Usability should be the focus point - 3DeLight CTO says 3D is still too difficult


(thomascheng) #1

I agree.

https://vimeo.com/263539245


(Ace Dragon) #2

From the perspective of the world of Blender, 2.8’s development is right on top of the concept of making 3D more usable and less difficult.

But the tricky part is this, how do you make 3D software far easier to use without sacrificing chunks of the incredible power that some apps. give you (via the huge array of tools and features they give you), how do you do it in a way that doesn’t hide the advanced tools away from the userbase when they need them? In 2.8’s example, some of the new design notes appear to try to ensure that the workflow will be better for newcomers and power users alike.


(burnin) #3

Ooops…
… and then somebody makes ‘an orange a plum’, simply because they were in position when opportunity occurred to express their PoU (perception on usability). Should i comply? Suddenly, I have no other choice. // In that regard i prefer path tracers over ray tracers (also have more naturalistic jargon and less bias). :eyebrowlift:

Another thing is respect, let techies be techies and artists be artists. Both learning and listening each other and occasional by-passer. Hand washing hand. At least we’re not witnessing dumbed down egoistic button fuckers and screen lickers, as it can be observed with mobile users and selfiepidemy. (On that note - no touchy gestures i consider a loss for blender)

Knowing how to hold a pencil and draw a line of figure, one doesn’t make an artist. Yet majority working, doing just that, seem to regard themselves as such and then so much more, especially when wage is higher. Same as developers making beautiful code. Was already mentioned by Ton. Thinking again, the Industry seems to watch closely what’s happening with Blender and the Community… it’s learning. So, good for them. :slight_smile:


(BeerBaron) #4

I think we need to draw a distinction between “difficult to learn” and “tedious to use”, both fall under “usability”. If something is unnecessarily tedious to use, nobody wins. However, if something is more difficult to learn than necessary, the people who have already learned it are at a competitive advantage.

Let’s say CG could be made easier to the point where the amount of people giving up on it is halved. If you’re a CG professional, you now have more competition in a market where it’s already difficult to survive. Why do you believe so many professions have licensure? It’s the professionals lobbying for it to limit the amount of labour, increasing what they can charge.

Why do you think people have so much respect for someone who becomes a lawyer? Quite clearly, if only laws were simpler, learning the profession would be simpler. Laws could be simpler, but they’re generally made by lawyers who have nothing to gain from laws being simple.

CG is just as much about art as it is about engineering, if you limit yourself to “just” being an artist, then you’re just not that valuable. You can ask any painter or sculptor how hard it is to make a living being “just” an artist.

Also, don’t forget: Any effort spent on usability can not be spent on other areas. So unless you have a volunteer that wants to work on usability (and only usability), you don’t get it for free.


(joseph raccoon) #5

I have to agree with BeerBaron, most things to learn in blender are not in and of themselves ‘hard’ , but but there is quite a bit of blender to learn and the documentation for it is umm…filled with considerable noise in the information. if anything I would say that what makes learning blender the most difficult to self learn is finding a guide to teach you enough starter commands to be functional in each of the editors.


(m9105826) #6

There isn’t a single rendering team in the world that’s going to pick the renderer that gives them LESS control. When there’s a deadline coming, you want every knob possible at your disposal.


(Ace Dragon) #7

There are many areas where CG creation can be made easier without shafting the power users (Zbrush for instance has some ideas on increasing ease of use through better quality and smarter features and tools).

The only people who would lose out on easier to use tools would be those who decide not to take advantage by pushing the software even harder (because you now can create more content in less time). I surely must’ve missed that worldwide economic calamity that started when GUI-based operating systems became mainstream and anyone (not just pros) could now use a computer :spin:


(BeerBaron) #8

You misunderstood the point. For the economy at large, for the consumers of labor (essentially all of us), it’s of course best for the work to be as cheap as possible. That is, everyone’s work except their own.

If you’re a CG professional, it’s in your best interest for the barrier of entry into the trade to be higher, not lower. Imagine being the guy that really nailed a photorealistic look with <complicated renderer>, something that only very few people could pull of. That skill is now practically worthless, because (almost) everyone can pull off photorealism with Cycles or Arnold.

Imagine being an independent game developer before Unity or UE4. You had to have a lot of rare skills, the market you had to compete in was small. Now that even amateurs can publish videogames that at least look good, the market is drowned in mediocrity and even the best games get relatively little visibility.

That doesn’t mean “progress” towards simpler tools can’t or won’t happen, just that it isn’t necessarily in everyone’s best interest. In my estimation, the people that have put the biggest stake into learning Blender are done complaining about the UI, but at the same time you have other people arguing that a better UI is the most important thing in the world. Developers can burn a lot of hours on making the UI somewhat easier to learn. Or, people can learn the UI that already exists and developers can focus on more powerful features, or more bug fixes, or more optimization, or better hardware support…


(Ace Dragon) #9

So are you actually against the idea of the masses being able to make 3D images/animations/games without years of technical training and practice or not?

I’m also trying to figure out if there’s a point to be made in assuming a confrontational persona against nearly every idea and concept posted in this forum, because it does sound like an attack on the very concept of lowering the barriers through better, smarter, more powerful, and less buggy tools without a high cost of access (both in software licensing and hardware costs). Can’t have those pesky commoners steal our revenue, gum up the forums, and bury our art threads.


(BeerBaron) #10

No matter how much lower the barrier is, it’ll still take years to become good at anything. I’m against the idea that “usability” (in the sense that it optimizes on that barrier) should be the focus. I’ll trade virtually anything else (functionality, performance, stability, …) for a steeper learning curve, because I already went over that curve and I’m ready to do it again. That’s my interest, yours may differ.

You’re making it sound like your idea comes for free, but my point is that it doesn’t and that such a focus is actually not in the best interest of the Blender users that invested the most. You also really should be arguing that all these newcomers will surely bring in more donations for the software to be better. Maybe that’s true, maybe it isn’t. In my estimation, the people willing to put in less effort to learn something are also likely to not spend any money (or earn any money, for that matter).

I’m also trying to figure out if there’s a point to be made in assuming a confrontational persona against nearly every idea and concept posted in this forum, because it does sound like an attack on the very concept of lowering the barriers through better, smarter, more powerful, and less buggy tools without a high cost of access

You can’t have cheap AND smart AND better AND more powerful AND less buggy AND easier to use. You have to trade off something for something.

Have you ever done any client work, at all? If so, this should be familiar:
https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcQEXVIA-jKKxFKoHbOupFkluLD3cvej_xbtvL8j3rfJTQlN-WxsZQ


(Ace Dragon) #11

On the contrary, Blender 2.8’s development is taking steps to provide nearly all of that at once (compared to where 2.79 is).

Godot is looking to do the same for game development (even the less buggy part). Unity and Unreal 4 gets piles of features in each release and are technically developed faster, but based on their community discussions you could say they are like a huge toybox with half the items damaged or broken. All a new engine would really need to do at this point is provide many similar tools and features that just work.


(cekuhnen) #12

So true - currently I fight with students who want to use SketchUp instead of Fusion360 which works the same but offers dramatically better deign tools that deliver perfect geometry results and unlike SketchUp are parametric.
No they want to use dump tools.

I really think often people mistake easy to use with cannot do much.


(BeerBaron) #13

There’s an idiom called “spreading yourself too thin”. To focus means to put your attention somewhere specific, as opposed to everywhere. Blender 2.8 isn’t done yet, they’ve already cut targets (as they should!) for the first release. Let’s consider all the money spent on Blender 101 (something which many users are unconvinced of), it obviously could be distributed to other things.

All a new engine would really need to do at this point is provide many similar tools and features that just work.

<sarcasm>Well, those other market leaders must simply be stupid to not to do everything everyone wants, instead of only doing some things some people want!</sarcasm>


(Ace Dragon) #14

Isn’t this why the Code Quest exists, so they can keep the targets cut for an initial release to a minimum?

The BF has a small army of devs. now for at least 3 months (and they aren’t GSoC students), I can see a ton of development happen as a result.


(thomascheng) #15

I think this is generally the direction we are going anyway. Zbrush exploded because they made it easy to digitally create extremely high detail organic characters. Before that, I saw some dude try to model in wrinkles in some old lady face for the XSI face robot presentation. Blender saw a big uptake in users when they adjust their UI in 2.6? Cycles, Vray, Prorender, and many others are now using a unified sample system and yet still very powerful. No more balancing and adjusting 10 different parameters for the fastest render. Auto rigger for quick secondary characters. Liquid sim with automatic splash and foam effect. Substance Painter and 3D Coat for PBR Texturing. So much easier.

In general, the presentation is trying to save artist time, by making certain things easier. Many time, I’m up against deadline and I can only do 1 or 2 solid iteration before the client finalize it. There still much I wanted to include, but can’t due to time. Anyway, as an artist, I rather make great art and spend more time learning better art skills. I don’t really care much for the technical side, except that it helps me reach my goal.


(theoldghost) #16

[QUOTE][Anyway, as an artist, I rather make great art and spend more time learning better art skills. I don’t really care much for the technical side, except that it helps me reach my goal. /QUOTE]

@thomascheng - Amen to that nor should you which was the entire point of the presentation.


(JustinBarrett) #17

Meh, one of the things I find appealing about 3D, programming etc etc…is that it is a little bit on the technical side…I like the fact that not everyones doing it…I also am not a fan of gimmicky software or niche software like Zbrush…do not get me wrong…I think zBrush is an amazing tool…probably not my best example…something more like ultimate unwrap or whatever dedicated to one specific use…those things just kill me.


(BeerBaron) #18

Zbrush has a really convoluted interface with tons of buttons and knobs and unnecessary features. Mudbox is really streamlined and easy to get into. Zbrush won over the market, Mudbox is almost irrelevant.

Before that, I saw some dude try to model in wrinkles in some old lady face for the XSI face robot presentation.

If all you have is a hammer, it’s really hard to drive in a screw. The crucial difference here isn’t “usability”, it’s functionality.

Blender saw a big uptake in users when they adjust their UI in 2.6?

Going from 2.4x to 2.5x was a pretty long transition period. I’m not aware of such an uptake.

Cycles, Vray, Prorender, and many others are now using a unified sample system and yet still very powerful. No more balancing and adjusting 10 different parameters for the fastest render.

Path tracers like Cycles are simpler because they don’t have the possibility of tweaking the algorithms that allow for faster rendering. For best performance, you still need to tweak the bounces, plus you need to optimize your node trees (effectively an infinite amount of parameters).

From a code perspective, there’s tons of parameters that could be exposed so users have the possibility to tweak or they could be hidden, to spare the user the true horror that lies beneath it all. Tradeoffs, tradeoffs…

Anyway, as an artist, I rather make great art and spend more time learning better art skills. I don’t really care much for the technical side, except that it helps me reach my goal.

That’s an okay attitude for a sculptor or modeler or painter, not so much for anyone who needs to solve novel challenges. It’s great if your program automagically can rig a biped, but what if you need to rig a centaur and you don’t understand how to do it, because you never bothered to learn the tool at a deeper level? You’ll fail the job.

Or on another level, if you can do just a little bit of scripting to automate tedious tasks, you can save yourself hundreds of hours of manual labor. You’ll be worth more, as a professional. If you want to focus on “just the art”, you better be really good, because competition is tough.

Anyway, I made my point. You can have your little thread back now.


(Ace Dragon) #19

Look at the first batch of Zbrush demos and then determine if they could’ve been recreated more easily using an industry standard 3D program at the time with a conventional interface. By the looks of things, the awkward interface wasn’t enough to cancel out the benefit of its innovative features.

Though what helped is that Zbrush back then had far less tools to fit into that interface (people on CGSociety are commenting on how the UI is getting worse through the continual accumulation of new tools and modes alongside their policy of not cleaning out their outdated and less powerful forebears). Eventually, Autodesk could put a serious effort into Mudbox again and take a bit of marketshare by promising a clean UI and minimal bloat while offering similar features (Mudbox by the way is actually receiving new development).

That in a sense could be how to ensure that better, easier, and smarter tools could backfire (if you don’t clean out the older cruft that barely anyone uses and is only there out of fear of breaking someone’s workflow).


(cekuhnen) #20

Yeah look at Photoshop vs Affinity Photo

It seems all those older apps are at one point falling victim to their old heritage / foundation

great that Ton has no issue pushing Blender through another UI face lift.