This is just a teaser, I will model this up to the level that I can use it to determine what colors I want for my car, just got my hands on a '69 model 2a that needs some love, so it will be a project for the years to come.
The model is/will be modeled around a simple 3D scan I did with scaniverse, so a different approach from the blue-prints I have used on earlier builds.
Modelling in R8. I do hope R9 will get some modelling UI love for SubD though. Doing simple stuff just isn’t fast enough in R7 and R8 and there are some bugs where if I delete some vertices odd stuff on other parts get selected after delete is hidden, and not having hide/isolate/show and simple stuff like scale1D available when in the SubD tab (I hate tabs) just makes it really slow to model.
I’ll use this thread to post some frustration, ideas and wishes as I go along.
I find it more pleasant and easier on the eye both in smooth and and boxy mode.
Why it looks darker in boxy mode I don’t know. @jeff do you know? Does it have to do with the number of polygons or how Rhino calculates the position of the isocurves when smoothed?
Great, and just for the record, I prefer the smoothed version if I can choose
It comes a cross as if the AA is better, maybe the lines are thinner or something, I don’t know. Looking forward to hear what you find.
The easiest way for me to look at these types of things is to provide a model and a display mode, that I can just bring into Rhino and then follow a series of steps defined by you.
Having to recreate specific scenarios can get (and is) very time consuming just to see what you’re seeing…and that’s assuming I’m interpreting things correctly and getting things 100% correct.
That might sound petty, but these things add up… Something I can stick in a bug report that I can easily pull up tomorrow or even months from now and still reproduce the problem without having to go back and re-read some post somewhere, is extremely helpful.
Ok, can you shows in more detail how does this setting work?
Here’s my current setup that I use to get a very similar result, but it works on the “display mode” level, I’m confused about how to set it on “the layer level”(?)
My setting is a derived from the Shaded mode, so no total material override or “rendered” environment lighting, just the default lighting from the Shaded display mode and some tweaks on top of the layer colors
@Holo I think the behavior you are seeing with the ‘white’ isocurves is that they are getting mixed with the background rather than with the material. You can clearly see this when you change the background to a different color. I suspect this is a limitation of working with two different transparencies, the ones from the isocurves and the material transparency.
I am thinking of 2 workflows - not sure if much more elegant:
(A)
use Gumball Extrude with snappy Gumball twice,
then _stitch.
(for the stitching you might need to randomly / “dirty”-move one edge first, to be able to do a correct selection)
(B)
_dupEdge (one of the 2 multi-segment-Edges)
_subDSweep1 (with Chain option, and “crease”)
_removeCreases
_join
Yeah, I know (all to well ), sorry.
Problem is that I type this stuff quickly down after a session I didn’t have time to do in the first place, so I don’t have time to add all these reproducable steps, nor provide the files. Busy days trying to fit in everything. So then I prioritize to quickly type it down as I do as I figure it is better that I send the message while I remember it than put it off for tomorrow that never comes
Here is a slide for you to put on youtrack:
Rendered mode: Turn on show isocurves (and or edges if you like)
Make a layer black and at 50% + add a material to it so it isn’t white and put something on it
Make another layer and add a glass material or a transparent something and put something on that layer so it triggers the display pipeline.
Small update on the modelling.
Lot’s of detailing to do, but the main shape is up and running and I can play with colors.
(Which is the main reason for the modelling, the LR needs a new coat and now I can start figuring out what the different colors does to the appearance of the car.
(I found the tires online and modified them to match the rims. The rims where modeled in Nurbs since it is just easier and faster. The rest is SubD)
There are some quirky stuff going on when deleting faces on subd’s. Some times when I delete internal edges vertices that should not be removed get’s deleted too. I’ll try to add a post on that one day.
And when bridging I wish that Rhino would not let me pick surfaces if I have started to pick edges. That seems like a bug to me.
This isn’t really a challenging car to model, and would probably be just as fast and easy to model in nurbs, but there are some situations where SubD just shines, and that is when I need to adjust complex parts, as everything is stiched together so I can easily manipulate parts of multiple objects at the same time. (This works since I try to use a clean and coherent patch layout) For sketching this is just great and much faster than working with nurbs.
Here is the model in Arctic mode, unsoftened with wires showing:
(The windows look strange since I just patched the holes and extracted the result, but they are planar so it doesn’t matter if the layout is strange and triangulated at some places)
And this is what it looks like when I model. The rendered mode resembles the raytraced mode quite nicely. The car is lit by two large planar lights and an hdri.
