Ok here's an other 'hey, I didn't expected that'.
For the story I was just showing MOI to a friend, I was really happy to show how easy this piece of software is.
Just drawing some curves, doing an extrude, sliding (boolean merge) ... and did a 'magic fillet' ... but I wasn't able
to explain this :)



Of course I've tried different fillet size. I know filleting is an heavy task etc ... but still, it's not like I was doing a crazy one. I would even prefer to have a failed output that a wrong result :P

The construction scene is provided.

(workarounds are: use g1blend fillet or rebuild the wrong part by hand (sweep works perfectly here) ... but I still think that circular method should work too, as it works in an other package starting with an R and finishing by INO ^^)

Hi PaQ - that one is quite odd, I haven't seen that one before.

Thanks for sending in the model - I think that next week I will be working on incorporating an update to the geometry library to be released with the first v2 beta. I saw that there were quite a few updates in filleting, maybe this one will be fixed there. If not I will send it over to them as a bug report.

Thanks,
- Michael

The beta don't fix this one :S

Hi PaQ,

That is a strange one. Messing around with your model and found some interesting results.
From a fillet size 0.1 to 0.5 same as your result, 0.6 failed so I thought thats the limit, typed in
0.7 works again with the same result, up to 0.8 then at 0.9 it's happy and a nice looking fillet up to
about 1.4 then fails.

Tip: When doing a fillet and it fails, change the shape option to G1 blend, sometimes it works, anyway it did in this case.

.