Search found 1396 matches

Go to advanced search

by realthunder
Mon Apr 03, 2017 6:23 pm
Forum: Archived Pull Requests
Topic: #619 Made TreeView stable
Replies: 67
Views: 6712

Re: #619 Made TreeView stable

Even debug build is fast. What OS are you using?
by realthunder
Mon Apr 03, 2017 6:10 pm
Forum: Archived Pull Requests
Topic: #619 Made TreeView stable
Replies: 67
Views: 6712

Re: #619 Made TreeView stable

Thanks for looking at this issue. Unfortunately your patch doesn't change anything. The load time of the STEP file is still at ~15 minutes. That's odd. You're right, it didn't make a difference, because I reverted the change and it is still fast on my machine! How did you load the file, by just ope...
by realthunder
Mon Apr 03, 2017 5:08 pm
Forum: Archived Pull Requests
Topic: #619 Made TreeView stable
Replies: 67
Views: 6712

Re: #619 Made TreeView stable

wmayer wrote:Here it is.
It is because of the cyclic check on adding object. This is the worst case I've predicated, all children first. But obviously I am wrong about the performance impact. Only 900 object, and it's that slow. I've submitted the fix.
by realthunder
Mon Apr 03, 2017 4:25 pm
Forum: Archived Pull Requests
Topic: #619 Made TreeView stable
Replies: 67
Views: 6712

Re: #619 Made TreeView stable

wmayer wrote:Today I found a major performance regression with the new tree view. In issue #1999 you will find the file FreeCadProblem.step which takes 15 minutes to load in release mode.
Mantis is down at the moment. Could you please post the step file here, or if it is too big, send it to my email?
by realthunder
Sun Apr 02, 2017 9:14 am
Forum: Archived Pull Requests
Topic: #619 Made TreeView stable
Replies: 67
Views: 6712

Re: #619 Made TreeView stable

This is not due to the presence of 2 bodies: it's the result of Body001 not being hidden as it should. Think of a Part Chamfer applied to a part: after the Chamfer, its parent (what you call "child", but IMO it's the other way around) is automatically hidden. This should be the case here I can see ...
by realthunder
Sun Apr 02, 2017 8:54 am
Forum: Archived Pull Requests
Topic: #619 Made TreeView stable
Replies: 67
Views: 6712

Re: #619 Made TreeView stable

I'm sorry but I find your reply confusing The logic you find incorrect lies in PartDesign. Before my patch, the same thing happens, that is, when you create a boolean of Body001, it is NOT removed from 'Part'. So, 'Part' is still effective displaying TWO Body001, you can try on your FC that reverte...
by realthunder
Sun Apr 02, 2017 8:41 am
Forum: Archived Pull Requests
Topic: #619 Made TreeView stable
Replies: 67
Views: 6712

Re: #619 Made TreeView stable

The file is provided in the topic I linked, but I created a new, smaller file that reproduces the issue I check the file, and the Tree View actually behaved correctly. Here is what happen, previously, in the Tree View, one object can have only one corresponding item in the Tree View, even if the ob...
by realthunder
Sun Apr 02, 2017 4:46 am
Forum: Archived Pull Requests
Topic: #619 Made TreeView stable
Replies: 67
Views: 6712

Re: #619 Made TreeView stable

NormandC wrote:Two instances of Body007 can be seen in the tree. IMO this is wrong.
Could you please attach your model file. I'll take a look
by realthunder
Fri Mar 31, 2017 7:04 pm
Forum: Path/CAM
Topic: How do I get rid of this bounding box
Replies: 17
Views: 1948

Re: How do I get rid of this bounding box

It is very handy to compare this box with the working envelope of the destination/target machine. Additionally, there is a need very often to square-off/face the chosen mill stock (especially extruded stock); thats why I pushed for the Facing Operation a way back. As requested by this thread, I've ...
by realthunder
Fri Mar 31, 2017 6:17 pm
Forum: Developers corner
Topic: Introducing App::Link/XLink
Replies: 12
Views: 1534

Re: Introducing App::Link/XLink

That's not exactly true. Draft Clone does a deep copy of the shape. In Lattice, I made array tools to do a shallow copy, to get extra performance and lower memory use. That worked brilliantly, and I got an unexpected speedup because OCC re-uses the triangulation too... to finally realize that after...

Go to advanced search