Posts filed under ‘Uncategorized’
So, I’ve managed to put in some BMesh time. Recently I ported a bunch of selection tools, select linked flat faces, select random, select sharp edges (as in edges with folded faces around them, not edgesplit-sharp edges).
I also ported select mirror to bmesh, and in the process I decided to try my hand at writing a topological-based function to find the best mirror candidate for a vert. It works pretty well, except I’ve later learned that Campbell Barton beat me to it in svn trunk. I’ll have to take a look at the code, see how it compares with my method.
A good mirror-finding function is important. In a perfect world, the function would be entirely topological (e.g. no need to have your model in a t pose, perfectly split by the x axis), but I’m not sure if I’ll be able to do that anytime soon. What I’ve done is coded it to detect the best match among possible candidate vertices in a sphere around the vert’s mirrored location, based on how similar the topology around the candidate is to the topology around the source vert.
I’ll have to put up some illustrations later. I’ve not yet hooked this up to x-axis mirroring, weight paint, etc; I need to look at Campbell’s code, and decide if I should modify my version to use the same method. Mine isn’t completely perfect, and I suspect his approach would work better in certain situations. Most likely I’ll end up combining the two. This isn’t something I’ll likely be spending a lot of time on in the near future, though, since there’s lots of porting work and debugging to do.
I’ve also been chugging away at aforesaid porting work, and merging in recent changes from the main development trunk. I’ve decided to put off the dreaded loop to region tool (ctrl-e, for example, loop select a forearm and loop to region will select the hand) until later, since it seems few people use it and it’d be more effective to spend time on other parts of the code right now.
The last three weeks have been insane for me, but I’ve managed to make some progress. I recoded the join triangles tool, though I’ve not yet tested the more advanced uv/vcol features. I also finished getting all the primitives to work.
The next big thing will be the loop to region tool, and after that it should mostly be easy monkey work porting small tools, merging recent changes from trunk, restructuring extrude to match the way it works in trunk, fixing bugs, etc.
Unfortunately I’m not going to hit my self-administered deadline of having something complete and stable by the end of this month, due to health problems and homework (and fixing bugs in the main 2.5 development trunk so I could use it for my homework) crowding out my time. But I shall get there in the end, no amount of adverse circumstances shall defeat me.
When I originally got the updated BMesh API from Geoffry Bantle, we both knew it was going to be a lot of work to integrate it. We had tried and failed a from-scratch rewrite (that started by removing much of the old tools code) in the first BMesh project, so when I started again I decided to keep unported mesh code working via a compatibility layer, until it came time to rewrite it. This worked very well in the beginning of the project, as I could write tools and improve the BMesh API. More then that, I was able to rewrite or refactor a large portion of the mesh code one piece at a time, and still have everything mostly working.
Unfortunately, as more and more tools and code got ported to BMesh, the compatibility layer got more and more unstable and buggy. From a coding point of view, it made the code more confusing and muddled the underlying design. I finally realized I was going to have to remove it entirely, and port the remaining tools to the new system.
I have now started this process. I’ve removed the compatibility layer and all the old code it supported, and I’ve started porting it to BMesh. Currently join triangles, a bunch of selection tools, and some of the primitives code (cone/grid/circle/tube) remain to be ported. I hope to have all of this done within the next two weeks; the hardest will be join triangles and region to loop, the rest is only one or two days of work.
BMesh should be much more stable after this, as no longer will things randomly break or tools unpredictably mess up your mesh. After that, all I have to do is fix the remaining bugs and BMesh will be ready for a merge into trunk (yay!).
I’m in the process of porting over the new sculpt/multires tools to bmesh. It’s not working quite yet, but I don’t expect it to take too long, brecht did a good job making the new code easily extensible to ngons.
I’m excited, because sculpt/multires was the biggest thing blocking inclusion of bmesh in trunk.
Someone recently reminded me about this in email. Originally I planned to give donators credit on this site. My plan was to launch this along with a system of incentives, whereby some random subset of donators would receive free bmesh t-shirts. I wanted to do this all at once, but I ran into problems with the tshirt manufacturer I chose that I’ve not yet resolved.
In hindsight, I suppose I got too obsessed with the gimmicky aspect of my plan and forgot about the important part. This is something I intend to rectify. Look for an official donators credit list in the next day or two.
I’ve decided not to stop working on bmesh. I’ve done some more work on it; fixed a few bugs, merged recent changes from trunk into the branch, and I’m tackling edge slide (which is turning out to be quite hard to do correctly, but should work great when I’m done).
I’m planning to have bmesh in trunk by march of next year (or the third 2.5 release, whichever comes first). This is a more then generous timeline, imho, and is achievable for me. I’ve also had several new developers show interest in the project, and of course Wael has been very helpful.
I’m committed to finishing this project; I love modeling, and I can’t stand the old toolset (after using decent modeling tools in an ngon modeler, it’s hard to go back). I don’t want to see this part of blender stuck in the 90s anymore.
I may have to take a hiatus from BMesh for a while. The donations haven’t really been enough lately, and unfortunately I can’t make enough money working on both BMesh and a real job, since at the moment I can only spend a very small part of my time doing either. I was supposed to start a medication today that may have allowed me to be more productive, but there was a glitch and it looks like I’m going to have to wait a bit longer.
I certainly will continue working on BMesh if people donate more, but I don’t want to get a reputation for constantly asking for money or the world will end. Either way, once I start this medication there’s a good chance I’ll be much more productive and will be able to spend time on it again. The truth is that I need at least $1000 USD a month to maintain my current level of commitment to BMesh, and if that isn’t possible I’m going to have to work exclusively on other projects to make more money, at least until I start that medication.
BMesh is now scheduled to be included in 2.5 Beta 2, the third 2.5 release. This is an exciting time for this project; I’m very happy with it’s success, which is largely due to donations and the time they’ve allowed me to devote to working on BMesh.
On the topic of donations, I’ve put up another progress bar which represents the current level of donations (which is rather low at the moment).
Don’t forget to donate! Even small donations help. This project is funded by the community, which allows me to devote much more time to it then I would otherwise.
BMesh is now in the alpha testing stage. It’s reached the point where it needs user testing to further progress and identify bugs.
Please use a build no older then November 4th.
Dissolve is xkey->dissolve, and vertex connect is ykey. Those two are used a lot, so it’s a good idea to familiarize yourself with them. Subsurf now works with ngons, which allows for all sorts of cool modeling tricks (and to get rid of the ngons you simply apply the first level of subsurf).
I’ve also made a start of feature notes on blender.org.
Blender will now save recovery files on crashing. So if blender crashes, look for a [filename].crash.blend file. Note that the recovery file will not include changes made in editmode, so I’d suggest cycling through editmode/objectmode every once in a while.
- GLSL draw mode doesn’t work.
- Edge Slide doesn’t work.
- Particles are buggy.
- MultiRes hasn’t been brought back yet.
- Sculpting may not work.
- The vkey rip tool isn’t quite finished yet.
Not everything works yet. Some parts need attention by their maintainers, e.g. particles and MultiRes, and are things I shouldn’t really be dealing with (meddling in ) myself. And I’m sure there’s a great many bugs I’ve not uncovered yet myself.
On the other hand, things are fairly functional. Most tools have been ported over. Only a few are missing.
Please remember to donate, even small contributions help. Donations are what allow me to spend as much time on bmesh as I do.