I had heard of the new V7 draft analysis and I thought I’d give ita try.
It failed on the first model I tried. The model was crappy but even Rhino version 2 was able to do the draft angle analysis. Rhino2 was also able to identify the the surface with no area that was causing V7 to fail.
How many years do the Rhino developers have to fall flat on their faces before they are going to develop decent error checking? or are they going to just keep releasing software that is riddled with bugs ?
Here is a small part of the model with the bad geometry. Bad_draft_analysis.3dm (2.7 MB)
The bad geometry creates more than just no draft angle display, While I was fiddling with the new dialog settings Rhino crashed and disappeared and when I open again there was no offer to recover the geometry (in this case that does not matter).
And there are more bugs. While I was extracting surfaces by window selecting areas and joining the result I discovered that Rhino7 was duplicating some surfaces (copy option set to No). This is related to the bad geometry because it does not happen with good geometry .
Another bug is the advanced options has an option to use the old version of draft analysis but that option seems to not work
tools failing on bad input geometry would not qualify as a bug imo…
For instance, if I try to boolean a set of object with mismatched normals, is it not reasonable to expect the tool will fail?
for instance this model is duplicated on top of itself in some places 7 TIMES.
explode, sel dup, delete the dups, then join. Look at the parts that won’t join… it’s because they are still duplicated but with errors that disqualify them from being considered a duplicate. Pick through it and get it down to one complete model without the duplicated surfaces, or anything joined to itself, and the tools works exactly as designed.
Anytime rhino fails, I tend to 1st assume bad geometry, then use the many tools available to try sort through and figure out what’s wrong. More often than not The problem lies with me and not rhino… Are we perfect? nooo far from it… If you do find a bug, we’ll do our best to fix it asap.
I’m not asking that Rhino do anything with this geometry other than analysis. I’m trying to find out what needs to be done to repair this geometry. Every previous version of Rhino can do the job - Rhino 7 fails and that is a bug. All the bad geometry in this will be removed when I add draft to this model but I need to be able to first analyze the model to even get started on fixing this model. I’m not trying to fix all the horrible fillets in this model . I’m trying to figure out what needs to be done to create a new model. I will create my own fillets. What I need to know is which of the base surfaces need draft and which can be used as they are.
The point I am making is that one of the reasons that Rhino keeps regressing like this is because Rhino does not have adequate error checking. Without good error checking the development process is doomed to be forever groping the dark. Its a bug that Rhino 2 can identify the one bad surface that is causing Rhino7 to fall on its face.
That is a bug I have been complaining about for 16 years and amazingly McNeel sycophants are still trying to argue that Rhino falling on its face when it encounters degenerate geometry is expected behavior.
In the model I sent if the first thing you do is explode and seldup Rhino reports no duplicates. However if you extract parts of the model by window selecting and and then join the result and repeat that several times and then explode the entire model and run seldup you will find duplicates. In other words, Rhino7 adds the duplicated surfaces. That is another bug that I reported in my first post. How is the user going to fix a model with bad geometry if Rhino keeps adding more errors as one proceeds?