Andrew Price is continuing his research into usability for Blender with a user survey.
Andrew writes:
In an effort to improve blender's interface, I've created a survey aimed at finding out how people use Blender and for what purpose.
It should only take a couple of minutes to complete but will be a big help to anyone designing the interface.
The results will be published publicly next week on Blender Guru, so that everyone can make use of it.
Thanks in advance!
47 Comments
What a good idea it would be if the Blender Foundation were to run a yearly survey like this, with more questions.
so if many beginners votes for something, developers will change blender for the beginners desire?
:(
There's no reason why a more intuitive interface will be a bad thing for experienced users. As long as any GUI element can be switched off and the support for keyboard input remains, you can continue to use Blender exactly as you do now.
But having said that...
More users means more eyes looking at bugs, more funding for devs, more ideas for improvement and it means that Blender is hitting it's goals as stated on the Blender Foundation homepage: http://www.blender.org/blenderorg/
More users means a better Blender.
So (in my own opinion of course), if some existing users are forced out of their comfort zone when these issues are addressed, then so be it.
I totally agree, Corsair
Just did the survey, some of the questions seem a bit vague to me.
The Question regarding which rendered I use: I said Internal not because I wont more development on it, but because Cycles needs more development before I can switch.
So would a high Internal number mean that developers should concentrate more on Internal or that they should spend more time on Cycles?
ye same me I used almost all functions every day exept vertex paint and BGE
I also do most of my work in internal, but this is more because my current computer doesn't support GPU rendering. I'd rather spend my time focussing on modelling and character animation for now, then update the textures for cycles once I upgrade my system.
Internal renderer has its charm, specially speed for NPR. Cycles is incomplete, ETA is vague. Abandoning BI seems very irresponsible of BF and devs. No wonder Blender got harsh comments from people in the industry.
Just read here:
http://forums.cgsociety.org/showthread.php?f=2&s=a21d86c0fd285baa00d81ec60c84ae66&t=1127034&page=1&pp=15
The idea of the survey is to find out what you actually use, not what you wish you could use.
Also did the survey.
When trying out selecting things with the left mouse bottom i found out that the posibillity to lock the position of the 3D Cursor would be extremly helpfull!
Maybe a good idea if you want to change the default button...
Since we're being asked our opinion about the interface . . .
I'd like to see Blender get a re-vamped Materials/Textures Properties panel, or go exclusively node-based and get rid of it altogether. It's cluttered, and 'counter-intuitive' seems too mild and unfitting a term for how unwieldy it is.
Otherwise, compared to that 2.43 thing of beauty, I like the interface just fine.
Survey would of been more helpful if it had of asked for feedback on the suggestions you put forward in your video series. Most of which I agree with but I disagree strongly on your views for the terminology used in the software. A survey on how people feel on little things like that would be great.
Lets fix Blender and make it user friendly, but just don't go too far and 'dumb down' the software for beginners to the point where it becomes frustrating for people who know what they're doing. I've seen so many of my favourite programs do that over the years and it would pain me to see that happen to Blender of all programs.
Blender needs a UI tune up, not a complete revolution.
Is Maya easier to use than Blender? From what I hear, it is not.
It is not (supposed lack of) ease of use that is holding back adoption of Blender, it is marketing, pure and simple. Companies selling proprietary tools simply have a bigger PR budget than the open-source community can manage. But our advantage is that they tend to have a short-term view of things (look at the number of applications Autodesk has acquired and then crippled or killed). All we can do is keep quietly plugging away for the long term, building up an ever-more-impressive portfolio of art, and community of artists producing that art. As the saying goes, it takes years to become an overnight success.
Andrew is wasting his time.
You couldn't be more wrong - http://forums.cgsociety.org/showthread.php?t=1127034
That’s a funny way of saying I’m right.
Looked at the first few comments, and they’re along the lines of “we don’t use it because we don’t”.
Thanks for posting the link. I think the majority of posters were giving a fair and honest appraisal. Main issue being as an artist looking for employment, it just makes sense to learn what the market is using. I hope with interface improvments in Blender that will eventually change.
In my view, Blender isn't used by a lot of people, especially to pro, not because it isn't good or great. It is just not what majority uses.
It is a false premise that having something good or great would make it popular or used by many. There is more to it than that. Marketing is actually the more strong actor than quality of the product. With marketing, you could make the product appear constantly in many 3d magazines and books, advertisements in many websites,ect. The marketed product has a lot of exposure and so is often choosen by the people who got interested in making 3d. Blender on the other hand won't even appear in tiny prints on magazines or books. An advert about Blender in non Blender related sites (but is related to cg) is non existent. It is just not possible to chooses a product you can't see!
My experience is in that direction. The very first one I liked to try was actually maya and zbrush. I had used linux in those times so I used the free versions and use it with wine, which create some problems. Then after some weeks, it comes to me to search for 3d products that runs natively in linux. The first one I look for was the linux versions of maya, max, rhino, houdini, etc. Then I found Wings3D and sketchup. The very last one I have 'found' is Blender and have since stick to it.
You see? For almost everyone (that includes me) Blender is just not that vissible. So by logic, probabilty, consequences, etc. many are bound to use the very vissible products and Blender gets the very few who choose to look further; but as you know very few do that.
It's like the store shelves and why companies pay huge sums to the stores to display their products at eye level where it is easily seen. That is the point of the shampoo commercials that keep popping up. It the Vissibity of the product that makes it sell.
My feelings exactly.
Houdini was created for Linux first and its the easiest to install too. All Foundry products came from Linux. Maya is on Linux for a loooong time but problematic to install on too old or too new distributives, have to manually re-link some libraries. Both have free versions with ads poping up from time to remind you in Maya and few restrictions in Houdini.
Blender is easily found if you go to repository manager -> graphics -> 3d, how couldn't you find it yet manage to find Wings? And Sketchup on Linux with wine is easier find about?
Its obviously you're not Linux user.
*face palm*
Thay ALL run on Linux. All major (and some minor) studios use Linux. This is not about Linux.
Huh! I had not expected this from this site being not linux focused. You kind of people are the reasons I don't want to post anything in linux related forums. You always have this high and mighty attitude, af if using linux makes you better than anyone.
I was distro hopping with linux like 10 years ago, that is also the first time I tried working with 3d applications. I was searching for 3d applications while using linux for just weeks.
Yes you could find blender easily in the repository, especially if you use Ubuntu, but not all linux distros are like that. The distro I used when I had started with experimenting in 3D is Puppy Linux.
You also make a mistake in assuming I go that way you would do when I searched for 3d applications. I actually have difficulty intalling softwares, and in my ignorance I had to look at the internet at instructions how to. I use the internet reading the tutorials available to decide what software to use. At that time and the most common tutorials use maya or max and zbrush thus I tried looking at them first. The same could be said when I stumble upon wings 3D. I was reading this tutorial: http://en.wikibooks.org/wiki/Wings_3D/Tutorials , which have changed vastly now days.
After some time using Wings. I changed distro, this time I used Ubuntu, and intall Wings there. That is where I also discovered Blender when it appeared beside Wings.
I'm sorry when I was super ignorant when I first tried linux, not to mention, can't even install Open Office without reading a detailed how to. My screen resolution is wrong. Printer doesn't work. WiFi doesn't work. Sound doesn't work. And all my sweat and blood trying to fix things all stamped and crushed by your "Its obviously you’re not Linux user." God no wonder many hated Linux with people like you around.
After all you are born a genius expert in everything linux and expected I a lowly average linux user should bow down before you and kiss your feet and beg for your scraps of guidance... Oh! Linux GOD. May you give me a pleasant Linux life by not talking to me or putting your holly attention at me. Your scorn is the most unpleasant experience I could have, so please stay away.
By the way I'm using CrunchBang linux now. Ubuntu just doesn't work well for me.
High and mighty attitude? Did you even read yourself now? I just pointed out where you were false and you couldn't disprove anything other than 10 years ago Synaptic or any good package manager didn't exist on distro that you were using. You were wrighting for all distros saying that all were outdated which is not true. And was basicly saying "on Linux i couldn't find any 3D software, no Maya, Houdini, Blender etc and that is why i think they should have better marketing" Even though its not entirely true is it?
I've used many distors myself(yes, not as marginal as Puppy) and still do but never had any trouble finding any software. And certainly was never searching new program like SketchUP natively being experienced(at least 10 years, did i get it right?) Linux user.
If you would say "i was experienced Windows user and couldn't find ANY software' i would believe you, because without web browser or going to stores its impossible.
With your messages(especially the one i am answering now) you're damaging not only Linux but Blender community as well, leaving the wrong impression about people using it. Its not popularity of Blender that is the problem...
First as I have told you, I was a very new linux user at the time I was searching for a 3d package. I had only a few weeks of experience so linux is a very limited. As I told you I can't even install a package I wanted to install without reading a tutorial somewhere.
I wonder how could you miss this, quoting myself, second paragraph:
"I was distro hopping with linux like 10 years ago, that is also the first time I tried working with 3d applications. I was searching for 3d applications while using linux for just weeks."
I clearly said that I only had weeks of experience in linux when I started dabbing in 3d. I was searching for 3d packages 10 years ago. yes, I could easily find softwares I may need now with some years of experience, but ten years ago I can't.
“on Linux i couldn’t find any 3D software, no Maya, Houdini, Blender etc and that is why i think they should have better marketing” - What? I was talking about more marketing for Blender not Linux. From the begginng, I was talking about Blender. How could you make such a mistake? In my first post I only mentioned Linux two times and Blender many more times, and then you still mistaken my comment as about Linux?
You seem to have mixed up the time frame again. Yes I can find things now. but years ago when I just started using Linux that is not so.
If you still have hard time understanding what I said, or still continue with your mistaked beliefs and assumptions, then it is better to drop out conversation. It is just plain waste of effort. You just are too biased that you see things that are not there. I have my load of people like you hurting and insulting me. If you really wanted to help your beloved linux, then please do actual help and not drive people away by your repugnant venomous words.
+1 Agreed.
I've used many packages professionally, and Blender has the fastest workflow.
Of the 2 features I've worked on this year (both to be major releases) 1 has been in Blender and the other XSI.
Blender is starting to pop up in professional workflows more and more.
^^^ This was meant to be a reply to Ralmon's comment, BTW
Also, just to be clear, is this survey anything to do with the Blender Foundation or anyone from the worldwide Blender development community?
I feel what Andrew is doing is very worthwhile. Blender is frightening and awkward for total beginners, so the few tiny little interface tweaks he is suggesting would save a lot of the teeth-grinding and hair-tearing that probably scares away many new users. The splash screen should have a great big 'Get Started' link to a 'noob-to-pro' tutorial (or similar) as a lifeline for new users.
After that, they'll have to get used to learning a lot. This is where the help comes in - or should - except that pressing the universally accepted 'F1' in Blender doesn't show the help file!
The biggest problem I have in learning blender is that I use it on a machine which does not have an internet connection. What is needed is more help actually in the installation itself. Maybe this could take the form of a local copy of the manual wiki - though I don't know what size that would be. Otherwise, we need some quality documentation, in readable form rather than video, which can be easily downloaded and transferred - by usb drive or whatever. Noob2pro has a lot of good info but it is also poorly organised and has a lot of repetition, making it tedious to work through. Needs to have a good "flow" - a logical learning path to cover pretty much all features plainly. I know this would be difficult though as blender is constantly advancing and changing, and tutorials quickly become out of date
Yes +1! I am happy this point has been brought up. When on a machine without internet, it is impossible to look up anything. There was a downloadable version of the Blender Wiki a few years ago, which unfortunately hasn't been updated since, it was somehow generated from the Web-Wiki with some script. It would be great to resurect this and run an update regularly.
One other similar problem I find is the prevalence of video tutorials. Which is all good and everything, if you want to follow a tutorial step-by-step. But videos are linear and very bad as documentation/reference, it is not possible to find the information one is looking for without spending a lot of time: some tutorials go on for 20-40 minutes, while the whole content could be written in about one paragraph. Some in one sentence. They are also unsearchable.
Second, I also use and have taught Blender in a professional setting at, at work, where Youtube is explicitely disabled, and watching videos (eg. vimeo is enabled), whatever they are, is seen as using work time for fun.
When was the last time you looked at the “Noob to Pro” Wikibook? You realize it’s a wiki, which means it depends very much on the community to contribute to it?
I discovered “Noob to Pro” early on, and found it frequently out of date and incomplete. So I set about trying to improve things. It now has a much more up-to-date introductory section, better overviews of materials, textures and lighting, a step-by-step introduction to scripting, and even some material on Cycles and Freestyle now.
Learn from others, and contribute what you know.
Hi
I have a few things to say about helping blender.
In 2011 the company where I used to work started a short film project and I proposed for the company to use blender, but after the tests blender fails because it doesn`t have hair collision. The director of project gave me a chance to find a solution of hair collision and nothing worked well, in that time I made a video with the solution and with the problem, I have posted the video on the blender nation. The comunity gave me a link for one addon with some collisions and many erors. The director chose to work with Maya.
-----Blender 0 - Other 3d platforms 1------
In 2012 another project started in company and I proposed again Blender as software to work and again the hair collision is the same. In one year not even a change.
-----Blender 0 - Other 3d platforms 2 ------
In 2013 I founded a litlle company, the Blender fails again in hair collision. But I use blender because I like it really much and is fast for modeling sculpting texturing and animations and I like very very much the compositing system. But if I`ll have the chance to start a biger project I`ll choose definitely Maya.
-----Blender 0 - Other 3d platforms 3 ------
The problem is not in how the user see the blender interface with tools or not, with shortcuts or icons, the problem is what is the final result. In the last 4 years the blender results are the same as the final result and not even one director of project spends time comunicating with the developers and with techical staff for software problems or developing something.
Blender developers: Try to fix what doesn`t work or make a list of basic things of what the industry use in that case and make those things to work! The list of Andrew is ok but he doesn`t include many parts of a 3d platform workflow like tools or simulations like muscle simulatins or particle effects.
I see it the same way - lack of complete features or problems is what harms Blender much more than interface concept.
While the UI can take some time to get used to - but thats the case with each application. However if a tool does not work a feature is not present or buggy there is not much you can do when it works in a different application.
"all that is wrong with blenders interface" : I facepalmed.
You don't have a serious tool suite unless it yields serious output. Awkward interfaces can be learned, and blender is better than most.
Blender is spotty. Simulation, depsgraph issues, library linking, consistent multicore/GPU support : these have had first-pass implementations for years. They limit the potential quality of output.
The survey can't get at this fact, so it's of no use, even if the developers were listening.
*standing ovation*
EXACTLY. This is a direct result of the fact that the survey was conceived by a guy who knows more about self promotion than 3d. As far as I can tell, he's never had a professional job in the field, and this is quite evident in his approach.
More broadly, you've hit the nail on the head regarding the areas of Blender that actually need work. The depsgraph and library linking system in particular are major hindrances to large scale production work. They have not received much attention from the community, as most of the hobbyists don't even know what a dependency graph is, and that combined with the fact that it goes to the core of the program and involves a massive overhaul of the code means that sometimes it seems these issues will never get addressed.
Fortunately Ton and co seem to be well aware of these issues and plans are in motion to deal with them. Just check out this year's GSOC projects for a ray of hope. I'm glad that they're working to address the REAL issues at the heart of Blender's shortcomings.
I sincerely hope that this GUI discussion doesn't distract them too much.
Yes so true !
More and more I see blender development going on, more and more I'm amazed by choice and strategy made by Ton and his team. See from the outside a lot of thing may look silly but these guys really know what they are doing. And I'm totally confident in them.
That's why blender is evolving so fast and so good.
Another thing pointed at in the first video of Andrew is why blender isn't compatible with commercial plugins like realflow or rayfire . I think the main reason it's because there are no easy bridge in blender that will make these application compatible (correct me if I'm wrong) . And that's something that stop middle sized studios to use blender.
In general one problem is being compatible with standards but not much in the UI side but in the workflow/pipeline side. That's a big thing that's stop a technical director who draw the pipeline for a studio or a project from using blender.
But I'm sure that will change one day or another.
For now blender is very capable and I see more freelancers and small studio using it.
Also, I've seen lot's of animators, modelers, layout artist learning blender on the go because they were forced to when working on a project. It wasn't much difficult for them to learn it. In fact generally they are quite impressed with the design of the tools compared to maya or max.
Ha ha. That is what I think too. I'm not a great artist and I would not claim to be but Blender just lack in many things.
I have not used other software much, actually I use only Blender for most of my time making 3D (just for a hobby). And I am always frustrated at the seemingly incomplete stuff. Texture painting, sculpting, simulations, hair (especially rendering it), cycles... all left something to be desired. I don't really use them to many of my work (except hair), I just dab at them just for fun and see how they work. For me, they are experimental features to experiment and play with; not for serious work and projects. There is no much tools inside Blender to make them easier and fun to use. They are there but their use is severely limited and difficult. They need vast improvements in terms of tools to improve workflow.
Actually the secret Autodesk doesn't tell you is that their tools don't get wholesale used for major projects much either. The big boys (Pixar, Dreamworks, etc.) use their own in house software, and while the others use Maya and XSI (and very occasionally 3DS Max) they always extend them with their own tools.
I'm currently working on a feature film using XSI and am barely touching the native interface, instead using the in-house tools that extend it.
Blender's interface is already streets ahead of the main contenders in most areas. Where it's really lacking is in areas Mr. Price has never even heard of.
To me too, Blender's interface work just fine. Yes, some improvements could be done here and there but it works well enough. In basic and medium (and even more complex) modeling Blender is very fine. But when it come to some more advanced tools like hair, or cloth or texture painting, it just way below par.
I always have known about this in house softwares. You always read them in magazines and interviews. Pixar for example has their own cloth simulator that they have used for animations like Monster Inc. and Incredibles. Dreamworks boast their own cloth simulator too for their animations like Brave and even have a custom clothing modeler when they make clothing and shader when it come to rendering them.
Just admit it. Most of the amazing things we see in these hollywood stuff is not out in the public. We can dream about them but it is very unlikely that we could have those things soon.
Yeah, sometimes we forget it's a tool, a TOOL for crying out loud, to making your dreams come true. hang tight.
In my opinion, this survey is just a joke , like the other ones...
For me it miss basic questions , like :
How long do you use blender ?
Do you use it professionally or just hobby ?
have you tried other 3D software before ?
what make you use blender instead of them ?
-open source / free software
-usability
-all tools in one package
-other
do you prefer use another software and why ?
what to you think of blender :
do you think blender is ok for what you do with it ?
do you think the UI could / should be improved for what you do with it ?
do you think blender is hard to learn ?
what can be improved to make it more easy to learn ?
-UI
-tooltips
-documentation/wiki
-other
what in your opinion blender lack to be more used ?
-usability
-bugfix
-improving features that are already there
-missing feature to match other software
-good reputation
-UI
For me these kind of question can lead to interesting answers, and maybe show where could some efforts can be made.
In fact, I think you made some good point in your videos , but I can't really trust you because you have your how idea of what is good, and you don't care much of other thought.
You just want to prove you are right, and it's easy to found people who agree with you.
I think all the energy you put in your videos and your reflection will lead to positive things and maybe improvements, but it could be better from the start if you would be more open-minded in the beginning...
sorry if I'm offensive but I'm a bit upset with the way you communicate your ideas...
You aren't alone.
I agree with you 100%. Moreover the fact Andrew doesn't bother reply (or read) to the hundred of comments he triggered here, on cgsociety and ba, shows his one-way communication habits.
It seems more a lecture on usability then a sincere will to improve the software that translate to code and bug reports.
As ignavaler wrote few comments above this, the problem with blender is not the UI (that certainly can be improved) but simulations and display of complex geometry. These are real problems that hinge wide adoptions, especially the viewport double side opengl problem make it impossible to use it in anything slightly complex. The Fast Navigate addon helps but not solve the problem (this is going to be addressed anyway in the near future)
The particles sytem / hair, fire, and cloth should really be the focus to bring blender to a professional level.
For wide adoption marketing is also necessary, it's clear that few studios use it because few people are capable, and few people use it because few studios ask for it.
To break this chain blender should be the de facto tool for some "niche" market like game creation.
Eli
It's not just the 3D window that slows down. Memory management seems to be an issue throughout. (I did notice that the first video Price presented mentioned the issue of viewport slowdown, for what it's worth.)
I have a file that slows down at only 6000 objects or so, even with no 3D window open. Outliner screens slow the system to a crawl with this many objects, especially when each object is loaded down with modifiers, drivers, parenting/grouping and so on. I don't have even the first idea how much effort it would take to fix such things but I know it isn't trivial.
Agreed that the UI can be improved. Uncertain where I would prioritize it in the larger scheme of things. Once some coders at the Blender Foundation get involved in the podcasts and other presentations, the air should clear a bit.
I thought he did a very thorough job of explaining that it wasn't about "proving" who was "right." On many issues in the past I have been one of Price's detractors, but not on this. I'm not even certain how much more clearly he could have made his points that all this was just a beginning, one person's opinion, etc etc, and I think he has a healthy level of awareness of his own limitations and tries to communicate this.
Of course he's not Jonathan Williamson, but what separates them is a matter of experience, polish and some level of innate gifts on Williamson's side (it's astonishing to me that Williamson is only in his early twenties, given his abilities on the microphone and as an instructor.) My best read on the situation is that Price's intentions are not something at fault here. Let us not attribute this to malice when it can more readily be explained by other factors.
I certainly noted his disclaimers but to to me that doesn't change the feeling that this is just more grandstanding from a marketing guy.
Why isn't he proposing changes through regular channels like the blender funboard mailing list? Why is he putting out videos with sensationalist titles like "Why Blender is Broken"? It looks to me like an obvious attempt at self promotion. I predict that if any of his suggestions make it into trunk, even the ones that have previously been proposed by others, he will claim the credit in a similar loud, bombastic way.
To me, he has lost what little credibility he had.
Perhaps the truth is somewhere in the middle, or you could well be completely right and it wouldn't be the first time I've been completely wrong. I don't know enough about the normal development channels or how the people who work directly with him at the conference or elsewhere regard him personally, so I take your point as someone who is more plugged in to all this than I am.
Rather than 'what do you do?' in blender, 'what would you do if it were usable?' would be more appropriate to development. I'd do a lot of simulation if simulation were robust, but it's not so I don't.