Call to arms: A rock-solid 2.49 while waiting for 2.5

Although I am as excited as anybody about 2.5, I don’t get the impression that it’s going to be available for professional work for some time.
In the meantime, I really hunger for a blender that just does the job, without the bugs. Feature creep seems to always be ahead of bug-fixes. I run into frequent ‘known bugs’ that seem to propagate forward without ever being dealt with. (personal peeve: broken wind forces on softbodies, cloth, and particles.)

So, I would like to propose an initiative to send the ‘old’ blender out with a bang. 2.5x will be so chock full of new features we will be glutted for years, but it will also open the doors to lots of mad developments, not exactly an ideal environment for stability. I need new features far less than I need a Blender that doesn’t seem to bomb on me once per session.

Instead, how about an initiative to push 2.49 development into an ultra-stable mode, well beyond the basic one bug-update version that is the norm. Get rid of the glitches, the work-arounds, anything that can be fixed without the deep recoding addressed by 2.5.

Let’s polish this baby to an ultra-shine before it goes into the museum!

Feel free to submit your patches to fix bugs in the current trunk for review.

And then 2.5 comes around and no one will be using it, and all those things will have to be coded again.:rolleyes:

I especially want the physics engine to have less bugs, as well as the particles. The kind of bugs where you constantly have to change a setting back and forth again to make it update, or where collision just doesn’t work for some reason and can’t be made to until you redo your project. I hate those bugs…

shadowphile,
It would be great if 2.49a was released as a finished project.
I do think the devs are working towards this.
Also there will still be a few patched builds on graphicall.
2.49a will provide a testing ground for some devs for a while yet.

Kind of makes me wonder what the plan is to turn the 2.5 branch into trunk.

If they merge it into the already existing trunk then all the bug fixes will not be a waste but if they just one day say that 2.5 is the new trunk then they will have to be ported over.

Been pondering over this for a while since there are quite a few bug fixes going into trunk that don’t also get committed into the 2.5 branch. Makes me suspect that once 2.5 becomes the development branch it will be merged with trunk so it gets all the stuff like the BGE enhancements and other assorted things being worked on.

its weird that so many people have bugs, but I haven’t had anything buggy since about 2.34 when blender would delete my meshes.

then again i build my own blender weekly and the devs work on bugs constantly.

you will never be 100% bug free i have bugs in photoshop, flash, sony vegas, open office…

maybe, and i may be wrong, they are working toward getting everything ported back into 2.5 and banging out the bugs as they intergrate before they release 2.5.

2.49a will be released in a day or 3, many bugs fixed. Will keep 2.5 in a branch for many more months.

I expect to see a 2.49b,c,d. especially since this will be the last release with the existing scripting api and python2.x, there are quite a few 2.49 patches postponed until after 2.49a.

Excellent scoop ideasman… I can’t wait for this to be released. For the record, since 2.48 I have only rendered one scene, I need more blendertime these days…

Is volumetrics going to be in 2.5?

I doubt it, but maybe I’ll be surprised! I’m guessing like 2.54 or something (after Durian).

Volumetrics are supposed to be used for Durian and there’s a big “volume” button in the material buttons for 2.5.

What do you really think?

Hum… cloth, and particle force fields were unified quite a while ago by the cloth coder.
Because he did not want to touch the soft body code he left a tiny note in the commit log, which did not really reach the soft body coder. So it happened that 2.48 came with 1000 times too strong soft body wind. This glitch was fixed for 2.49 and wind works as it should for all three systems, well at least it did when i checked the last time.
So could you please explain what is “broken wind forces” ?
And chances to get a bug fixed raise drastically if the bug is reported.

argh! It’s hard to keep up with Blender. This has been such a diverse response! My call-to-arms turned out more like a hornet’s nest :slight_smile:
I guess in this beta-testing world the idea of a fully-debugged software is a thing of the past because who can afford to sit still anymore?

Overall, I guess I feel like I should, er, apologize (?) for stirring things up, although it did stimulate some interesting if chaotic discussion.
I have discovered for myself that the broken wind forces in 2.48 were fixed in 2.49, even though I kept trying it out in many of the post-2.48/pre-2.49 builds. In the past I tried to chase this down in the bug tracker and elsewhere but had no luck. Why no ability to search for keywords in the bug-tracker? (Or at least I couldn’t figure it out.)

This thread however did raise an interesting question. Just how much code is getting ported from 2.49 to 2.5? One would hope that all of it is and the game engine will work just as good if not better in 2.5. Of course a more realistic scenario is that 2.5 will have all the code but is going to be buggy as hell up until 2.51 fixes a lot of issues with it.

The worst case scenario is that not all the code is getting ported over and half the features in 2.49, like the game engine, will not be in 2.5. To top it all off 2.5 will still be buggy as all hell.

FOSS Development is a vary scary place.

I’m not sure this scenario is any different than any other development model honestly. As far as I know when SoftImage rewrote their flagship product it took them a long while before they got it functional to the point where they could stop shipping Softimage3D to customers who bought XSI. I don’t even know if you could fault them for this seeing as how it was a ground-up rewrite. I’m going to be very suprised if Lightwave CORE doesn’t ship with the same sorts of issues.

That being said I think its important to have some faith in the Blender development process. Theres a lot more right with it then there is wrong.

Cheers,
Briggs

How is it scary? Certainly not more than Lightwave’s development, where they have an actual company hinging on it.

Campbell fixed a bug a few months ago, and told me that even though it was for 2.48, it would have been an issue with 2.5 as well. Besides, even if all code was undergoing a rewrite, I’m willing to bet 2.49 will be used for a while after 2.5 is out.

It’s probably not that big of an issue. I was just hoping that it was not as hard to port all the code over to 2.5 from 2.49.

I concur. I’m in no rush to even touch 2.5 at this point, anyway. Not till Durian is completed. That will be about a year later. I also hope that the dev team won’t rush things. Make Blender 2.5 (or maybe 3) as industrially strong as possible. Well, after successfully completing 2 film projects, the main players should be more than capable to produce the next one in flying colors and in parallel development with a 3d app known as Blender. It will take time, of course.

So I’m focusing on the current Blender version (2.49). And I’m actually waiting for the release of these interesting video tutorials. See if I could extend my current Blender know how and stick with it.

You’re not being fair. Look at what has been reached in four months. Think about what could be reached in the four months to come (until Durian starts?).

I don’t think it will be a year until 2.5 can be used for more than just play stuff.