September 29, 2013

Is Blender's UI Really "Broken"?

This is from a post I just wrote on the blender artists forum in response to the new furor surrounding Blender's "broken" User Interface:

Well, I just want to offer my own take on the whole blender UI debate, so here goes.

I think Andrew brings up some excellent points about Blender's usability and intuitiveness from a UI standpoint. He's not criticizing an immature piece of software quickly developed as part of a Stanford thesis. He's examining an application that has been developed over years and years by a very dedicated core group of people and with lots of input from its user base.

Why are his suggestions creating such an uproar? I believe it's because Blender is currently experiencing a growth period. The one area I disagree with Andrew is that Blender isn't broken. Blender is actually quite amazing. And now that it is relatively stable and packs features (cycles, smoke sim, 3D tracking) that are comparable to more commercial software, it is collapsing the divide between "open-source" quality and "commercial" quality.

And herein lies the problem: Blender is a developer's dream, and a company's nightmare.

Because it is open-source, the people who develop blender want it to be a playground for the latest 3D tools and advancements in this field. And that's a good thing! However, it was never designed as a product. Blender is missing one essential ingredient key to any company's success: the marketing department.

Blender is not well marketed, nor was it intended to be. The interface was not primarily designed to engage new users, but as a upgrade to improve the previous version. The UI was a vast improvement over the 2.4x series, but I believe it was still was not designed with the goal of attracting new users and/or porting over users from other software. Now that its feature set is quite extensive, the cluttered interface is all the more apparent. But that's not because Blender is bad, it's because there has never been a person in charge of making it more user-friendly - that job only becomes necessary when your product is meant to be profitable.

So, as a consequence, I agree that blender should start to consider itself more of a product. Its website should resemble that of other 3D applications. But I think we as a community would benefit from calming down, and approaching the interface logically and with baby steps.

1) If people desperately want the interface improved RIGHT NOW, create another crowdfunding drive. The stated goals should be voted and agreed upon ahead of time. The money would go towards a full time UI and UX developer at the Blender Institute once the money would be raised.

2) Create a master document (blender wiki?) where each page attacks a different aspect of blender. Once through a system of voting, the UI with the most votes is selected, we pass on the design revisions to the developers and they help identify a plan of attack from "most able to implement in the next version" to "will take several versions to implement."

These are of course my suggestions, and I merely want to continue the discussion and see what people are thinking.
If you take away anything from my long post, it's this: Blender is amazing, and we don't need to revolutionize it all right now. The interface can gradually improve the same way Blender has always been improved. Let's focus on the less difficult tasks now (reset default splash screen, update blender wiki, update blender.org website design), and move our way up to the trickier ones (e.g. icon and layout redesign).
And if you're interested in seeing what another Blender user suggested for a cleaned up UI design, check out his blog post from a few years back here.

September 13, 2013

Blender Ink Drop Simulation - High Rez

Latest test, trying out my new PC. Simulation is maxed out at 512x512. Took only about an hour to render 120 frames using Blender internal. Some post-prod in after effects (slight blur, sharpening). It helped that I rendered out 16bit png's so I had a bit of latitude when tweaking the colour.