Blender Compared To the Major 3D Applications
LetterRip writes "Recently TDT 3D published a comparison of the major 3D digital content creation applications such as Maya, 3DS Max, and XSI, and of course Blender. Blender came out surprisingly well, although it definitely still has some weaknesses."
I've always thought Blender to be a solid but completely useless application because for whatever reason, the developers created the most heinous god aweful UI known to man. It's a freakin eyebleeding headache that leaves one happily shelling out the hundreds or thousands of Dollars for a modelor with a usable GUI.
It's a shame. Because Blender could be a contender, but since the developers live in their own little world with the attitidude that their app is made for a "certain group of people and not everyone", the application is basically a sick joke. If you're looking for a free 3D package and don't care how painful it is to use it, Blender might be for you. Otherwise, go with Maya or Lightwave. Also, Modo is a good modeler with a great UI, if you just want to model and not animate/render.
First off, the name is a tad misleading, it's more "Major 3D Applications Comparison (including Blender)" as opposed to a focused against Blender - but that can be easily forgiven. What is good is that the author has used a wide range of industry standard tools (Maya, 3ds max and Lightwave etc) in everyday tasks, so it's not a fanboy style review where the outcome was and always would be 'Blender best'.
Blender is maturing well, especially considering if you look at the progress they've made since the code was first open sourced, and I'm confident that they will be able to continue this progress in to the future. I won't comment on the feature to feature comparison, I will just reiterate what the author said early on in the article - it's the artist not the tools - and in this case the an open source and free tool is sufficient to create some stunning art. Check out Elephant's Dream to see an example.
And my one note to those commenting later - the interface. Many people complain about the interface, how it's difficult to learn. Unfortunately, many of these people are trying to 'learn 3D' over the weekend - and I'm sorry, that won't happen, regardless of the package. To become truly proficient in any 3D software package takes a long time.
Overall however, I'm glad that Blender has progressed so well, and look forward to seeing it's progress in the future. Without it I would never have begun my exploration into 3D at all.I also see many people compare the UI disaster to that of GIMP - I don't think that's really an accurate comparison. Read reviews of people who have actually mastered the Blender UI and they will confess that once learned it is very effective. The author in the article also states this, with him saying that it's a "Fast workflow, (but) can be more intuitive".
I'm not saying the Blender user interface is perfect, or that improvement can't be made, merely that even in its current state, once learned and mastered it is a powerful tool.
Comparison lists like this are very misleading. The devil is in the details. It's not about how many buzzword columns an app can check, it's about whether the app has had a userbase that has thrashed the software through years and years of real production work and had the software evolve into a powerful tool. It's very easy to add a feature in a 3D app. Most interested hobbiest weekend coders could whip up most functions found in any given 3D app. Whether that feature is production ready is another story. Going by the list it would seem that something like Blender of C4D is on a par with Maya and Max. They're not. They don't have the huge in-depth expert user communities and the benefit of thousands of users pushing them to the limit day in and day out. If you're choosing a 3D app, talk to some real 3D artists who have been in the industry for years and depend on their 3D application system for their living. Don't compare checkbox lists.
Some people don't like blender's UI like some people don't like vi.
Once you spend the time (say a day or two) it's really quick and productive. The UI is a major asset.
I don't therefore I'm not.
pushing polygons is wrong. It's a short term solution to a long term problem. Keeping to a poly budget is grunt work that should be done by the tool not by the artist. Programmers who think bitrot is bad should have a look at the "asset rot" of 3d models. The deprecation of graphics assets is so fast that it's a stretch to use the word "asset" to refer to them at all. A character model from a AAA title from six months ago has some value, to B quality games, but mostly none of them are reused.
For 3d models to earn their name as "assets" they need to be created with infinite resolution. This is not hard. Constructive Solid Geometry is a well understood technique for modeling and is typically used in CAD applications. An object described even at what would be considered a course level of detail of typical CSG modeling is orders of magnitude higher resolution than the typical game model. What's more, they can be incrementally improved, whereas the b-rep that is typical for a game model today is a one shot affair.
Today, the vast majority of CSG models are created for raytracing architectures which, although they give stunning results, are too slow for realtime applications such as games. Even the attempts to create realtime raytracing systems are aimed at rendering b-reps because of the opportunity to perform an acceleration stage which greatly reduces the scene complexity. B-reps are a win for realtime applications, but just because your application requires a b-rep doesn't mean artists need to get their hands dirty pushing polygons.
Two solutions exist which can render CSG models in realtime.
The image-based rendering algorithms with implementations (such as OpenCSG) which take advantage of z-buffer and stencil-buffer hardware in popular GPU cards. The image-based systems are more compatible than raytracing with the current 3d graphics rendering state of the art, but still require intricacies to integrate that have prevented them from appearing on the market.
The other option is the automatic generation of a b-rep from a CSG model. This has the advantage that it requires no change to the 3d graphics rendering in games and the algorithm can be parametrically tuned to produce b-reps that are fast for different applications. For example, the same CSG model can be used to generate b-reps of different Level Of Detail for when the object is close up vs far away, or for pre-rendered applications such as cut scenes or trailers.
It just happens to be really hard.
How we know is more important than what we know.
I use 3DS Max and Blender myself and I definitely have to say Blender's UI is horrible and I mean that. I don't know what the developers are thinking, but damn they need serious help in making that user interface more friendly. The only time a I use Blender is for when I do some modeling and art for open source games and mods that have and export function for Blender (such as xmesh in Vega Strike). Other than that its 3dS max for me, though from what I've seen Maya is great too. Lightwave used to be a great program, but it seems to have performed worse in this comparison than I'd have thought.
Again, I can't stress enough what a pain Blender's UI is compared to these other programs. I normally try to recommend FOSS programs to people, but this is one of those programs I'd only recommend if the person couldn't afford one of the other programs here.
I think Blender's UI is very efficient, not friendly but efficent.
Its good that Blender is looked upon a viable option these days!
I've used 3DSMax and Blender extensively though learnt to be proficient in Maya.
Blender's interface is difficult at first but once learnt I found it more efficient for many tasks - especially rapid mesh-modeling. Moreso, I found Blender's interface much easier on the hands over long periods of working, namely because it prioritises keyboard useage over mouse, has one-key accessible menus (3DSMax has terrible problems here) and distributes mouse input over all three buttons. RSI is a real problem in 3D modeling/animation related work and it's here I think Blender has the one-up. Blender's many keyboard shortcuts need to be learnt for Blender to realise it's real capacity as a tool.
Are you referring to TDT3D as pro open source, or Slashdot? The article isn't *by* Slashdot. Slashdot also posts articles which lambaste open source projects or often put closed source projects in a good light. While Slashdot is definitely pro open source, the general community isn't blind either. We recognize Blender has flaws. What makes us pro open source is that we are willing to give it extra weight in our personal preferences simply due to that fact. We know Firefox has some serious issues and don't ignore the fact she's putting on a little extra weight - we still love her anyways. However, considering '07 *still* wasn't the year of Linux on the desktop, we also recognize the face that others - the majority in fact - don't share our appreciation for open source, and found it pleasantly surprising how Blender still ranked up with the best of them even when it's open source-ness wasn't taken into consideration. I can tell you this much: after struggling to get used to Blender after learning 3D Modeling on Maya, *I* am surprised Blender ranked well.
"A witty saying proves nothing." - Voltaire
... the fact is, Maya's UI sucks. 3d Studio's UI sucks. Power Animator's UI sucks. Cinema 4d's UI sucks. Rhino's UI sucks. Nendo's UI sucks. Silo's UI sucks. Softimage's UI sucks. Lightwave's UI sucks. ALL 3d application UIs suck ass until you get up on the learning curve, because all 3d applications do fundamental things differently.
I "grew up" with 3d Studio MAX - Maya, Rhino, etc. all make my brain BLEED - not just because they're Not MAX (the way The GIMP is Not Photoshop) but because they're Not MAX the way Emacs is Not Vi.* 3d is HARD, dammit - hard to code, hard to learn. I'm picking up Silo pretty fast, but mostly because Silo seems to make it a point to do intuitively a lot of really basic things that are a massive assraping pain-in-the-ass to do in 3d Studio MAX.
That said, I have only two complaints about blender.
1. The open/save dialogue (as has been said elsewhere) does suck a load of flaming ass. Weapons grade Anthrax ass.
2. Like almost all modern 3d apps, you need a three button scroll-wheeled mouse to get ANY use out of the app at all. Which makes the app unuseable if you're using an input device, with, say... two buttons. While that's all fine and good, it ought to be fairly trivial to poll the bus, check for pointing devices, and pop up a nice little "FOR OPTIMAL USE PLZ ATTACH 3-BUTTON SCROLLWHEEL THINGER. [DIAGRAM WITH FUNCTIONS OUTLINED]" if inadequate hardware is detected. If Blender did this, it would put it well ahead of commercial apps for first-impression useability.
* Pick any two apps that generate the same results but go about it using completely contradictory and counterintuitive methods. Same principle.
Robust has more options than good.
:P
Much the same way 3d Max's implementation of mental ray has a LOT more granularity and options than its default scanline renderer. The problem for people like me is that you can flip a few switches in the scanline, wait a couple of hours, photoshop for ten minutes, and get something that would have taken you DAYS of messing around with mental ray.
Just because you can learn to use a harder system, doesn't mean it isn't harder. Since car analogies are popular let's go for another one: The difference between a stick and an automatic. I drive a stick, always have, always will if I can get away with it. I'm an ace at it, never causes me any problems. I don't even think about it, I just do it. However there is no way I can argue that it isn't harder than driving an automatic. There is simply more to do, more to learn, and I remember how painful learning to drive on a stick was. It certainly isn't an unusable or unlearnable system, and once you've learned it well you are plenty efficient with it, but it is not as easy as an automatic.
Well same deal with interface. Not all choices are arbitrary, it isn't as though all UIs are created equal. There really are things that make it easier and things that make it harder. Just because you spent the time learning it and now don't have a problem, doesn't mean it is easy.
Also, some things are easier for computer people than non-technical people. I've seen many things that make assumptions in relation to a level of understanding of how computers work that if you lack that, you are really screwed. A good example would be regular expressions. If you understand how a finite state machine works, regular expressions are likely to be easy for you. I remember when I learned about them and as soon as I was shown the syntax, a light bulb went off. I found them quite easy, once told how they work, but then I'd been tinkering with FSMs before I knew that term. However a non-technical person could be hopelessly lost on them because they just don't understand the logic behind them.
Trying to defend bad UIs in OSS does no good for anyone. People don't want excuses and they certainly don't want to be told that it is their problem.
I dont see why that is such a big issue ?
Apple's implementation of X is fast, stable, easy to install and blends very well into OS X. It's freely available and you can download it from apple.com.
Sure a native application would have been even better, but this is really very far away from "unusability".
Probable impossibilities are to be preferred to improbable possibilities.
Aristotele
Uh, I'm using blender on Mac OS X, and it does not use X11. It's implemented using OpenGL, so the interface is completely custom, but no need for that X-server to be open.
blender's user interface is a bit rough to get used to, but once you're used to it, it's very efficient and fast (I once attended a Maya course, and always screamed out in terror when I had to hoover around in 2 levels of menus to access a simple thing like extrude -- in blender you just press 'e').
I'm an active blender's user. And i find the interface very productive too..like any other "real" blender user. However, we are in a big world and , yes, THERE IS other 3D apps over there.. People spent a lot of time learning these apps, and these apps tried to UNIFY their UI. Actually,If you do ctrl+c it does copy something . When you alt+mouse , the 3D view does something.. Well, not in blender. And that's the problem, and that's why the next version is all about the new CUSTOMISABLE UI :)
Take care of blender.. it's one of the best real open source projects actually.
congrats tom, you're a big guy !
I'd like to know how many of you complaining are actual 3D artists (of some sort, be it pro/student/hobbyist) of even an intermediate skill level, who actually made an honest effort learning Blender. No, opening it up, finding your left and right mouse buttons swapped, then giving up and going back to your pirated copy of Maya doesn't count.
I _HAD_ to learn Blender for an animated short I worked on over the summer. I had about 9 months of 3ds max experience behind me, and grasping the basics in Blender took me less than a week, using Tony Mullen's book Introducing Character Animation with Blender as well as whatever tutorials could be found online. Yes, the documentation is sometimes lacking (nothing new there, considering it's free software), but that mostly applies to more advanced features (NLA editing, linking animation files together, stuff like that) where you end up having to discover many things by trial and error (which is horrible if you're working to a deadline to be sure). In my experience there's no shortage of resources targetting beginners though. Just don't expect to become a master sitting around reading tutorials -- the reason 3D graphics in general and animation in particular is hard is because IT IS HARD. And it's hard even after you've mastered whatever application you're using. Getting past the user interface is the easy part.
At this point the only thing I could say 3ds max is outright "better" at is pure modelling, since Blender lacks n-gons, and even then I'd rather use Blender because the interface is just that much nicer. I haven't really used hair and fur much in 3ds max, but my understanding is that it's a fair bit more advanced than what's currently in Blender as well, although you can expect that to change soon as the Peach project gets underway. Textures, rigging, animation, lighting, rendering, compositing are all things I find much easier to do in Blender.
Yes, the Blender interface is "unintuitive" in the sense that it doesn't look like anything else, but compared to the hodgepodge that is max it's a shedload more consistent and logical. Once you grasp the basics (A selects/deselects all, B border selects, G moves, R rotates, S scales...) all 3D windows, node views, curve editors and what have you work exactly the same way which is just lovely. Add to that a completely customizable workspace layout and completely zoomable/pannable windows. I cannot tell you what a shock it was to go back to 3ds max and discover that to move and zoom the timeline you had to use combinations of ctrl, alt and mousebuttons, when you're used to doing all of that the same way you zoom and pan 3D windows: scrollwheel and middle mouse button. Or that the damn thing loads for a minute and a half even on a top of the line machine, while Blender takes about 2 seconds to start. :)
That's not to say the Blender interface couldn't be improved. The default keyboard shortcuts are WEIRD -- Ctrl-W to save, anyone? -- and more importantly, not customizable. The default view control options are pretty confusing as well if you're used to basically anything else ever, but these can at least be changed. I think changing the defaults (but making it easy for advanced users to change back) would go a long way towards improving Blender's image at large, but I'm not so arrogant as to assume that the Blender developers (who btw are excellent, pay attention to users needs and release new features at a mindboggling rate) aren't aware of these things.
(Struggling with the formatting here -- first post, sorry.)As I started out as a mechanical draftsman, Your post is not quite as sarcastic as it could have been. I've sharpened my rulers with a whetstone, used a scalpel to amend ink, and made my own Ink.
I can use a sliderule, also.
Why, yes, I AM a Pagan Libertarian.
Should you use text to italicize rather than em? <em> is for text you want to emphasize, rather than text you want to force a specific slant on.
That's all well and good but the fact is, you never entirely "master" a 3D application. There's just too much stuff to learn, and if you work in an environment where you use other apps as well, there are so many things you need to remember it's impossible. A good UI should be self-explanatory enough so you can find things quickly even if you don't remember *exactly* where everything is. And blender has one big fault here, which is that because of the way its interface panels are limited by size, it has to cram too much information into not enough space. So you get in each panel a hodgepodge of barely-sorted buttons which are labeled with concatenations and meaningless invented terms like "spin dup" and "innervert" and "shadbuf" because there isn't enough space to write the real term. The tool tips are a quick fix, but are not a good solution. Also, the interface is supposedly customized for speed and so is very different in basic operation (context menus, mouse selection and such) from most applications. But if you use it in conjunction with other applications - for example, when you use GIMP and Blender to texture a model - the mental gear shifting is so jarring you just get slowed down by the completely different interaction paradigm. And you almost always use more than one application in any normal professional workflow.
That's all well and good but the fact is, you never entirely "master" a 3D application. There's just too much stuff to learn, and if you work in an environment where you use other apps as well, there are so many things you need to remember it's impossible. A good UI should be self-explanatory enough so you can find things quickly even if you don't remember *exactly* where everything is. And blender has one big fault here, which is that because of the way its interface panels are limited by size, it has to cram too much information into not enough space. So you get in each panel a hodgepodge of barely-sorted buttons which are labeled with concatenations and meaningless invented terms like "spin dup" and "innervert" and "shadbuf" because there isn't enough space to write the real term. The tool tips are a quick fix, but are not a good solution.
Also, the interface is supposedly customized for speed and so is very different in basic operation (context menus, mouse selection and such) from most applications. But if you use it in conjunction with other applications - for example, when you use GIMP and Blender to texture a model - the mental gear shifting is so jarring you just get slowed down by the completely different interaction paradigm. And you almost always use more than one application in any normal professional workflow.
True, that - when switching between the two I often find myself wishing Gimp worked a bit more like Blender... And certainly there are a lot of people out there who wish Gimp worked more like Photoshop...
But this whole idea of "one paradigm to rule them all" doesn't work for me. I think it's premature. People are still figuring out better ways to do things in UIs and - here's the clincher for me - what's best is domain-specific. If you're writing, say, an e-mail application - there's a certain set of tasks that program has to do, and probably there's an expectation that lots of people with all kinds of different backgrounds will use the app. In that case, sure, familiar UI concepts are a strong asset. But even then, even in something as simple as e-mail, the emphasis on making the app "easy-to-learn" emphasizes things like hiding or removing lesser-used features, to simplify the interface.
But a 3-D modeler is a different beast. You don't expect everyone to learn 3-D modeling, not everybody needs or wants to - which is quite understandable due to various difficulties involved. (For instance, dealing with 3-D space but 2-D display and input, using a giant library of different modeling tools and techniques to create the desired effect, and even dealing with things like individual vertex placement, since the computer can't quite handle that on its own in "specialized" cases, like low-poly modeling or animation...) The task of 3-D mesh modeling is inherently complicated - at present, anyway. People, in general, aren't sculptors or model-makers. In any medium (clay, plastic, polygons, whatever) the process of learning to model isn't simply "how do I make this", but "how do I make this with these tools and materials?" So users of a 3-D modeling app have a considerably higher initial investment: this suggests a higher level of sophistication as well.
So, here's the thing about making the interface fast-to-use vs. easy-to-learn... How many times do you have to use a given function? How does that weigh against the one time you have to learn how to invoke it? Let's suppose, for instance, that I had a polygon model with "smoothing" turned on. (That is, smoothing via shading algorithms, like Phong or Gourad shading) - And I wanted to specify edges that would appear "sharp" - discontinuities in the "smooth" surface. Now, it may be simpler for me to find this functionality if it's laid out nicely in a button bar somewhere - and I honestly have no idea if Blender has such a button, since I use keystrokes for most functions like this... If this is a task I have to do lots and lots of times, it's worth my t
Bow-ties are cool.
Hello, :-)
:
:-)
cheers,
Ben (BSM3D)
I'm the Benoît Saint-Moulin (the badboy) who writed this article
First thanks to all for comments and email, I hope my article goal is reached : discuss about how improving softwares and what's can be offering us for our needs.
I'm CG technical artist and teacher in 3D College (HEAJ-Belgium) and try to stay independant into that's writed article. (sorry for my english...)
To writed this, take a long and painfull 12 months of works, I hope it's well understanded.
I read all comments, so here is few replies
- I'm not Pro Blender or not, I tried to stay indepedant into my article, I'm using Blender because there is powerful tools inside, not because it's free or not. As teacher / trainer I can say today industries interface use ""standard"" shortcut, F1 for help, well organised docs,... in Blender it's a bit uncentralised and new users never done 3D before must probably loosed to found the info when he need...
To Blender be an industries and Film used softwares, replacing maybe some old dinosaurs, he must adapting his interface to be more CG artists friendly and open to artist's today kind of pipeline working, artists say not technical geek like us.
- I'm not anti-Lightwave, I have it (9.2 dvd), used from LW 3.5 and the famous LightRave dongle until today for such task others can do ! I also writed this words on LW with LW-europe dev.support.
- I'm not cons to text interface, I really prefer text interface, that's more faster to work with than vivid Icons as each softwares icons are not the sames ! text stay the sames...like in xsi, lw,...
- This article don't to tried to compare Blender opposed to the world. Blender is include in because to industries can't simply ignore Blender potential and great work who can be done by Opensource software. (no licence troubles, no network installation issues, no per year subscription...)
- Blender have a now a place to be in this kind article and I really don't like work with Nurbs in Blender, so each softwares are good for one tasks, try to do all with one is mayeb not the best to do !
If blender is really good today, he must evolving (as free it's the best) he as somes lacks on somes BIG datas projects. I know just because I'm worked with, why somes highend 3d softwares is again today so highly priced and in somes case that's really justified ! in many not
Will it blend? http://www.willitblend.com/
-- Boycott Shell
That may be true, it may have a bad UI, it may turn new users off, but theres nothing that can be done to fix it without people who have the skill, and the will to change it.
So, in the short term, if you don't like blender's UI, go buy a package with a good UI. There is no free alternative. In the medium term, your best bet is to help to fix it, and in the long term, eventually, this issue will be addressed by somebody else.
Its a matter of how much time have you got to contribute vs how much time youre prepared to wait for Blender to get a new UI.
I mean, if i never bothered trying to edit text files in POV-Ray (oh yes, i've been there too), i would never have gained the knowledge i use daily with Blender and other 3D apps. I never once threw up my hands and said 'too hard' - i was too busy learning and creating stuff.
So while it may be 'the slogan of bad user interfaces everywhere', its also the cry of the project that is trying to be too many things to too many people, with not enough hands on deck.
Its a simple matter of resourcing and a lack of consensus - so if Blender is losing users because of the UI, well, thats regrettable, but unfortunately not addressable without compromising other goals in the short term.
Blender can, and will be, so much better than it is now - and its not like anyone is stopping anyone else from reorganising the Blender UI in a newbie-friendly way like GIMPShop does with GIMP.
Inkscape is wildly popular because a bunch of developers took the oddball sodipodi interface and changed it to their liking, not because the sodipodi developers were influenced by the UI complaints on slashdot.
And really, when you look at the size of the Blender community, does it really seem like there is a problem with the number of users?
I gots ta ding a ding dang my dang a long ling long