. Lighting is spot on.
@TLBKlaus - if those things were cogwheels, that could be a look into a very complicated machine.
Thanks very much GussNemo and Horo.
I have done some w/gears before... :D
Here's another new one, some tiling again...
@TLBKlaus: Another stunning image. I think that image is either an exquisite ceiling painting or an exquisite floor tile design. Nice work.
I used the information from David's latest tutorial on the sphere I've been experimenting with. Besides my couple of duh moments, I think that information is what I was looking for in the first place--creating glass but not having to load an HDRI file to use for reflection during rendering. It's used, but as a spherical mapped image which is used to control specular of the material for an object; hope that's a close paraphrase. :-)
Anyway, the three images below are what resulted after watching David's video and using Dave's color suggestion. Unless I can think of something else to try with this particular experiment, it's probably best to hoist anchor and sail elsewhere. Thanks to all who contributed.
@TLBKlaus - left and right like masks with eyes and a small mouth.
@GussNemo - yes, that's a clever idea of David's. I haven't had the time to experiment with it. Your renders all look very nice. My scarce experience tells me that many settings that work for one type of object don't necessarily for other types.
@Horo: the trees look every nice. I managed to download and few but have not started playing with them. Do you mean that you just put a blob in the transparency channel and change to Blend Transparency. I think TheSavage or Chohole suggested this.
@Stuart: cool cars.
@TLBKlaus: all your abstracts are lovely.
@ GussNemo: You managed to get cool results, I tried David’s tutorial too but not much success. I must have missed a step here or there.
I think Horo is correct that could be the reason why I did not get good results.
Yes, that was all I did.
It (in theory) should work well for all smoothed mesh objects and curved surfaces through reflection mapping. Not so for anything with flat sides because of the limitations of the reflection map mapping mode. Which I have belabored in other video's.