Top 10 Reasons to Love Blender

I think the keyword here is consistency (which is a notable weakpoint in Blender at the moment).

At the least, all common functions should be shared between every window type or mode where they are relevant (that would include not only hotkeys, but tools as well).

For instance, UV editing should ideally give you every selection and transform option that edit-mode in the 3D view gives you (unless it explicitly can’t work with no Z axis). This should also be reflected in the code itself by having all relevant functions shared between said modes.

Another thing, Blender currently has multiple implementations of curve objects (bezier curves, curve widgets, grease pencil, and masks). It would be good to see just how much the code can be unified (so, if possible, largely make each object type a variant of each other).

Blender is very good at doing different things, but at the very least those different functions could be better tied together, on both the user level and the code level.

That seems like a solid assessment. Thanks, Ace.

i won’t really list 10 reasons because i haven’t used other programs but…

  1. the unwraping/modeling is VERY fast
  2. layers and grouping is fast as well, very convenient
    3.supports many many formats
    4.comfortable for painting and sculpting once you get used to it.
    5.scene management and materials is simple and easy
    6.has mesh cleanup and automatically detects errors in the mesh which lots of other programs don’t currently have
    7.blender is JAM PACKED with stuff. you can move your entire pipeline to this small 200mb program.

Glad you could made the switch, but I couldn’t, even if I tried hard.

I even took notice of all the stuff I changed through the months I was trying to use Blender for some projects.

Here’s the pdf with what I changed, it should be updated with the latest stuff I changed.

There are many functions that Blender lacks, so for me it’s really a downgrade. There’s no workplane, the 3D Cursor can’t replace Cinema 4D pivot point editing, the outliner is almost unworkable (scene organization is a pain), no wireframe on shade, clunky manipulators, disorganized UI (most of the stuff keeps rearranging and you can’t find where that button was, because it moved or is not available in that particular mode), and many more. But if there is a “feature” that is unacceptable for a 20+ years old software, that is the undo system. If you use it in Object Mode and by mistake you do something, you’ll lose what you have previously done in Edit mode. I can’t believe someone thought of this way of using the undo function, unless it’s the usual way of making the features more Blender.

I want to address some of the points Ace Dragon made in his previous post, because I believe they are misleading.

I don’t see the problem on other major 3D software to use QT for their UI, like it’s some kind of fashion in the industry that everyone follows for no reason, just because you think that Blender has the best customizing capabilities because you can create new windows simply by dragging them or that you can script your own menu or buttons?


Look how ugly Cinema 4D looks with Blender Layout, but it’s possible to do it quite easily, though, I couldn’t reproduce everything because of obvious differences, but there’s no code involved, so well, moving some windows around and spending a month to code a new tab menu, is not really what I intend for customization friendly.

The 3D Cursor is one of that “weird” things, it’s considered both innovative and a major Blender selling point, but I have to disagree:

As you can see from the video, the pivot point in Cinema 4D can dictate where an object is created or how you can position it in 3D space (with the parent-children feature) , and it’s quite easy to reorient in the correct position.

At the moment, Blender 3D Cursor feels archaic on how it works, it needs a selection to be positioned, and the manual movements are imprecise, you can click on the viewport but you can’t snap it to the various components, so what’s the point to assign it to the (default) LMB if you have to use anyway the RMB to select your position (well, if you want precision) and then SHIFT+S to actually move the 3D Cursor where you need it?

I think I’ve said it another time, but I feel it needed to be said again, this time with a video proof that explains why I don’t see the 3D Cursor as a selling point.

This post is not mean to bash anyone or Blender, it’s just a reality check, considering that other software are doing what is believe to be the peculiar things that make Blender different, way better than it.
I don’t see why Blender development should follow the same path that it has been proved not to be optimal for the success of software.

As on the other thread, people with suggestions are completely ignored, but that’s fine, waiting to see the final product (101 and 2.8) so that fixing the stuff would be easier with all already in place.

I never suggested that QT is poor quality, I only mentioned that there’s some out there who would consider Blender’s UI as a poor piece of design simply because it’s not QT-based like almost every other app. Though there have been complaints by Autodesk users on CGTalk on how the Maya UI seems to get a ton of bugs whenever the library is upgraded (and that’s another thing, Blender’s custom UI code means the devs. are not constrained by whatever limitations or bugs a UI library might give them). I know there’s the old piece of advice where it’s said to use 3rd-party libraries where possible when programming, but the idea of rolling a custom solution for various areas give the benefits of complete control, less time spent keeping everything up to date, and a much smaller file size.

Also, the 3D ‘cursor’ in the C4D video took more than a minute to set up (and I’m not sure if it’s worth making Blender users do that if only the existing cursor could get some improvements like orientation/scale attributes, N-key panel options for the placement operator, and a placement control other than left-click).

Also, i do agree that Global Undo has crappy design, which more or less needs a complete rework because of its origins as an ugly hack back in Blender 2.35 (and unfortunately, no one seems to want to give it so much as the ability to do reverse operations, as the current method of copying the .blend file adds a gigabyte of memory in large scenes when doing so much as change a value).

By the looks of things, Blender 2.8 will take care of a number of usability issues (though some like the Undo system and Multires have no timetable for an overhaul at the moment).

Where is the Qt-discussion stuff? I remember a thread on BA, but I haven’t seen it discussed elsewhere.

Actually, that was not what I meant with the video, you don’t need to set up a fake 3D Cursor to do that, I showed that you can create a new object aligned to another already existent on. I mean, every object has its own pivot point that acts like the 3D Cursor would work in Blender, if you look at some of the examples you’ll see that, also, you can use the Workplane to do the job, there are many possibilities to position an object in the scene.

Even if you really wanted to use that null, you can easily save it on the Content Browser to be used in another project (drag and drop), or save it in the default scene and you’ll find it every time you launch Cinema 4D.

To answer your question, I was using Blender simply because it’s free and legit, nothing more, but currently I’m not using it, I’ve found Softimage Mod Tool 7.5 to be a perfect budget solution (as long as you work on a non commercial project), the viewport supports millions of polygons without problems, and especially with what I’m working on, the slow Blender performance is a big problem, not only, the inability to have colored wireframes it makes the work even more frustrating, with references pictures of complex objects you won’t see much, and also, I couldn’t figure out how to turn off the grid on the orthographic views, and it appears that it’s impossible to do.


I’m planning to give 2.8 a go, but in the meantime I need to get stuff done and Blender, unfortunately, fails to deliver.

Regarding the viewport. Yes, that is one of Blender’s biggest weakpoints right now (but fortunately, one of the major points of 2.8 is a new engine that will have far better performance and better features than the one in 2.7x).

If you’ve seen the video from the Blender Conference, you will see that one of the things looked at was a better wireframe shader.

I watched your vid, and I’m still stumped as to what your problem is. Blender has snapping, shift S for 3D cursor positioning etc. And if I’m honest, you could probably knock up a python plugin to have it behave the same in Bl as C4D.

Maybe I can take a guess. The 3D cursor is a good tool, but doesn’t work as efficiently or as extensively as it could. Personally, I have two dedicated hotkeys for snapping the cursor to selected/origin and another for snapping selected to the cursor. This is better but still clunky, and leaves some things to be desired, like snapping the cursor more organically, without preselecting geo, or even (especially?) keeping rotation data on the cursor.

Yes, many or all 3D cursor improvements can be hacked in with an addon. But there are some things which really ought to be part of the core design paradigm, i.e. Vanilla Blender, and not left to third parties. Also, the viewport limitations have been legitimate deal breakers for many folks who work with large scenes or high density sculpts. I’m optimistic that these things will be corrected in 2.8

I’ve seen that part of the presentation, but that is still not available, so at the moment I don’t find that convenient to having to select every object in the scene, go into the Object panel, and clicking on Display—>Wire (remembering to hold down ALT while doing so) every time you want to turn on or off the wireframe on shade.

Don’t worry, I’ll try to explain it better.

The 3D Cursor lacks any kind of data about it’s orientation, it only store the position, like if it was a point.

To set the 3D Cursor you need to have something selected, this means that if you had already selected something, you’ll lose it, considering that you have to constantly reposition the 3D cursor while working, this situation happens quite a lot.

What I mean is that the 3D Cursor should be an entity that can be repositioned and snapped to any object independently of what mode you are or what is selected. At the moment it doesn’t snap to anything at all, unless you tell it to move to a specific coordinate in space, no reposition on the fly like every other software dealing with the pivot point manipulation.

You think I wouldn’t have made an add-on, script or other kind of hack if I know how to do that? :smiley: I know that a semi working solution exists (but it comes packed with a bunch of stuff that probably I don’t need) which is Sensei Format, though it only solves the snapping issue.

Also, this kind of things tend to lost compatibility every now end then with newer versions of Blender, many tools are already being abandoned by their creators, so I don’t believe an “hack” would be the correct solution, considering it’s a core feature.

I know that many doesn’t find any problem with the 3D Cursor (and many other functions) but what I’m suggesting (and I’m not the only one) is not to copy another software, since I tried to emulate Blender functionality showcasing how an improved 3D Cursor could be used, but to implement in it new features. If you look at that add-on (remember to backup your Blender settings first) you’ll notice that is it possible to drag around the 3D Cursor and place it near a component, and it will automatically snap to it, I’m sure you’ll find very fast to use and I’ll improve your modeling experience. As I said earlier, if I could figure out how it worked, I’d be already using a similar solution.

edit:

Yeah, probably I worded it poorly, but the message would have been to not ignore what other software are doing, because they may teach you new way of thinking about your workflow and even help you in the creation of useful tools, like the Pie Menus.

A recent 2.8 commit has made the multi-object editing mode the default (ie. the alt key is no longer needed, and people who just want to edit one object will now need to have just the one selected).

The reason why it was tucked away like that in the first place is because 2.7x had a policy of keeping backwards compatibility in .blend files, usage, and documentation (but that will not be in effect for 2.8).

No, that’s not really the reason why multi-object editing was hidden behind the Alt key. Improvements to the UI should be part of the general development, we can’t avoid breaking documentation or user habits sometimes - it just has to be for the greater good (though, we are careful and try to bundle UI changes, especially now that 2.8 is around the corner).
.blend compatibility wasn’t a big issue here either.

The reasons why it was a bit hidden were mainly open design questions and limitations. How should it work when editing modifier or material properties when each object can have completely different modifiers/materials? How should it tell the user that its changes are applied to multiple objects, not just to a single one? Which settings make sense to be applied to multiple ones, which don’t? …
We got to some agreement on the design in the 2.8 UI workshop, you can read about it in the write-up. It is definitely something we want to see tackled during the 2.8 run.

I have no other legitimate experience with any other 3D application. I’ve been hobbying with Blender for 10 years now (give or take) …getting pretty serious about it for the last 2 years. It’s really the only 3D app I’ve used for the most part. That being said, I’m too old to be a fanboy LOL.

I totally agree with this statement. Cursor on LMB is very frustrating. Drag box click would be ideal.

Just my 2 cents.

Emo

XRG gave some great responses to Derek’s posts.

Keep in mind a lot of professionals use graphics tablets and graphic monitors. These artist are interacting with a screen, not doing keyboard olympics/twister. The same kind of vague reasoning you are using was used to attack Blender’s development of Pie Menus.

Pretty sure those same people are now happily using pie menus (which btw is found in a lot of those “industry standard” apps you are dismissing).

Interface driven control schemes can be just as fast as hitting a few keys on a keyboard, it all depends on design and feature sets. Nearly all major industry standard apps feature hotkeys, but they are also not stupid enough to forsake the GUI either.

A former art director over at EA who uses Maya can work at the speed of light (obvious hyperbole) using a cintiq and pie menus. I have yet to see a blender artist work as fast and as good as this guy did. What made it fast? The pie menus were nested, so it was pattern based in workflow.

Modo artist like Tor Frick can generate mesh items and model based on just a few macros and some pie menus. Smart design choices in Modo have Extrude, Inset, and Bevel all being the same tool with context sensitivity enabled. Smart design is faster than a literal ton of hotkeys for every single possible function available.

Its about perspective and mastery. One of the major reasons Blender is not embraced by the industry is due to is licensing and development structure. Commercial applications offer more security and the ability to really protect anything they develop in house for the 3D application in question. They also get access to tools and technology that cant go into Blender because again of the licensing.

Blender’s licensing is a double edged sword. It makes it more accessible for the average user, but not necessarily the most logical choice for commercial studio.

“But convention only proves popularity, not usefulness or efficiency.”
No. Do you know what makes something intuitive? Familiarity. Familiarity is also a factor in efficiency. Convention is efficient, especially when you build upon it and doing so knowing that your job pool is full of artist that can get up and running right away…especially between applications. Training cost money, time is money. You have to be efficient or else you lose money and that makes or breaks both small and large studios.

Maya’s navigation and selection method are conventional, but it is also the most efficient. It is designed with both the mouse and the tablet in mind. One modifier key, and the mouse or pen doing the rest. The artist can pick up 3D Coat, Unreal Engine 4, Unity, Substance Painter…ect and that approach will work. It creates cohesion, and thus the speed is there. Try not to dismiss that benefit.

Usability is directly tied to convention, thus intuitive design, and therefore familiarity. That’s just how it works.

On a side note, Blender is very popular, but I think you are forgetting why it is popular. Its price. Thats it. Not some great magical design thats better than everything else. The fact is Blender is FREE. FREE means high accessibility. When you have Free software in a world where the competition cost anywhere from a few hundred dollars to a few thousand, yeah it kind of drives up the demand and thus appeal. This isnt to say Blender sucks either, its just that the popularity has a lot to do with accessibility rather than the lack of convention.

Food for thought.

Very true.

Well, that’s a lot to unpack. Here goes…

Keep in mind a lot of professionals use graphics tablets and graphic monitors. These artist are interacting with a screen, not doing keyboard olympics/twister. The same kind of vague reasoning you are using was used to attack Blender’s development of Pie Menus.

I use a Cintiq at work, so this confuses me. How is it that touching my pen to the screen, rather than using a mouse, negates the utility of the keyboard? Blender does a fair job of keeping hotkeys left-hand accessible, so not sure where the twister comment is from, but will agree that hotkeys in some other apps are pretty dodgy and inefficient. Also, I’ve got nothing against pie menus, and actually really like them. But what I like most is that you don’t have to actually select the items with the cursor, but instead can still nest hotkeys for faster access.

Pretty sure those same people are now happily using pie menus (which btw is found in a lot of those “industry standard” apps you are dismissing).

“I hate all things found in ‘industry standard’ apps. Everything. Blender is such a sellout, what, with its polygons and vertices. It even has buttons, and menus, and text (stupid Autodesk). Amirite?”

So, at this point I want to say that my generalizations (intended as brevity) and snark have caused a divide in what I meant and what you’ve taken me to mean. This is not your fault. I should clarify: I’m not against a feature or workflow simply because it was first implemented by a commercial app. Maya != Bad, Max != Bad, etc. However, I’m not willing to adopt a feature based solely on the same metric. Maya != Good, Max != Good, etc.

Interface driven control schemes can be just as fast as hitting a few keys on a keyboard, it all depends on design and feature sets. Nearly all major industry standard apps feature hotkeys, but they are also not stupid enough to forsake the GUI either.

On this we can agree, generally. Where we probably disagree is on whether or not hotkeys are good enough in commercial apps (I only use Max/Maya at work, but on them I’d say not) and whether they’ve actually adopted, curated, and consolidated their a UI schemes effectively (same qualifier as before, and I’d say not). There are some good things, and some really mickey mouse things about all apps.

A former art director over at EA who uses Maya can work at the speed of light (obvious hyperbole) using a cintiq and pie menus. I have yet to see a blender artist work as fast and as good as this guy did. What made it fast? The pie menus were nested, so it was pattern based in workflow.

I’d be interested to see how he works. I find that Maya takes roughly 30% longer to do any modeling than what I get done in Blender. It’s a lot more cursor movement and a few more clicks to do any one thing. This is mainly why I will not model in Maya anymore. If you have something game changing, I’d love to see it.

Modo artist like Tor Frick can generate mesh items and model based on just a few macros and some pie menus. Smart design choices in Modo have Extrude, Inset, and Bevel all being the same tool with context sensitivity enabled. Smart design is faster than a literal ton of hotkeys for every single possible function available.

Never used Modo, but I’ve heard good things. You’re not going to find me complaining about better tools if they really are better and also fight bloat. Again, there’s nothing wrong with pie menus.

“But convention only proves popularity, not usefulness or efficiency.”
No. Do you know what makes something intuitive? Familiarity. Familiarity is also a factor in efficiency. Convention is efficient, especially when you build upon it and doing so knowing that your job pool is full of artist that can get up and running right away…especially between applications. Training cost money, time is money. You have to be efficient or else you lose money and that makes or breaks both small and large studios.

Sorry, convention describes how something is done and its pervasiveness, irrespective of the why. In my family, my grandmother would cut the ends off the roast. Following convention, my mother did the same. One day my wife asks me why I cut the ends off the roast. “It’s convention”, I say. But she insists, “no, why do you cut the ends off?” I call my mother, and she doesn’t know the answer. So I call my grandmother. It turns out she cut the ends off because they were poor and could not afford a larger pot. Should I continue to cut the ends off the roast?

Convention does not address the why. Preserving convention for its own sake is idiotic. We should be diligent about doing things as best we can, not about maintaining compatibility with something stupid that a lot of people happen to be doing. That said, remember, all apps are doing a host of things very well and if they have tools/workflows that are worth porting, we should be porting them.

Maya’s navigation and selection method are conventional, but it is also the most efficient. It is designed with both the mouse and the tablet in mind. One modifier key, and the mouse or pen doing the rest. The artist can pick up 3D Coat, Unreal Engine 4, Unity, Substance Painter…ect and that approach will work. It creates cohesion, and thus the speed is there. Try not to dismiss that benefit.

Maya’s navigation is Maya’s. I use my tablet to navigate seamlessly in Max and Blender as well, who’s navs are more similar to each other than to Maya’s. The problem with ‘one modifier key’ for nav, is that you now have cut your modifier keys, for other uses, by 1/3… for no actual benefit. Not exactly a triumph.

As far as selection goes, it would not bother me to switch L/R select. I switch between apps all day, I already know it, it’s not an issue. However, I’m adamant about keeping select and tool functions separate. L/R makes no difference.

Usability is directly tied to convention, thus intuitive design, and therefore familiarity. That’s just how it works.

These are all related, but in a way that doesn’t really pertain to our conversation. As alluded to before, convention is a result of innumerable factors, of which intuition-friendly design is only one, and a subjective one at that. Just because you’re used to something doesn’t make it right, efficient, ethical, etc. Quantifying familiarity is a meaningless metric by itself. But, again to be thorough, I’m not implying that the prevalence of a convention somehow diminishes its value. Nor am I ignoring the obvious fact that something of high value has a greater chance of being popular than something of lower value.

On a side note, Blender is very popular, but I think you are forgetting why it is popular. Its price. Thats it. Not some great magical design thats better than everything else. The fact is Blender is FREE. FREE means high accessibility. When you have Free software in a world where the competition cost anywhere from a few hundred dollars to a few thousand, yeah it kind of drives up the demand and thus appeal. This isnt to say Blender sucks either, its just that the popularity has a lot to do with accessibility rather than the lack of convention.

I have not forgotten. I know my own roots in the industry, and I know those of my peers. However, what got me in trouble in the conversation was the offhand comment that Blender needs to worry about Blender, not other apps (+ snark). By that I mean to say that Blender should make changes which make sense for Blender. Some of those changes, absolutely, should be to incorporate more functional UI, and to repurpose ideas from apps which really are leading the pack. But also to say Blender should not be attempting to clone antiquated paradigms, and maybe also a hint that those same paradigms should take the plunge and reform.

TL;DR

We probably agree more than I lead you to believe. “Conventional” says little about function without other metrics.

Food for thought.

nomnom

Its this same debate again. Hasn’t all this all moved on a bit now? Blender is already being used in serious production environments and several of the projects were talked about in detail at the last conference… Whole feature films are being made on Blender based pipelines.

I also use Blender with the default key map and hot keys and love it. Its lightening fast and natural to work with once grasped.
Tangent Animation advised people to learn the native key map when getting them up to speed with Blender as it worked better that way.

I use Blender and Maya right now side by side at work and jump between the two constantly. Blender is my main modeler at work now. Its never been a problem.
Its gaining a !ot of respect in the industry now and there is a lot of interest.

One thing not taken into account though is that when Blender became open source, there were number of applications out there (commercial and free) that were either comparable to Blender or superior (both in terms of features and workflow).

  • Art of Illusion was one of the early free software solutions and had several features Blender did not (like GI), that project’s about dead now.
  • Equinox 3D back then was able to go toe-to-toe with Blender on features (it’s still going, but rather slowly and is now well behind).
  • Bryce was one of the solutions for Hobbyists at one point and considerably easier to use (it stagnated when DAZ picked it up).
  • Carrara was basically superior to Blender in several ways and was one of the major hobbyist 3D apps. (again, a solution that stagnated under DAZ’s ownership).
  • Cheetah 3D (does anyone use it these days)?
  • Aladdin 4D (another low-cost app. It also tried to go Open Source several years after Blender did but failed)
  • Truespace (was geared to the game industry but had access to render engines like Vray, but Caligari went under and Microsoft couldn’t make it work).
  • K-Ray also had comparable features and its rendering solution was arguably superior (due to its POV-Ray integration), supposedly it had a poor release at one point (one from which it never recovered).

The BF must’ve done something right because Blender more or less killed off the hobbyist tier of 3D apps. (or at least reduced their relevance to a minimum). Sure, there were a few that had quality issues, but I don’t think that would’ve affected them near as badly if Blender was as shot through with bad decisions as some allegations state.

When people say Blender has a steep learning curve I can’t help asking if they have seen Maya, 3ds Max, zBrush, Cinema 4d and other software in this field. I have been using Maya for around 10 years and then switched to Blender because it better matches my needs (and Autodesk made Maya crash every 2 seconds in recent years - there is always that as well). I have to say it took me surprisingly short amount of time to switch. I expected it would take a lot longer for me to fully learn Blender to the point I feel more conformable working in it than in Maya. So I think there is nothing wrong with the learning curve. Except if one is new to 3d.

The other thing I personally love about Blender is the UI that is so complained about. I really love the ability to have what I want where I want it whenever I want it. It takes a fracture of a second to have whatever panel wherever I need it and they never overlap one on top of another - how good is that?.. Sure there is space for improvement, but it is pretty good in my opinion. I loved Maya’s interface as well - I learned to use it first and I never knew anything different, it became standard to me and I don’t think there is something fundamental wrong with it, but Blender’s interface is in many instances more advanced and more fun to use in my opinion.

The other point I feel is left out is how stable Blender is. Everything just works. You can use all it’s functions one on top of another and it would not crash and if it does crash you have autosaves by default. I just love how you can trust it.