Let's pick on Andrew's GUI proposal

I am sorry to add yet another UI thread to this forum, but so far I see a lot of support for his proposal and also a lot of hate. But we don’t seem to be saying specifically what we like and what we do not like about it.

It would be helpful and more organised if we state our concerns in bullet points, instead of completely disapproving with it.
Also write what are the good points about it. What would you change about the proposal to make it better?

Some folks on a blender group made a few good points and I would like to add some of my own points and challenge the design to come up with solutions. However as Andrew states in the end of his video, if you don’t do it in a constructive and organised way, then we will get nowhere with the discussion.

Here are some things I want to question myself:
-Mouse travelling distance to execute tasks- how much your mouse must travel from icon to icon

  • Giant icons and wasted screen estate around them
  • Splitting the viewport into 2 or 3 viewports (common thing when you work within a CG pipeline)- how does this work out in this one tab per task proposal? Maybe it’s worth looking into modo 701’s new tabbed layout preset.

Now please lock this thread or move it if you think that nothing good can come of it. I am hopping to get all the pro-excited people and nay sayers to write down:

  • What does the current UI design lack and this proposal absolutely sorts out?
  • What does this proposal break, that the current UI design has?

Brecht made a related document, looking into the limitations of the current GUI:
http://wiki.blender.org/index.php/User:Brecht/UI

Please be specific.

A conversation in FB

"There is 1 concern of Blender Guru proposed UI.
Mouse travel.

The common scenario is like this:

  1. Select Object/vert/edge/face (middle of 3d view)
  2. Select task, then tool (top of tab)
  3. Fine adjustment (right sidebar)
  4. Repeat 1 to 3 until done"

“Other software can use the proposed UI because their tools have toggle behavior. But most Blender’s tool is like “machine gun,” once the bullet is out of the barrel you need to fill the chamber with another.”

“I agree. This is just for beginner, we are in expert mode. I don’t even use the toolbar.”

“And that’s the reason why beginners stay beginners, UI is forcing the idea of where the mouse should be traveling next. But I see Andrew’s proposal working beautifully with tex/vert/weight paint and sculpting.”


What is the point of analysing this GUI. I doubt it will be implemented at all. Maybe my mind has too much tunnel vision and only has a view of design from the perspective of Engineering I can’t for the life figure out how people undertake a GUI design without writing down any design specifications or parameters. The way I know design is that the specifications come before the concepts or mock-ups. If you don’t have any specifications how do you know that your concepts meet the requirements you set out.

The last GUI from 2.4x --> 2.5x IIRC had a specification document. If you going to design a GUI than you need to tackle all 17 windows of Blender.

Personally I can’t take a design proposal that looks a plagiarised Frankenstein collision of Max and Photoshop seriously. I would personally love to know if Andrew uses Max. If he does than fair enough if not than I think this design is been informed by what looks cool and general inference rather than actual first hand knowledge that this ribbon interface works for a 3D suite.

The only place I have seen a serious attempt to analysis our interface is here by Breacht …http://wiki.blender.org/index.php/User:Brecht/UI

So my worthless two cents is that, I would seriously love to see some kind of design specification regarding the interface not mockups. Photo bashing different guis and saying Blender’s gui should look like this gets old fast.

It was this kind of GUI design way that brought us the old 2.5 GUI. Which is the same GUI where so much people have trouble with. So why not start at the users end this time? And not at the programmers end, where the programmer decides what is cool or not. This didn`t work really well for Blender in the past.

The user has to stand in the focus of the GUI. He is the one that uses the software then. Not the programmer. And Blender has a userbase that is made to 99.9% of hobbyists. So it`s more than time that the GUI reflects this.

I am sure that we will see your specification document at one point. All we have seen so far were just general ideas. Now follows the process to decide in wich direction the development really has to head to.

To me the wasted screen estate also is a bit strange. And it is possible to debate the necessity of having both text and icons.
But as for the icon size: I think it is useful if often-used buttons are bigger. Because that requires less strain when moving the cursor from totally a different area on the screen to exactly that button. To make a much-hated comparison to M$ Outlook: Reply and Forward are big buttons, clean up and junk are smaller buttons.
I’ve got the feeling that a button would thus apply for being big if it is much-used and used after the mouse was totally somewhere else.
So not necessarily all big buttons, like in his mockup, but some big and a lot of small buttons would be nice.

@tyrant monkey, agree 100%. The worst thing is also to note Brecht UI page has been accessed about 50 times (in 2-3 weeks), where there are the real points to be addressed concerning UI.

How about…

WE’VE ONLY JUST IMPLEMENTED THIS GUI! DON’T CHANGE STUFF ALL THE TIME!

Also - I use Blender mostly for animation, VFX and video editing. None of these are really covered in this proposal - though we see the curves window once. Also - not too keen on the wizard-based setups… I don’t like software to be double-guessing me the whole time.

Ha! My first laugh of the day. The irony of this statement is astounding.

-LP

I like his UI ideas. And when you dont need those (menus/current selected menu) visible, just ALT toggle them to hide/show.

It is great proposal for one handed pirate!

Well then at least we know that all of the Blender user that are one handed pirates will like this.

Maybe I am in the minority here because I think the 2.5 interface is pretty cool. I just hate the ‘N’ and ‘T’ panels in the 3d view, how difficult it is to read the properties and how if I switch to LMB some inconsitences emerge.

So minor things really. I think Blender has no trouble in attracting new users a lot of them and the seem to stay and stick around. I Blender users need to move beyond just blaming the UI as to why Hollywood or whoever is not using Blender.

The are so many alternative programs out there so I think we need to remember that people who use Blender use because it’s not Maya, Max, XSI, Cinema4D. The are using it because the actually like it. There is something like over 3 million downloads every year of Blender.

Without even taking a side I too urge everyone to read Brecht’s comments on this matter. I didn’t know the page existed before tyrant monkeys link above. Although the developers probably have very little time for actually using Blender as a artistic tool they don’t live in some vacuum. I’m a night owl and see Brecht on the forum here. And, did not Cycles come about because of a movie project. Where Brecht and the other developers were getting constant feedback from the artist.

Also did we not just have a major program overhaul including the UI. Well at my age it’s recent. Jesus folks they did that with 2.49a. And, by the way I got upset with that change. A attitude I laugh at now.

Maybe Andrew has some valid points. But it seems he just became aware of the size of Blenders development team with the last podcast. And, I don’t think he’s unique to this program. Over at a Autodesk someone is probably saying we need this and then the developers have to step in. And, hell they also have bean counters in the loop which thank goodness we don’t in that sense.

However while we don’t have bean counters as such we do have a money problem as I understand it. Leaving me damn happy we somehow have the number of developers we do and Ton who somehow keeps it all together. Anyway all of this will be hashed out at the Blender Conference by folks way more tech savvy then me.

Oh and while I’m here. Couldn’t a opening screen explain to any new user the right mouse click and the reason for it. Hell for that matter tell them they have just opened the fastest 3d program in existence. A little self promotion and not some apology there.

New users will find their way if they are really serious about CG Graphics and chasing Autodesk be damned. Modo might be doing that for the next decade probably with a way bigger war chest then Blender will ever have. Maybe with us Blender has a identity problem.

Yes it would be nice if learning Blender early in life would lead to a big studio job but it isn’t happening. And, from what I read the professionals using Blender are doing so for some strength. With that said I’d be reluctant to make changes for new users and in my mind Autodesk owns the professional world simply because of investment and being entrenched in the educational world. If Modo challenges that I don’t think they’ll do it by imitating Maya. After all the word innovative is front and center in their ads. But, then who’s more innovative then Blender. Did I mention the right mouse click. As a hobbyist and not a tech type that’s my two cents worth.

Added a link to Brecht’s document in the OP.

Here are CGsociety’s thoughts on the subject

I’m so sorry Andrew but I just couldn’t resist


First of all I want to say great job to Andrew for describing logical problems and solutions with the Blender UI that I feel would simplify and streamline the user experience. It’d be foolish for people not to take note of these concepts.

Unfortunately when people get used to working a certain way and with certain rituals they lose sight to objectively judge those things. People are afraid of becoming beginners again (which is ironic because the intent of the redesign is to expedite the process of getting beginners to functional users).

I wouldn’t say people are using Blender because it’s not ‘Maya, Max, XSI, Cinema4D’. For many people Blender is the only realistic option to start with. If we were to start over and all of these programs were free and open source and everyone was a complete novice to 3d, I seriously wonder what the distribution would be like. I can imagine someone downloading blender clicking around a bit and not being able to figure out how to position a few spheres with a texture or shiny material and then giving up. Those fundamental tasks require you to already have a pretty involved understanding of the interface.

I seriously wonder how many of the 3 million downloads (given they’re unique users) actually stick around long enough to become regular users. I’d say a lot are trying Blender because it’s free to download unlike ‘Maya, Max, XSI, Cinema4D’. More experienced users who’ve had the luxury to sit down and pound away at learning multiple programs may favor Blender because it can do a few extra things that a lot of the other programs can’t, but that’s clearly a minority.

As someone who came from Maya I can say that the Maya interface is much simpler and easier to learn on, and the reason why I was probably able to succeed in switching to Blender is because I already had a foundation with Maya and I had some basic experience with the 2.4x interface.

Let’s not be blind to what’s an obvious problem. We’ve all heard stories of people who’ve only ever tried Blender.

There are some nice idea. But my concerns are:

  • How would multiple windows work? This have been one of blender biggest strenghts, no more 2d painting with a 3d view and node editor open?

  • The tabs seems inconsistent, Animation, Rendering, Modelling seem logical, but then there’s Modifiers, Background etc. Seems to me like inconsistent mix of tasks, properties and tools.

  • One of Andrew’s argument is that the current design is crowded (I agree), but in his proposal hasn’t added half of blender yet. I’d would really like to see him pack everything without it being crowded as well, especially when he’s using big icons and font…

  • Might be less scrolling, but instead there will be a lot more mouse traveling in the horizontal direction and switching between tabs.

Here are the thoughts of Leigh, CGS Staff. Just scroll down to his post: http://forums.cgsociety.org/showthread.php?f=2&t=1127034&page=1&pp=15

I believe you mean her post

Sorry. To be safe, I should have said their. LOL