Developer Meeting Notes, January 15, 2012
Better and faster Booleans? HALLELUJAH! That only took, what, 10 years? :) Oh, and and a port to Android is in progress.
Ton Roosendaal writes:Hi all,
Here is a summary of topics for today’s IRC meeting.
(irc.freenode.net, #blendercoders)1) Blender 2.62 targets
- Current targets.
- Longer names for Blender data made it in svn last week!
- Bastien Montagne: Fix for posing, related to hinges or special restricted relationships. Needs testers!
- Sergey Sharybin: Carve branch (much better, much faster booleans) is ready, can go to svn this week.
- Brecht posted cycles roadmap update on code blog. Further all goes according to 2.62 plan!
2) Other projects
- Alexandr Kuznetsov started a wiki page to monitor progress for Android porting.
- Campbell mentions SDL is ready for Android, he added SDL support in our GHOST library for that reason (= low level OS, window, drawing and input handling).
- Interesting opengl-openglES wrapper.
- Luca Bonavita heard Freestyle integration (non-photorealistic render) might be near (2.63?). People try to convince Tamito to finish it up. Question is then where to put Freestyle docs in wiki.
That was it, short meeting!
-Ton-
Port to Android? I recall a while ago when people wanted a port to iOS the problem seemed to be that using Blender with a touch interface wasn’t going to be a happy experience….
I agree, it’s probably not. It might make a nice viewer though?
Touch is the primary, but not only way to use Android. It might well be your desktop OS in another year or two. The line between platforms is going away very, very fast…
bluetooth keyboard should help … also take a look at the asus transformer.
I can’t imagine working on Blender on even the best tablet but doesn’t that port also mean that it will be possible to play Blender Games ?
This is what I am wondering.
Aye. Me too.
The GPL2 is incompatible with Apples restrictions (app store and SDK license). At best it could only be available for jail-broken devices.
Exactly. It will never happen on iOS. It’s why VLC player was removed from the app store.
any ideas on an ios viewer??? :)
EDIT: also, any video/screenshots of the new boolean modifier?
Apples grip on IOS is so strong that it renders a Blender port unrealistic to impossible.
Apples grip on OSX is even stronger, and that’s what I use Blender on.
The problem is having to go through the Apple store. It conflicts with Blender’s licence.
OSX is a general computing device and, by the essence of it, much more welcoming.
Case in point: VLC player
Not sure what you’re trying to say. Apple don’t approve/reject apps on Macs (yet) to the degree that they do on their mobile devices.
They are MAC fans, don’t expect common sense from them. The reason blender is so easily considered for porting to android is because it essentially a modified Linux. IOS however is not.
iOS is very much like linux because it is a modified OSX (Which is a lot like linux), the only difference being that they have set iOS up that you can only install software through their app store, which does not allow GPL. Most of the command line tools I use on OSX work on a linux OS. The only major difference between the two is the UI and it’s dependence on Objective-C and the price of the hardware.
]#
port blenderplayer to devices, is more altruist, i think…
I think that there is some shared code between Blender and BlenderPlayer, so it would be tricky (impossible) to port only BlenderPlayer
correct. blenderplayer uses loads of blender code.
go Tamito, go !!! for the sake of Freestyle on Blender…
Long live nicest paradigm on NPR
Seconded!!!!
Indie homemade games everywhere !!!
Please make it happen, Tamito! I really would love to have Freestyle on Blender 2.63! Thanks!
Oh, and good works, folks! All of you, I just thank you so much! You’re giants among mere mortals! lol
Freestyle in 2.63? Yes please!
Yes! Freestyle!
I’ve been waiting for so long
It seems that I will have to prepare some casual game concepts with my team in case blender´s game engine will be ported to Android too.
I still have a dream that blender will one day equal enemy to the great Cryengine :o). Rendering and physics speedup and Blender will become the ultimate 3D tool >:-)
I don’t want to sound rude, but there is no way BGE could ever do half of what udk and CE do, just have a play with those, it’s so advanced… And they are developing them continuously.
I’m sorry, BGE is not intended to ever compete with those programs, and the blender foundation don’t have enough money to finance all the developers needed for that.
I mean, when you play with udk, each submodule is a big software on its own, it’s really impressive.
But I do have plenty of hope for blender as a production tool, if the development continues to follow its momentum.
The good thing about blender is : WE can make it better, for that we just need to support it, and there are plenty of ways of doing it :).
Its true that the BGE for now do not give the possibilities that those big game engine give. Though I saw many great games in development now.
But I do hope that in the future that Ton would give it some thought and that the BGE would progress enough to handle bigger projects and that small studio could have the possibility to create great indie games.
It’s a catch 22 situation. If we get someone making a great game that gets people’s attention (yes, even with all of the current limitations), then devs will get more interested, improve the code and people will make better games again. Alternatively, it could happen by a dev implementing something impressive nough to encourage someone (like the guy who did Super Blender Galaxy) to make a great game. Then the cycle picks up.
Currently there’s a malaise just because dev’s (other than a noble few) don’t code for the BGE because users aren’t creating exciting things with it (other than a noble few like Haidme, et al). No dev means no exciting new features to tempt users.
Personally I think that what the BGE needs to break the cycle is a Blender Foundation open game project. I find it so sad that the last one was so much trouble, not because of any limitation of the BGE (in fact it saved the day at the end) but just because they tried to use CrystalSpace with the BGE and CrystalSpace was just all problems and headaches. IMO, now Ton’s been scared off open game projects as he (understandably) doesn’t want to risk Blenders financial future on something that risky. :| Hopefully we can convince him otherwise for Gooseberry+1 ;) I would be curious to see what kind of budget would make it feasible for the BF and what would happen if we ran a kickstarter to pull the funds together for the Foundation….
Carve library is really appreciated. 2nd best thing went in unnoticed – OpenMP for fluid particles :)
I wish Ton and company would be a little more outgoing with what’s going on with the Development Fund. Lack of updates and the feeling of being in the loop will be the main reason I don’t intend to donate to it.
What can be found on the Dev Fund page seems clear enough to me. What do you expect ?
I had one question when they did put the Depsgraph sprint on ice. So I asked and got my answer.
Right now and up to March, the money is affected to finishing the sculpting tools and the particles.
I find that my little 10 euros/month is well placed.
I expect updates on progress on a more regular basis. I don’t question the value, I question whether they are loosing money by not engaging those who would donate. Updates a la Google’s Summer of Code being an acceptable level.
It is a reasonable question to ask whether or not that should be important; regardless I know it affects whether or not I am willing to pay for the time being. I would guess that I’m not the only one.
I understand and the ‘(link follows)’ mention seems to say that Ton, for one, agrees with you. There’s likely a reason. I believe that both are still students. Their schedules may not give them the opportunity for regular updates.
In the meanwhile know that they are more than the average GSoCer. Lukas and Nicholas are two of the best developers we’ve ever had and we have many. They are already responsible for our present particle system and our sculpting tools, respectively. There can’t be any doubt that they will deliver.
Best regards.
http://www.blender.org/blenderorg/blender-foundation/development-fund/
October 2011 – March 2012
Granted EUR 3000 to Lukas Toenne, for further work on particles and node systems in Blender. (link follows)
Granted USD 3000 to Nicholas Bishop, for applying GSoC sculpt/paint branch and other modelling related todos.(link follows).
I guess Ton doesn’t have to update it until April.
Surely they have deadlines to meet by dedicated dates, if not they should introduce them in the next round of funding because they will increase interests in the fund an may be get more people donating.
If I refer to the 2.62 current target page, it is mentioned that the cucumber branch should be implemented (Harmony) but when you go on the cucumber WIP page, the last entry date from august 2011. Is it because it is in the hands of svn code reviewers. Will it be in 2.62 ? That would be cool.
That said, I don’t think the BGE will ever equal GE such as Unity or UDK or CryEngine. I mean those commercial licences are so advanced compared to the BGE. The only way that COULD change is by doing an other OpenGame like they did with Yo Franky in 2008 or some year near it.
I could be wrong though. I’m not an expert with the BGE
As a matter of fact I’d love you to prove me wrong for I’d like to try to create one myself.
I agree. Even Ogre which has its own dedicated team probably isn’t as powerful as Unity or CryEngine in terms of game development cant compare. Unless Blender some how gets a savvy dedicated team for game development, I doubt it will ever happen. Even if there was I’d prefer resources go into the renderer (like cycles) imo
Harmony is already in trunk.
Also we do have a dedicated and savvy team working on the BGE.
Try to understand Blender for what it is.
Which is modelling and a animation tool, with texturing and rendering capabilities.
Yes it has editing capabilities and a Game engine, both are probably more of a distraction rather than a asset.
I use both of those ‘distractions’ intensively and so do a lot of people. Actually, the BGE and VSE are of use to people who barely touch the other parts of Blender if at all. Neither are a distraction since the developers who work on them wouldn’t work on other parts of Blender. That’s how an Open Source project works: devs work on what interest them. It gave us Blender as it is.
Blender won’t ever become another Maya & Co.
That’s how I understand Blender.
The possibility to get a renderfarm made of cellphones and tablets is near ! ;-)
What about improved selection (either to replace GL_SELECT or make it work fast with ATI/AMD GPUs) ?! When is it going to be in Blender?
Also, there is Darkplaces game engine. Instead of making Blender jack of all trades, just make sure Blender fits well into the pipeline with Darkplaces.
_ttp://www.youtube.com/watch?v=g1bbTCcBLU4&hd=1
I agree completely. Blender is great software but in mu opinion Blender Foundation guys are pushing it in not perfectly suited direction. They want to have all-purpose product – that is fine but the way to accomplish that – is not clear and I would say – heavily influenced by incidents instead of solid forward thinking.
I personally miss the strategic vision on how the blender will be stabilized and developed in the future. Almost everyone I know who works in 3D business considers blender as a handy tool, which is missing a few features which if covered – would fit blender nicely into the production pipelines. Making blender fitting in the production pipelines – that helps to get the user base. Not amateurs but professionals. And that is something worth considering in my opinion.
Game engines is a good example – making blender better tool for people working with UDK, CryEngine, Darkplaces, Unity etc. that would be quite an achievement and the profitable one. Profitable in terms of increasing the user base. Come on – nobody will consider blender as a platform to develop professional game. And for making amateur ones – isn’t enough what is there already?
People understand blender can be used as an asset creation tool, not a platform or engine by itself. It has possibilities, but not from the real business perspective. Try asking yourself – you want blender as a better tool to develop games or as a better game-engine nobody is using but still with poor tools?I’m sure we can find many examples and we can start a huge discussion which way the further development of Blender should go – but that’s still in blender foundation guys hands…
Hm… I have to agree with you!
But, I think that the blender foundation IS actually focusing on the rights things at the moment. The huge wave of merges is bringing a whole new world to blender.
And there are not serious projects to enhance the game engine but there are for the modeling pipeline and VFX pipepline which is great.
Cycles + Bmesh + freestyle + camera tracking/ object tracking + ocean simulation + GSOC merges ….
These are not little things.
However I would like to know more about the long term plans (after 2.6?)
Exactly, attracting VFX guys is a good target like attracting any other professional market segment. And I miss the big picture plan for next 2-3 years too…
How about a working build that will work with Intel HD video cards..Priorities seem very tilted.
Intel needs to write better OpenGL video drivers, that’s all ;) Anyhow, have you tried running Blender on Windows and Linux with Intel HD ?
Yea, It does run with Ubuntu, but booting back and forth is not really an option. Just wish they would do a build for this video. Want to buy a Toshiba laptop? I7,6g win7, 17″Lcd, harmon/kardon. Oh,,, and an Intel HD 3000. = [
I think T.K. has very good judgement for one thing. Secondly, I think edge intersection lines need to be implemented before merge. Regardless, I’ll be happy when I can have all of Freestyle’s goodness in every blender download. If it were not for Freestyle, i wouldn’t have started using Blender until about now. Thanks to Freestyle I was able to switch to blender a year earlier to make Clocked In. Very grateful to that dude.
Hello, I’ve been using blender 2.59 carve branch for a while to test carve booleans operations and found that in some cases, just doing simple union operations it crashes (http://www.youtube.com/watch?v=BFLqK7aZjzg&feature=youtu.be). Maybe it is interesant to try this kind of operations before publishing the next blender release. I’ll be pleased to try them if I have access to the source code.
2.59 is quite old. You need an updated 2.61 version with carve. Graphicall have some for windows. Don’t know for other OS.
Thanks Stargeizer, I can wait to next blender official release to test the carve booleans!
I’ve build the 2.61.2 (r43464) version and got similar error (“Can’t execute boolean operatio”). The example is very simple, try substracting a cylinder with radius 0.4 and heigth 2 to another radius 0.5 heigth 2 cylinder. Then substract the result of this operation to a 2.6 x 2.6 x 2.6 box.
I still don’t understand why improved selection code isn’t making it into 2.62 o_O
It’s the same dev who also just finished improving the UV tools. He had to set priorities. Also the selection problem isn’t trivial and there are still problems to solve.
aye. thanks for the explanation!
Freestyle ^^ integration. As a fan of NPR render this a very good news for me. How many months before 2.63.
Yaw!! Go! Go! Tamito ^^!!!