I dislike people not supporting Joedh with his BMESH branch and commiting code into parts he wanted to work on, only to complicate things for him. Is there any real statement from the core developers about the integration of BMESH. Are they interested in an integration anyway?
I heavily disagree with you on this.
-
if you don’t want others to commit code, dont make it opensource.
-
he is half-way there with his donation to move to another state, for his health reasons. And hopefully will be able to continue developing on BMesh so, that’s quite a support $1000.
I dont really follow the BMesh development , excepts on youtube videos. But isn’t there a set common roadmap for this ?
http://wiki.blender.org/index.php/Dev:Source/Blender/Modeling/Bmesh
just make it -_-* ! doesn’t matter whom does it, as long as it gets done.
Umm. Have you ever heard of svn? I’ve been checking out 2.5 since it was just purple and green boxes. In the last month or two it has become fairly usable in certain aspects and I check out a new revision and build everyday if possible. So yeah.
I dislike people who complain about stuff they know nothing about.
Martin
I’m sure they’re open to suggestions for how to solve this. Perhaps you could come up with a proposal?
But given that the context is a totally different section (just as all the other panels are) I’m not sure how this is really solvable? To me this is more illustrative of why the horizontal layout was very limiting in 2.4x to begin with, and is clear indication of why it makes sense to make the vertical layout the default. They’re not sabotaging the horizontal layout. It just has inherent limitations with regard to laying out the kind of information that the buttons panel needs. It was even already showing this limitations in 2.4x.
See the video of William’s talk at the 2008 BConf for more on this.
It is true that I think in terms of organization 2.5 is an improvement.
That I do not even question.
My only point was - and I guess I was not clear - to the current state
visually the new button system is less easy to read than the old system.
Question is what happens when people start to theme it. And it is maybe
the current gray in gray tone theme that makes it difficult for elements to
stand out.
I agree with you entirely. 2.5 feels so much better that going back to 2.4 is difficult because you notice where everything works so much better in the new version. The new sliders are great. Same with the color picker, menus and the whole interface theme.
The preferences are in the outliner, that’s a good change and it seems that the material/object block linkage has been removed. 
Not being able to create materials outside of an object selection context maybe not so good though.
I’m not keen on the panel dragging either. It seems more natural that clicking the background of a vertical panel would allow you to move it up and down without using middle-mouse button. This is important for touch-screen displays. I know alt-click works ok and avoids mistakingly hitting buttons but people are more likely to drag a panel than reorder them so reordering can perhaps be behind the modifier key.
Which is how it should be as it’s a better way to work. The proposals have clearly explained in detail why vertical panels work better for both developers and users.
Naturally there will be some clutter in the transition because buttons were ordered to work with the limited horizontal panel although I can’t really see a problem.
The big problem with horizontal panels is that to maximize the use of the space, you have to set a fixed height and this enforces unnatural grouping of settings. If you run out of space, you can only go sideways because wrapping vertically introduces another dimension of movement so it’s much harder to get to the panel you want.
Horizontal scrolling is not a good thing because not everyone has scroll wheels that scroll sideways fluidly and there is no ‘page sideways’ on keyboards, it’s page up and page down.
Vertical panels maximize the use of space, adapt to varying amounts of settings (constraints, rigging etc) and work well with input devices. People need to adapt for the greater good.
How many websites do you see that require you to scroll left to right? Imagine if Gizmodo made you scroll sideways to see each varying height article. It just doesn’t work.
There are so many good changes in this release, people need to stop getting hung up on the vertical layout.
This release is going to be amazing. Shame it might not be ready for a while but it’s probably best to stress-test it on Durian.
I hope that in future they allow to create just materials without a linked object.
And in addition also supply Blender with what is a standard in all other systems
A material library including a browser like for example in C4D.
The current way of saving the materials on objects in a scene as a replacement
is not really usable and I did not see any of the students being convinced about
its usefulness.
Sorry, for this. Reading my own post it sounds like I would want to offend someone. I only wondered why Brecht commited code to the mesh structure after Joedh requested not to do so.
I know this is open source, but without coordination, development can not work. I was a software developer for the last 3 years and I know what a problem it is to merge from a branch where someone altered files you were working on, too.
gdp2000, Yeah you gave the impression of that.
I can honestly say, NONE of the changes in 2.5 I don’t like. The inter-facelift is 100% all throu needed and in my most sincere comment… blender.25 is just after Softimage in GUI but ahead of Max and Maya in my opinion.
The event based system , and new MVC (clone) design pattern will make 2.5 take a GIANT LEAP in development aspects.
If I script in py2.5 or is it 3.0 , will those script work in blender2.5 ? what goes from now on. There’s a version change that’s not backwardscompatible with blender2.5
I feel really sorry for this! I hope the devs will forgive me as they are doing a great work!
OK, but there is really one thing I dislike. It’s the header windows. There was some talk somewhere about refactoring them. I don’t know if it’s still the plan. For example the timeline window could have a header integrated with the timeline itself. So for example the controls would be on the left side and the timeline on the right. This would save a lot of pixels vertically.
This makes for a pretty wide header though, which would have to get cut off on smaller views. I think there will have to be some work on the headers. For example, the main menu bar panel is empty as the user prefs moved to the outliner.
I personally never saw a need for a separate timeline - it could even be integrated with the 3D view because you generally only ever need to control an animation that you watch. The IPO window has it’s own controls as does the sequencer. The new sequencer control is actually really nice and not intrusive space-wise if you have multiple 3D panels.
Plus, given that you generally need to scrub through animation anyway, it removes the need to have a split pane to do this.
One thing that worries me in the recent build is that alt-left-mouse button doesn’t work. I found the viewport controls to be awkward before as they weren’t consistent. It was alt-LMB to rotate and shift-MMB to pan. I can’t even remember the zoom one.
In the latest build, it’s MMB to rotate, shift-MMB to pan and ctrl-MMB to zoom. I have a stiff MMB (not just because I’m excited about 2.5) so this is not very good at all to be the main form of navigation. IMO, it should be the fairly standard setup whereby alt is used with each mouse button. alt-LMB = rotate, alt-MMB = zoom and alt-RMB = pan. I think it’s more MMB to pan but scroll-wheel zooms anyway.
If it’s configurable there’s no problem of course but it seems like an extra hurdle.
I think one of the big problems with adjusting to the 2.5 interface is that all the panels in the buttons window (I believe we’re mainly talking about the buttons window) don’t have a very visible separation. While all the 2.4 builds had rounded transparent boxes surrounding each of the button categories, 2.5 just has a line and a bold title to recognize by. Before you just had to recognize the boxes and what order they are in. I also liked how some of the boxes had tabs, instead of just expanding to more text. Trying to read the words and titles is slow. In my opinion, its looking too much like Maya, and I hate Maya’s interface. It makes everything too non-graphical. Its too text-based. I like the new buttons and all, but they they really just look like they’ve been organized. I was looking for a simpler interface. I don’t really like the checkboxes either. What used to be a nice button that switched on or off is a tiny checkbox with some text next to it. It took forever to find some of the options. I’m not trying to be over critical with the new interface. And they’re still working on it, I understand. I’m just trying to point out some of the problems with it.
The interface may seem nice after you’ve used it for a while and got used to it, but its still too intimidating to beginners. The buttons and options should be able to be quickly identified, instead of spending time trying to learn a new interface. My point is really that the interface should focus on simplicity for all the new users.
One of my big complaints with 2.4x was actually that toggle buttons look exactly the same as buttons that perform an action. And that made it hard to visually scan things (for me, at least). So IMO the check-boxes are a really good thing, because they clearly identify toggle buttons as such.
That’s not to say that they couldn’t be large text buttons as well, with the check box inside. But I strongly oppose actually removing the check box itself.