I am on Twitter too.

I will be posting more often. At least short updates via my twitter.


5 things Blender should do PART 2

This is PART 2 of my previous article:
5 things Blender should do to be succesful in the industry
Thanks for all your comments on my previous post. There were over a 100  and I read them all. I can see there are users from all levels and ages. I also perceive that most of the misunderstandings came from talking details when I talked principles and talking principles when I talked about details. Many of the replies were also fallacies of all types.

It was also rewarding to read that some users, mostly professionals, agree on some points and want mainly integration with Nuke and other apps. This leads me to conclude that this is the most important feature that must be worked upon.

Disclaimer: I don’t like the word “industry” because it sounds elitist but I use it for practical purposes.

Because I feel I was not fully understood I want to clarify my points again:

1. Blender should solve one problem.

What I meant is that Blender does has more width than depth. Some understood that Blender should do only one thing and forget the others and some thought that this goes against Blender’s philosophy.

  • I am talking principles and not details. Blender should have a killer feature that makes it a must have tool. Being a “all in one” tool is not a killer feature and as far as i know it didn’t work in the over 10 years of Blender’s existence.
  • Now you say: But Blender runs on my slow computer and it allowed me to learn 3d, it is free, open source, down with Autodesk!, Blender is for the people!
  • I reply: The post is about how Blender can get into the industry. That could bring immense benefits to Blender and to you. Why?

People in the industry are advanced users with advanced needs. If they start using Blender they will probably have to expand it and they will contribute back to the community.

The game engine and the sequence editor. Two of the features that make blender so different.

“Blender should solve one problem” means that it should be introduced to do one task first. It should be seen as a perfect solution to do a particular job. Then it will considered to do other tasks. Now blender can do many things but advanced users always find a show stopper that makes them revert back to try and tested tools.

2. Accept Maya is king and steal users from it.
Personally I think Maya is an archaic piece of software. Its main strength is that it can be adapted to do anything and most studios I have seen have custom tools to meet their needs in one way or another.

  • I have been following Blender since the NaN days and year after year I considered it for several projects. But I never used it. Why? It was always too scary to convince a team to use a strange new software and risk that the project might not be finished on time.
  • That is the idea behind my second point. If it didnt’t feel alien to them they could produce results faster and the switch wouldn’t hurt anyone.
  • All animators I work with use Maya or know how to use Maya. We are not discussing if Maya is better or worse. It is what they are used to. If you make Blender CONFIGURABLE enough so that they feel at home when they try it then it becomes production ready.
  • You say: But blender is open source. Write the code yourself!
  • I reply: That could be applied to anything. Why redesign the interface in 2.5 if you could write a new one yourself?

But here is the real reply:

  • Individuals at studios can be strongly convinced that Blender is great and can be used to finish a project but they cant convince management that they have to spend X amount of money to make Blender usable at the studio.
  • Blender is unknown territory and the unknown brings fear. They will ask: Who is using it? Which other studios are using Blender? Why spend this time and money? How many professionals can use it?

User facepalm commented on the subject:
“Ever consider that on tight production schedules there is a little time to muck around the source code? There is barely time to whip a MEL script together much less edit C and recompile and pray for no errors.

So true.

I could make this experiment: A friend of mine is a feature film animator. He has used Maya since the Power animator days. I want him to try Blender to see if he could use it for a project. He installs 2.5 and sees that all the shortcuts are different and that he has to change them all one by one. He tries the graph editor and sees that he can not configure it to behave the way he has been using graph editors for the past 10 years. He gives up and tells me if he can use Maya because that is what he is used to.

The point is not to make Blender like Maya but to make it configurable enough that it can behave like Maya…if you choose to.
Bring Maya shortcut presets and allow small but important configurations in the graph editor and animators will happily switch.

Nuke comes with shortcut presets. Modo too.

3. Don’t become obsessed with simulations.
Most people overreacted on this one. I dont want the development of simulations to be stopped. I have looked at it and it is brilliant work. I can see that it could be used on several fields and a lifesaver for smaller teams.

  • The motivation behind my argument is that although I might be wrong I get the feeling that simulations are promoted as star features in Blender. I saw particles being implemented, cloth, fluids and now smoke.
  • And here is the thing: There is no need to have simulations to break into the industry.
  • You say: but if blender becomes a superb simulation system then it could be that one killer feature you are talking about.
  • I reply: there are easier ways and you are closer than you think. Having simulations in blender is fine. But there is no need to have simulations to  make blender break into the industry.

Which leads me to…

4. Make Blender talk to industry standard tools!!!
This time I wrote it with 3 exclamation marks.

I am happy that many users agreed on this one. Several professionals commented that this is the main reason they are not using it in production. I strongly believe so and witnessed that this is the reason too.

Some users wrote interesting comments:
MickiPixel wrote:
I fully agree with you on the points you made. I work in a large Animation film studio where we use XSI, Nuke, Lightwave ect,, We have some good Blender users here, but we cant use their Blender skills in production as the cross-app work flow is not smooth enough for our pipeline.. Our top 3D guys keep an eye on Blender’s progress and I am sure that if some of these bugs can be worked out,, they’ll have no problem fitting Blender into our pipeline.

Tim, LA:
I agree with all of your points. And I was saying this for years!!
To put Blender in CG-Workflows there is a necessity to be compatible f.e. with fbx exporter AND importer!

Nuke’s support out of the box seems really important to me too and Blender integrating well with other post-production software seems a no-brainer.

We are not talking about moving OBJs between apps. We need animation data and cameras as a minimum. We need this integrated in Blender as a starting point so integrating Blender in a pipeline looks less daunting!

Import and export menus in 2.5 alpha. This is work in progress.

5. Make it easier for newcomers

The most commented subjects were having a material library and a PDF manual.

Some users complained that this could make Blender too big. I doubt that having procedural materials like metals, water, car paint etc would make it much bigger.

If you want to include tons of rigs and meshes then make it a content package. You have Blender and you have another file with the content. Modo does it like this.

Still I believe that including several materials to start with wouldn’t hurt anyone. But first Blender needs a material browser to export and import material libraries.

The material manager needs a better browser to import and export libraries.

6. Fanaticism

Ok, this one is new. I would like to reflect on the fanaticism of some users.

  • Blender fanatics scare professionals.

facepalm summarized it nicely:
This is typical open-source horse shit. OS fanboys spend all their time screaming about freedom of speech and when somebody attempts to practice it on their personal blog it turns into a shit storm.

People keep pointing to those IRC channels full of elitist amateurs as a resource hub for blender related questions. No TD who has dumped 14 hour days over the past 10 years into feature film production has the time for nasty little 15 year olds (whether that is their real age or the age they reflect) when trying to get legitimate questions answered but instead is shit on because he uses at his JOB.

The industry is not your enemy. It is made by guys like you who probably spent too much time in front of a computer. Because they are idealists at heart they strive for perfection, that makes them become good at what they do and become hired in complex projects which are usually the big ones that demand good software.

Making professionals interested in Blender can only be beneficial to Blender and to you. Don’t scare them away with fanaticism because it is hurting YOU in the long run.

Blender is the best! Yeah!!!

  • Blender is a just piece of software, Maya too, not a religion. You will not find happiness by choosing one or the other. People are not your enemies, your mind creates your enemies. Try to be nice to others and life will treat you nicely.
  • Remember: A fanatic is someone who knows everything but his own ignorance.

Conclusion: Having followed Blender for years I feel that 2.5 could be it.  Blender devs, you are closer than you think. Good luck!

Some links:
Blender Official Wiki converted to PDF
Blender to Nuke camera export. Exports .chan. Only for 2.49

5 things Blender should do to be succesful in the industry

1. Solve one problem

  • Blender should focus in solving one single production problem. Now it is a jack of all trades and master of none. Blender should focus on one thing and be very very good at it. The best if you wish. In production we dont want another tool because it is free or has a great community. We want our lives to be simpler and go home earlier. Give us a tool that can do that and we will use it.
  • Cinema 4d was where Blender is now. Learn from what they did. How do you get into an industry where you only hear Maya? Do something completely new that makes peoples lives easier.One example is Mograph. Similar to the array modifier in Blender but with many more variants that can be combined. Now it is widely used in TV for motion graphics. They applied the same philosophy with ProjectionMan, a tool for matte mainters, and BodyPaint. One problem, one solution.

    Mograph in C4d 11. Check the toolset on the right panel. All this can be combined for unlimited results.

2. Accept Maya is the king and steal users from it.

  • Most animators use Maya so make their lives easier and dont force them to change their habits. They expect a channel box and marking menus. Allow the graph editor to look and behave the way they are used to.

The graph editor should look and feel the same as Maya if you want. Now clicking on a channel doesn’t isolate it from all other channels. There is too much switching on and off and collapsing and expanding.

3. Dont become obssesed with simulations.

  • Smoke in 2.5 looks cool but TDs will always use Houdini or custom tools. Simulations are a difficult field to compete. More people will become animators than TDs. Try to make them happy first.

4. Make Blender talk to industry standard tools.

  • Support Nuke right out of the box. Now exporting cams to nuke as fbx doesnt work. Multilayer exrs dont come out as expected outside of blender. There is no good support for custom passes and so on. In studios we already have ways of doing things, team members expect files in a certain way and blender has to be able to adapt to that. If compositors cant get cameras from blender to nuke in an easy way Blender will not be used. No matter how fast that AAO is.
  • After Effects is the same story. Widely used but support is low. Integrate scripts into blender that allow easier communication between the programs. Most people dont have the time to dig into forums to find the right scripts and install them.

Render passes come out inverted in multilayer EXRs. Cameras dont align properly when exported as FBX no matter what you do.

5. Make it easier for newcomers.

  • First impressions matter…a lot. The first thing most people do is create an object, add a material and press render. Impress them! Include with the software a lot of presets. High quality materials, character rigs, hair presets, lighting rigs. Now all this is scattered over different sites and forums and takes time to find. Include them with the software and create an official gallery where users can share assets.
    This is Cinema 4d’s content browser. It is full of good content.

Luxology’s modo has a great site for sharing assets. With the size of blender’s community this could grow exponentially. This should be on blender.org and not on external sites!

  • Include shortcut presets for Maya, XSI etc within the software. They will be using Blender in no time.
  • Create short official video tours. With professional voices preferable. Include links to them in the splash page.
  • Use standard naming conventions. What are lamps? Lights right?
  • Rethink your defaults. Default renders going to /tmp are hard to find. Settings being saved as .b25.blend are hard  to find too. All very unix like but not user friendly.
  • To blender users: Dont expect everyone to be a coder. Dont expect people to code features for themselves. Dont be a fanatic! Blender is not yet “the best”. Be nice to newcomers.
  • Remove or hide old stuf. Mist and stars were all wonderful in the 90s. Not anymore! They are too visible in world settings.
  • Create an official PDF manual. All apps come with a manual, blender doesn’t. The wiki doesnt count.
  • Try to get pro training by digital tutors and others.

Conclusion: I know the work being done in 2.5 is making studios have a look at blender again. Many did in the past but left frustrated. Make the first impression a good one and try to remove obstacles between blender, the user and other programs and it could become a major tool in many studios worldwide.

>>> Click here for  PART 2.