ZBrushCentral

ZBrush GUI - Principle of Circularity

Principle of Circularity

I have added an article to the main page of the ZBrush Wiki that talks abit about some of the ways I see ZBrush’s interface.

You can find it here.

I would love to know what the ZBrush community thinks. It is an early draft but I would love to hear your opinions on what I see as a defining aspect of ZBrush’s GUI.

Ryan

Hello Ryan,

I hardly post on ZBC, but lately I have been using ZB a lot. And I find this topic interesting. I do agree that the circularity workflow is good, and the way one can mix 2d, 2.5d and 3d is pretty nice. I do think that this principle allows the combination of multiple tools as never seen before.

My only complain, as well as most of the users, is that maybe we could have a better customization of shortcuts and general navigation. Thats nothing to complain that much, but sometimes I find myself trying to operate Zb like I do with my other 3d softwares and vice-versa.

I cant think of anything to change on this GUI, as after lots of working hours, one gets used to it. I think its more about features, and in this case I would prefer to wait until ZB2.5 before talking and complaining.

Main thing is definetly a layered texture as well as a multi-brush painting… I’m sure Zbrush already have its own kind of multi-brush, since you can paint displacement, color, shading…all at once…but having this managed as layers would be nice.

Now, back to the topic, what I would like to add to this concept of circularity is:
Displacements already treated as a texture. I dont know, but imagine that while you sculpt your model, you have a texture slot, called displacement… so the exportation could be easier…just like you do with your regular color texture. Again, having the concept of layers would solve this.

Also, I know that this changes with way we use the software, but maybe we could have 2 layouts/ways to work.
1 - just like it is.
2 - sculpting layout - this way, all you would need is the tools used to sculpt and texture your model. Without the Document stuff… so you can open your tool, save it…as a regular windows application does “File>Open/Save”

I’m sorry for the long post…and I hope I’m not out of topic here… In case I’m, I ask you to reply me, and after that remove my post.

All the best,

Interesting view. It never came across like that. In my quest to get a handle in ZBrush I deleoped a slightly different way to look at the tasks at hand. The described ZCycle is an interesting view but a lot is obscured by the UI. It takes the noob a long time before he learns that the intensity setting for the smoothing action is only available when I open an unrelated tab and select a sub function which then, without any further indication, allows me to adjust the smoothing intensity with the intensity slider… That is well hidden and breaks a gap right into the smooth circular workflow. Another ‘trick’, which I only saw Rimason apply masterfully is the axis limitation. the little xyz Button is hidden nicely. Another one is the application of displacement maps. That is missed for quite a while before one stumbles across it. It’s hidden in the ‘mental’ toolkit one has to build up as it is merely a methodoloy rather than a tool you can select. It’s another deformer but cannot be found there. The unison of deformers and controlling image maps is there but has to be imagined before it can be applied. It’s not a function which is selectable. Another hidden gem is the depth limitation function for paint activities. THAT is a cool features and can be used to achieve really cool effects. But again, you have to put your thinking cap on to put a few things together to get it going. If those functions mentioned, and a few others, would be integrated in another abstraction layer, then the tool would be even more powerful without actually changing the underlying functionality. An abstract Idea I had for quite a while is an ‘iI’. An intuitive Interface which analyses previos work steps and then builds a pop up menue which is populated with the most logical functions for the possible next steps. It would have to be activated as it would be rather annoying to have an assistant all the time floating around the mouse harrassing you with wild guesses. But an alt mousedrag could open a menue floating around the mouse which after an object was loaded for example would invite pesets for modeling or texturing e.g… If modeling is chosen the UI would populate with the primary modeling functions or a user defined set of those. And if the alternate function has been chosen it would conform to that. And if modeling has been chosen and a certain subdivision level hsa been reached and the previous modeling activities point towards a possible map generation (detailng and pinching etc was used) then the alt mouse drag would be preset with the functions for the normal/displacement map generation. Imagine that Ohhh and Ahhhh when that would be available.
I have a few more ideas into this direction but this mail is already long enough… Actually… I should patent that idea… :wink:

Lemo

PS:Hiding the mesh import options (Weld etc.) behind the star fish is also an evil act and seems only to be designed in that fashion to make fun of noobs. “Hahaha, let’s see how long it takes them to find that out… lol” :smiley:

PPS:Why don;t you guy’s make some screenshots of the 2.5 gui. Maybe we can have some last minute ideas. Or are you just starting to create it? (Taking cover:cool:).

Lemo,

Your “adaptive” UI is quite a nice concept… pick up the mostly used commands and arrange them on a menu, side panel or even a RMB pup-up…

As you mentioned…some features are so deep inside or only appear within certain other fuctions that one will take some time to find out.

I use Zb since dec,04…and I’m sure there are lots of features I dont even know they exist.

the ZCycle as you properly named it, would benefit from a clear layout and better destribution of the toolsBut what I find “logical” may not sound “logical” to another user…so…at the end, being able to Customize is the logical step for a good “iI” GUI.

Maybe, a feature that could be implemented:
Dockable/Undockable toolBoxes…the same kind of stuff we can see on 3dsMax…its quite interesting to have one monitor with the viewports and another one will all sorts of commands (if you can put in a toolbar, you can convert into a toolbox), this is useful to clean up your main window but also tends to stress you with lots of large-range mouse movements.

FOR SURE some screenShots of ZB2.5 would be nice.

all the best,

Lemonnado - you should know better!! :smiley: Please no comments about ZB 2.5 here. That seems to be a hot topic. I am very interested in exploring the idea of circularity and certain things that help me conceptualize the ZBrush interface.

Your thoughts are very interesting and, sorry, but you lost the patent by posting on the web. :wink: That said, I will modify it slight and patent it myself! :stuck_out_tongue: I’ll take you to your favorite Mcdonalds, though!

One thing I want to avoid, however, is a why this, why that debate. I have no control over those things and no real part. Also, it obscures my very genuine desire to explore the understanding behind the ZBrush GUI as it exists. :smiley:

I got to run out of the office so I will revisit this and write more in a bit. :slight_smile:

Thanks for the awesome thoughts guys. :slight_smile:

r

A good way to conceptualize th ZB GUI is thru the concept of Neural Networks, but analizing it as a concluded neural process.

1: Ofer taught ZB how to react to several commands.
2. Zbrush now reacts to the user actions showing the user the specific components used by a given command.

The ‘ZCycle’ is included on this general concept… Like a neural network derives one information from another, and interconnects neurones to form a bigger information so does Zbrush.

Where you can connect and “assemble” blocks of “ideas” and use them into different parts.

The only thing that doesnt happen like it does in neural networks is:
Zbrush doesnt learn anything from the user…or…it does at a certain degree, when you customize your GUI.

If we think in a very abstract way…this neural network concept does exist in ZB…or… I’m getting brainless…what is also possible.

Lemo’s idea of iI would complement this concept…for sure.

regards,

I think it is easy to learn what tools do in software.

The hard part is finding out what things a software is capable of and learning the processes necessary to achieve these.

By talking about circularity in ZB, I think you are essentially saying a similar thing - one achieves things in ZB by following processes that involve elements/tools from the different palettes.

The limitation of this is that it doesn’t explicitly tell you what processes can be achieved. Experimentation and discovery is a grand thing, but not when you just want to get something done.

It also isn’t helpful when the process isn’t self-explanatory or obvious.

A simple example of this is trying to paint with the simple brush when Zadd is off - nothing happens.

A user might then go around in circles saying, I have a colour, an alpha, a stroke type, an rgb intensity, so why doesn’t it work? Nothing in the palettes will tell them that they have to fill the canvas with a ground layer first.

An interactive UI could be used to overcome this, for example, when you try painting with the simple brush without filling the layer first, a pop-up could instruct you that you need to do this (much like the warning about switching tools in edit mode and dropping to canvas).

The CTRL help pop-ups could also be expanded to include a list of what is needed to use a particular tool and what palette you have to go to to find it.

Another example is the Cube3D and other parametric tools. You might want to subdivide these and access the different subdivision levels. It isn’t obvious, however, that one first has to make a polymesh of them. Such a tip could pop-up when the Divide button is pressed for any parametric object. (On the subject, I think there is an error in the CTRL help for parametric objects; it talks about the Modifiers:Initialize sub-palette, which should be the Tools:Initialize sub-palette.)

Such hints could be set up as optional for those who understand the intricacies and quirks.

Hey Ryan the concept you illustrated is very interesting. never thought about that cycle thing.
this being said, this is more related to ZB approach to creation, of objects, textures , alphas etc rather than being focused on how the user could deal with ZB UI. I do agree with Lemonado. some things are somewhat buried in the ZB UI and the user cannot understand em quicly, and sometimes, he doesn’t find those optinons. little example.: whwn working with z spheres if i wanna increase the Zsphere subdivision, i have to be sure i am in draw mode, and go to transform menu so that i can see and use the x, y, z resolution sliders. otherwise i don’t see them. so a noob couldn’t find em so easily at a first glance. maybe he’s in move mode and asks himself where the dam resolution sliders are. :smiley:

UIs are about categorization, and what many of you think you know about UI construction is wrong. In particular, old ways of attempting to apply categorization are a bad fit for ZBrush which needs to be built on the most elemental semantics. There are of course mistakes, whoever built the UI struggles for time like all of us, but ZBrush does, thankfully, avoid habits of mind that are left over from earlier strategies.

ZBrush’s UI feels like a fully abstracted application programming interface, and ZScript relies on the the fact that this is the case. Various functions are available they all accept various data types and don’t care where the data originated (there’s your circularity).

As experience with ZBrush grows combinations of the elemental semantics will appear (ZScript, Tutorials, etc) but these strategies belong to their own compound category and really shouldn’t pollute/poison the mind of the ZB purist, the principle of circularity seems to me to be another concession to those that don’t-or-won’t get it. Nice try though.

From day one (of using ZBrush) one of my strongest wishes - if anybody had asked - would be that nobody had a reason to reflect about the GUI of ZBrush.

This has not changed.
We all have some different apps for our grafical needs, and all have a SLIGHTLY different GUI. But, most of them seem to follow a nondeclared standard.
Makes it easy to switch between (just an example) Maya and Silo, back an forth.

So, when I see a discussion about the Principle of Circularity that seems to be present within the GUI of ZBrush (sorry about the change of the font just by copy & paste:)) - well, I guess it’s alright to have this academic discussion.
But for the every day work I would like to stay (or become) this GUI a part of a tool. Something that needs no extra thought of how to handle.

Believe me, the first hammer (a tool like ZBrush, maybe a little less in complexity) that needs the discussion of its principles will be thrown out of the non-digital toolbox.

Anything that will make the sheer working process inside Z2 more easy (because of a more ergonomic GUI) will be better news than “… and now it can handle two gazillions of poly’s!”.

But, I might be totally wrong (and with me I do not know how many users).
Thanks for reading.

The BEST UI ever is the C4D one. Maxon has clearly maxed it out. It’s the only, and then complete, object oriented UI without any compromise. You can turn the whole thing upside down. Colors, Menues, Submenues, Commands,Windows, styles. EVERYTHING can be redefined for any purpose. God, even the layout definitions you like can be put as buttons onto the workbench. Floating, as tabs… you want it, you have it. I tailored an XSI exquivalent layout together and have now real problems distinguishing between them hahahaha. That totaly distracted me from my XSI/C4D animation transport hahaha. It’s a lot of fun actually. All drag and drop amd flip and flop. Not tedious at all. You can also have dual layer hotkeys. That means you can press one character as hotkey and then a second one in quick succesion and both lead to a hotkey function. ‘M+R’ lead to ‘close polygon holes’. Wild… Every other UI builder should check out Maxons way to organize things. It’s a lesson in what can be achieved if you folow through with the OO paradigm.

Anything less is just cutting corners… :wink: Wink, wink, nuge nuge, know what I mean… nudge nudge…
Lemo

Sometimes it’s more like the Principle of Hexaflexagonality (especially with some of the more obscure easter-egg functionality that Lemo points out.)

One interesting byproduct of the incredible recombinant capabilities of ZBrush2 are the various ZScripts which cluster operations that normally require running up and down several different palette staircases to accomplish a task.

Some of the more ambitious ZScripts manifest these otherwise non-obvious tasks (most of the Marcus-civis scripts do this.)

Seriously though, I think you are on to something Ryan.

Sven

Haha will have to download myself a C4D DEMO and try that one. i love flexible UI concept and surely Maxon is good on that. however i don’t have great problems about ZB one too. its more a matter of workflow and some limitations, but definitely not great problems.

Ryan,

Your article finally woke me up.

I teach Photoshop for a living, and so always strive to teach them the theoretical thing, so that the rest follows logically. And that’s what you accomplished for me. I bought Hex2, have been checking the Nevercenter vids, reading the Mudbox threads, etc., all out of a fear of the ZBrush interface, hoping to find something better. Last adventure was thinking that I could achieve the same thing with BodyPaint 3D. Hardly.

Because it’s circular, it’s not always apparent where to start, but there’s freedom in that! Thanks, man, for teaching me how to fish, instead of hooking me up with one fish sandwich like most instructors do.

BTW, your brick pavement texture demo is a good example of this way of thinking.

Best wishes,
Chris Collins
www.chriscartoon.com