Show Us Your Bryce Renders! Part 2
This discussion has been closed.
Daz 3D is part of 
Connect
DAZ Productions, Inc.
7533 S Center View Ct #4664
West Jordan, UT 84084
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2024 Daz Productions Inc. All Rights Reserved.
Comments
What about boolean modeling? Haven't you ever seen the Bryce Master series The Great Hall where the entire scene is built out of booleans made with primitives?
Do you really consider that modelling?
In the end you can make a shape that hast the volume of all shapes involved and its even possible that you cant put the final shape properly on the ground because when pushing the "Land objects" button it follows the bounding box and not the object. The final object can be smaller than the bounding box.
Based on this definition from wikipedia "In 3D computer graphics, 3D modeling is the process of developing a mathematical representation of any three-dimensional surface of object (either inanimate or living) via specialized software. The product is called a 3D model. It can be displayed as a two-dimensional image through a process called 3D rendering or used in a computer simulation of physical phenomena. The model can also be physically created using 3D printing devices." Yes I consider it 3D modeling. There is nothing I've seen that says the model has to be able to be placed on the ground at the push of a single button to qualify it as a 3D model.
To me all it has to do is look like the object it's a model of, in 3 dimensions. Placement on the ground can be achieved by other means and so what if it takes slightly more then the click of a single button. Now is it a model in the same way as a model from say 3D Max, Lightwave, Maya, ZBrush, etc. probably not but it's a hell of a lot easier to pick up how to do it and create something worthwhile then it is with any of those other programs. Provided of course you can visualize the result of the boolean operation you plan on performing.
Still tinkering... here I have tired to use curvature as a substitute for SSS - hence the "faceted" appearance of the model. Curvature suffers most from Bryce's mesh smoothing issues. Don't know if this would be a viable approach if the mesh was smoother... or not. Just an experiment really.
Here's a way around that problem (if it is indeed a problem and I suppose it would be to some people)
Using the example of my barrels from earlier:
First Picture
Step 1: Make your shape out of primitives and set the positive and negative attributes to get the shape you need and make a group.
Step 2: With the group selected, go to 'Objects' in the top menu and select; 'Convert Boolean to Mesh'.
Step 3: With the newly made mesh selected, go to File/Export Object and export your mesh as a Wavefront Obj file.
Step 4: Go to File/Import Object and navigate to where you saved your mesh and re-import it back into the document.
You will now have a perfect mesh who's bounding box will not exceed it's actual volume.
Second Picture
Render of the screen grab showing the finished barrel sat on the ground plane as well as the other two stages of construction.
Hope this helps.
Here's a way around that problem (if it is indeed a problem and I suppose it would be to some people)
Using the example of my barrels from earlier:
First Picture
Step 1: Make your shape out of primitives and set the positive and negative attributes to get the shape you need and make a group.
Step 2: With the group selected, go to 'Objects' in the top menu and select; 'Convert Boolean to Mesh'.
Step 3: With the newly made mesh selected, go to File/Export Object and export your mesh as a Wavefront Obj file.
Step 4: Go to File/Import Object and navigate to where you saved your mesh and re-import it back into the document.
You will now have a perfect mesh who's bounding box will not exceed it's actual volume.
Second Picture
Render of the screen grab showing the finished barrel sat on the ground plane as well as the other two stages of construction.
Hope this helps.
wow thanks I learned something new about bryce.. fun fun fun
Maybe it would be an interesting feature to modify the function of the Convert button so that Boolean shapes are immediately simplified so that the user doesnt have to go through the export and reimport steps.
All this talk of modelling inspired me to give an idea I've had rattling around in the back of my head a trial.
Turned out to be a bit trickier than I imagined. I thought a pocket knife would be fairly simple, but it turned out to be a bit of a challenge. On the plus side, I learned more of Wings 3D in the process.
I can see where I can improve my design. The gimping on the back of the blade could do with being deeper and also some corresponding gimping on the handle would be nice. The back lock is a bit high. The finger choil could do with being larger and the flipper is not quite long enough to reach through the back. The hole at the back is actually a bit of a hazard I see now, since you could put a finger through that while inadvertently closing the blade - what would be bad.
The blade is supposed to be bead blasted black - perhaps a bit too shiny.
The handle is intended to be G10 but it looks more like FRN.
The metal inlay is more a fancy of my own invention - would probably make this a rather expensive proposition for a pocket knife.
Comments welcome on either the design or the render. I don't mind.
Horo's HDRI Treppenhalle 2 used for backdrop and lighting via obscure light method through TA gels.
BTW: I just found that if you create boolean shape, export it to [NAME].obj and re-import it then you dont have a single shape but a group of fragment which can be ungrouped and torn apart.
Wow David - you and Savage are doing some of the best photo-realistic images I've seen in Bryce...
Well it only takes seconds to do, so it's not like it adds loads of time to the process and in my opinion, is still easier and quicker than trying to extrude a shape in Carrara or the other modelling applications that I've tried out in the past. But I am so familiar with the Bryce interface that I will be biased about it's ease of use. :)
I'm convinced... after I'd got my head around the fact that it's the straight edge that's sharp. I was originally looking at it as if the curved edge was the sharp bit... which was confusing.
Materials and lighting look great though.
Well it only takes seconds to do, so it's not like it adds loads of time to the process and in my opinion, is still easier and quicker than trying to extrude a shape in Carrara or the other modelling applications that I've tried out in the past. But I am so familiar with the Bryce interface that I will be biased about it's ease of use. :)Actually it adds loads to rendering time.
7 spheres render faster than one sphere minus 6 spheres.
It will be interesting to see the wireframe of your scene :)
Thanks Dave for the explanation, much appreciated.
I like the knives David, they look like someone retrieved them from the bottom of the ocean, sunken treasure.
No it doesn't have to look realistic, but if I could get it 100% realistic that would satisfy one of my aims with this scene.
On the topic of converting, exporting and re-importing, if you want to experiment further and you don't already know this, the wireframe mesh resolution is involved in the conversion calculation (image 1). I'm not sure which it is... vague memories suggest it is selected - but it would be worth testing all of them to be certain... being selected makes sense though, since the object has to be selected before it can be converted.
Dan, thank you! I keep trying to push things a little bit further - but for skin, the lack of inbuilt SSS is proving a real stumbling block.
Dave, yes, well, I'm not really into "stabby" knives. I wanted to design something practical and also - easy to sharpen - which I thought the straight blade might be good for. For work I have a variety of folding locking knives (which stay in my tool box unless being used), but it is difficult to get ones that haven't been designed to have some kind of "tactical" element - the result of which is that I end up stabbing myself by accident. Because it is useful for when scrambling through roof spaces, I have a UK legal SOG slip knife (with a tanto blade - again a pointlessly stabby shape - but the blade is about 1 1/4 inches long so inoffensive) and a Blackhawk safety cutter - which handily incorporates a broad flat screwdriver just the right size for turning off isolation valves.
And I realised too, I missed out provision for a thumb stud for one handed opening. Which is very useful when you have your hands full! I'd also put one on each side to make it ambidextrous. And likewise a back lock over a frame lock or liner lock, because not only is it stronger, it's also ambidextrous.
mermaid010, oh, they were supposed to be new. I have some corroded metals I could use... but I'd have to eat into the blade to do it properly.
Thanks for the feedback!
No it doesn't have to look realistic, but if I could get it 100% realistic that would satisfy one of my aims with this scene.
Thanks for the feedback!Have you ever watched Babylon 5?
Even in the Space station itself you see the actually low resolution of the meshes and look at the materials, and people found it great.
It looks much better what we can do now
Yes, the way you were doing it.... however, If you make it into a mesh, it is no longer one sphere minus 6 spheres. It is one single mesh... The same as you would get with any other modelling application such as Carrara, Blender or Hexagon.
7 spheres render faster than one sphere minus 6 spheres.
converting the booleans to a mesh then doing the export, re-import does not appear to render slower than the boolean group. times pretty much seems the same to me. one there are a LOT of boolean groups then things seem to slow down, converting them to then doing the .obj export/import would speed it back up, as once its a mesh, the engine doesn't have to do boolean calculations anymore.
7 spheres render faster than one sphere minus 6 spheres.
I think he meant it only takes seconds to export it and then reimport it as an .obj
Maybe the programmers found it to be a real stumbling block too and that's why it's not inbuilt?
ok boolean to mesh render time test 2
ok with render option set to Regular, render time is 10 seconds
with render option set to Premium, 16RPP Ray Depth 6 I get this
Seems to me that just tests the difference between render settings and the fact that booleans are the object being rendered is incidental. Wouldn't a better test be to compare and object made from booleans in bryce to a similar object made in say blender or wings3d and then imported into bryce?
ah but those are not booleans anymore, those are the imported meshes of the booleans created in bryce,
also did not provide enough information in the previous post.. 1 of those objects renders in 8 seconds, 11 of them in ten seconds,
I'll have to redo the scene in just boolean groups and check render times,
*addendum
ok similar scene with just booleans no mesh conversion magic, premium renders time with the same settings as above 26 seconds..
clearly its time for me to shut my mouth since I need to run tests on both my computers before posting as they are affected differently by scene changes.
Okay well you didn't make that clear. The way you worded it I thought you just took the same scene consisting of booleans and changed the render settings. It seems that would skew the results though then. A fair test would be to use the same render settings but the difference being in one render booleans are used in the other an identical object made without the use of booleans. Since the premise being tested is that booleans significantly increase render times compared to a similar non boolean object.
I appreciate Rareth's tests, but I think that the mention of render times was a red herring anyway.
The original discussion was about modelling in Bryce, not rendering.
To use my barrels as an example again, from blank screen to finished scene was less than 20 minutes.
Actually making the first barrel was about half of that time (so 10 minutes to build and texture the model).
If we're going to start talking about render times, that then starts to depend on so much more than simply the model.
If a model was built in any other application and imported into Bryce as a mesh, it would take the same time to render as a mesh made in Bryce.
I appreciate Rareth's tests, but I think that the mention of render times was a red herring anyway.
The original discussion was about modelling in Bryce, not rendering.
To use my barrels as an example again, from blank screen to finished scene was less than 20 minutes.
Actually making the first barrel was about half of that time (so 10 minutes to build and texture the model).
If we're going to start talking about render times, that then starts to depend on so much more than simply the model.
If a model was built in any other application and imported into Bryce as a mesh, it would take the same time to render as a mesh made in Bryce.
I agree in theory. I just felt that what prompted the test was eireann.sg saying something to the effect of it takes longer to render 7 spheres minus 1 then it does to render the same shape made differently.
a return to bryce modeling..
Entrance to hell
A few times now I've had a look at saving the renders as .hdr files.
Here is a comparison of the file saved as .hdr and .tiff saves.
Both were then taken into Photoshop and colour curves and saturation levels altered and the .hdr was converted down to 8 bit using the hdri options in photoshop.
Top pic; HDR file
Bottom pic: TIFF file
Nice idea.