search

CGCookie / retopoflow

A suite of retopology tools for Blender
2.54k stars 251 forks source link

Strange Glitch - Polystrips #237

Closed mx1001 closed 8 years ago

mx1001 commented 8 years ago

While testing the tool I experienced some strangeness at this moment in the video.

https://youtu.be/R6xiBkaeORQ?t=4m54s

This is what my screen looked like. http://i.imgur.com/OM1NDDU.png JW also pointed out the help isnt supposed to look like that. That could be something with the blf not being tied to the dpi or something. We experienced something like that too.

vxlcoder commented 8 years ago

quick question, @mx1001: what is the scale on that mesh? did you apply the scale (so scale is 1.0)?

vxlcoder commented 8 years ago

ah, i think i know the problem! there seems to be an issue when you draw a polystrip from an existing strip that connects back to itself. if you add an extra endpoint before connecting back to the existing geometry, it seems to work. i'll think through this a bit more.

mx1001 commented 8 years ago

that could be the issue. I did scale it down by 100x before beginning sculpting and retopo. I guess since I did it uniform scale didnt prove an issue during dynamic tesselation sculpting. Ill try it again with that in mind on my next video.

vxlcoder commented 8 years ago

screenshots highlighting the issue here:

works fine

sketch of what does not work

broken result

i think the stretching of incorrectly positioned strip is just a strange consequence that will be corrected with a solution, but it might be from something else.

mx1001 commented 8 years ago

ah excellent. Ill give that a try. Im sure it was something i did moreso than the tool malfunctioning

vxlcoder commented 8 years ago

confirmed. using mesh scaling does weird stuff in this scenario.

image

mx1001 commented 8 years ago

excellent. So quick to find the source of the issue. Nice! Ill keep that in mind. Awesome work! Youre a python detective haha.

On Wed, Mar 23, 2016 at 5:56 PM, Jon Denning notifications@github.com wrote:

confirmed. using mesh scaling does weird stuff in this scenario.

[image: image] https://cloud.githubusercontent.com/assets/3790699/14003420/e62d2386-f128-11e5-8a89-2306ccd1fccb.png

— You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub https://github.com/CGCookie/retopoflow/issues/237#issuecomment-200574831

vxlcoder commented 8 years ago

@mx1001, the above commit is a workaround for this issue. it's not a great solution, but it should correct any strange behavior. note: to create a loop from existing geometry (as intended in the second screenshot above), you will need to add explicitly an extra end point between the two endpoints of the strip. see below

image

let me know if this solution causes any other issues. thanks!

carter2422 commented 8 years ago

@vxlcoder this seems to be a bit erratic still. I need to test more, though. Going to tentatively put towards 1.2.1

vxlcoder commented 8 years ago

@carter2422: can you repo this?

carter2422 commented 8 years ago

Just replicated this in release/1.2.2 @vxlcoder: https://cl.ly/ghGU

carter2422 commented 8 years ago

This appears to be fixed from the changes in #276. I cannot replicate any longer, considering closed.