Skeinpypy allways crashing

[jet 0]

when i try to generate g-code for some models (specifically the large dinosaur skull on thingiverse) skein pypy allways crashes somwhere after the fill stage, and no g-code is created, ive even tried repairing the models in netfabb-basic, and scaling them smaller and larger, but the g-code still never completes, any help would be awesome

[Daid 306]

Yes. It's a known problem, on some models it happens 100% of the time, on some only once every 10 tries. And some models work fine.

It's caused by a bug in PyPy, and it's not related to a problem in the model. It's just some times of models that cause it, but no direct relation to what type.

https://bugs.pypy.org/issue1045

If you download pypy-c-jit-latest-win32.zip from

http://buildbot.pypy.org/nightly/trunk/

and replace the pypy folder with the contents from that zip it will be fixed, and you'll get crash free slicing. I'll soon release a new version with the fixed pypy included.

[jngf 0]

Yesterday i had a meeting with 5 Ultimakers. They were not familiar with Skeinpypy and I introduced it to them. At home I had no problems with the packaged pypy, but elswhere ther was a error for slicing. It was on the forum that a removal of the pypy folder will work, but slower with Python. That is correct.

With all installations (on fairly new laptops) the error occurred and we removed the pypy folder. I hope that error this too is a result of the bug in pypy and will be over in the next update.

In the meantime I'm enjoying the use of skeinpypy and printrun. Especially the Joris plugin for a smooth bracelet scored very good!

[Daid 306]

Great to have 5 extra users!

If you get the:

 

RPython traceback: File "jit_metainterp_optimizeopt_optimizer.c", line 9176, inget_constant_box__pypy_jit_metainterp_optimizeop File "implement_6.c", line 53453, in dispatcher_459 File "jit_metainterp_optimizeopt_optimizer.c", line 7708, in OptValue_import_fromFatal RPython error: AssertionError

 

Error, then that will be fixed in the next version, which I hope to have ready this Friday (and the error is fixable by upgrading pypy yourself)

Also, this error doesn't happen on all models. It's a bit random.

If you get the "An unknown error has occurred" windows popup message, then you are lacking SSE2, and there is nothing that can be done really. The next version will feature a better error, but that's it.