I don’t intend to sidetrack this topic further but will just suggest you to search around, and read the developer’s own words. If you’ve used animation layers from Maya and want to claim how Blender does it is intuitive and sufficient… I frankly can’t help you.
If you take the time to learn how the NLA editor works, it’s quite intuitive, very similar to your typical video editing software, or the video sequence editor.
I even skimmed through what you posted to find the part relevant to animation layers. Oh, look, the NLA editor is meant for that.
- Note : For Blender, it was always intended that the NLA system would act as a bit of an Animation Layering system. What’s currently missing to support this is some internal code to keep track of all the channels that are currently being animated, what values they have, and a way to set/control what the default values for these channels will be (i.e. when they are not being animated). Since this is such a frequently asked question/request, I thought I’d address this up-front here.
And the issues that would require dev time to make it better are actually quite simple to work around. If you take the time to learn how the NLA editor works. If you want to use blender for animation and are unwilling to learn how it’s tools work - you shouldn’t be requesting other tools.
Once everything nodes is added…then yes.
The NLA editor does both of those things
i clearly wont convince you. Don’t get me wrong I love blender and use it daily even though at work i have max,maya,motionbuilder, houdini, zbrush, substance all installed as the standard pipeline at our studio. (and i know howto use them) Maya is a crashy piece of… $$£*** but when it comes to rigging and animation it does almost everything in a few clicks rather than blender’s 20… animation layers are so fast to set up, they work with the dopesheet and graph editor seamlessly and just work… in blender the constraint system, drivers and nla feel clunky and over engineered by comparison. you can dismiss me as being ignorant of how blender works. I’ve been using blender out of choice since the 2.4 days over ten years. i know it inside out. it has many strengths but there is a reason most studios use maya for animation. rigging and animation are just a lot easier in maya.
see this is a better argument
They are a little clunky, but once you’ve learned them in and out I’m not sure they’re quite as bad as you suggest, however that doesn’t preclude improvement, or making the more immediately useful aspects of the NLA editor (such as layering) more accessible and faster/easier to use.
Improve the flow of the NLA editor, rigging, animation systems - sure, I’d be all for that, but it does currently all work, its not god-awful hard to use, and it takes dev time.
Adding redundant systems out of ignorance as has been suggested, no, I’m not in favour of wasting time on that.
I’ll add here that the ‘push down’ button plus the fact that pressing tab adds the strip to the graph editor and dopesheet, make it pretty damn seamless.
this whole thread is a bit ignorant. just read the title!
blender CAN do so much,what it does well it does really well but in some areas it is just a chore. its frustrating because in many areas it is so close. i’ll diverge a little. a lot of the quirks can be ironed out with addons and some cant.
in animation: constraints with useable offset, easy setup of driven keys, layers, joint based rather than bone based rigging…all are missing, broken or extremely cumbersome. in blender’s favour we have bendy bones, armatures and vertex groups and a quite polished weight paint.
in motion graphics (i do much less) but i’m pretty sure animation nodes are way clunkier than cinema4ds mograph but ill admit im no expert in this area.
in sculpting we are remarkably competitive I hope multires with Sergey’s changes will be much more stable… people say blender isn’t because it can’t handle the polycounts zbrush can but when you can sculpt “like butter” on 128million poly mesh i wonder what they are on about… but putting the fanboyism aside try doing a boolean between a couple of meshes in the 1-2million poly range and you will wait several minutes wondering if you have crashed… and even then the results usually still have non merged areas, holes and bad geo. zbrush does it almost instantly and much cleaner… and people will still want sculpting layers, a quad based remesher, more brushes and the curve toolset…
UV mapping: blender used to be ahead in this area but packing and relaxing are woeful
texture baking works but is convoluted and misses many passes in cycles… BI used to be simpler and quicker but in 2.8 is gone.
retopology:blender is actually quite strong in tools but woeful in viewport visualisation… retopoflow has a lot right but also a lotwrong (selection is often painfull and you miss a lot of the standard edit tools butt least you can see what you are doing)
i could go on… you can do substance painter style smart materials and substance painter is clever in making you bake a suite of maps first… but blender would need a baking upgrade and a massive rethink in library systems and asset management for brushes to come even close to user friendliness here…
Not really, no. When choosing a piece of software to use for a given task, and asked which is the most powerful, the fact it can do things you’re not interested in is generally irrelevant. If I’m interested in sculpting, it doesn’t matter that zBrush can’t edit videos or create 2d animations, it’s brute force power in the area I’m interested in eclipses Blender.
Something being powerful is not the same as something being versatile. Blender sacrifices power across the board in order to be more versatile to the end-user.
1 Like
Not at the moment anyway.
Having said that, a 3DMax user who works at a studio that I know saw one of the 2.8 overviews was very impressed.
Prepare yourself for Blender to become a lot more accepted within the industry.
Agree with you on the animation front, but you really nail the reason why even smaller studios need ZBrush with this post
I’ve been working on a bunch of miniatures for 3D printing and, for the most part, the sculpting tools in Blender are up to the task. However, when it falls down - it falls down
hard. Booleans of any reasonably high polycounts are just not worth the time (on one model, I knew to go make a coffee and chat to the wife before checking back on it’s progress). Ambient occlusion sounds great, but zoom in too far on a subdivided mesh and performance dies. 1-2 million polygons vs tens of millions of polygons. And so on.
You don’t always need these features mind you. The issue is that, when you do, ZBrush can handle it and Blender cannot. For the hobbyist who cannot afford other tools, or idealogue who refuses to use the commercial variant, Blender is more than adequate as they’re likely to work within the bounds of what Blender offers. For those working professionally and so need to include their hourly rate in their cost/benefit considerations - being able to do something “out of the box” is more powerful than sitting there for a few days working around the lack of a feature/performance.
You can’t ignore that blender can effectively replace an entire pipeline of applications, not requiring clunky import/export operations.
That’s a power of it’s own. You’re ignoring blender’s strengths by splitting it up.
just one example: The video editor can link directly to the 3d scene in the timeline.
Everything you said is pretty much on point, but I still think you’re wrong about how difficult it is to use the NLA editor for animation layers.
How many paid developers have blender right now?
Only if every other 3d package stops their development…
6 at the moment, but not sure they are all fulltime, and also they don’t all have a year contract.
You can see their weekly progress in the meeting notes :
Haven’t seen you post in awhile. Good to see you’re still around.
Blender cannot be the most powerful 3D package for a lot of reasons.
Blender is different from other software because he can do more on some areas.
But since there are not enough devs on those areas, they aren’t be improved and by the time they are not competitive with other software.
Who will use the VSE/compositor when there is some professional software for free that do a way better job.
VSE is only on one core, the compositor isn’t powerful enough too.
Blender viewport is really far from Maya’s viewport.
Try to add a lot of assets to the viewport, it’s not responsive.
Edit mode is bad too.
On a lot of areas, blender isn’t strong enough compared to other software.
2.8 should change some but not all and that will take time.
At the same time, other software will be even better.
If the BF had more founding and devs, that could change.
The code quest showed that people are ok with this kind of founding and for me, this code quest is a win, blender is better than previously.
Maybe if they do one code quest per year and per area Blender could be better faster.
Another important thing is that blender doesn’t use powerful Open source features, it should be the first implementing them, UDIM, ALEMBIC, OPENSUBDIV, OPEN VDB, USD etc, but it’s, in fact, the last software to implement them.
Alembic came only because tangent wanted it and proposed a patch, UDIM is coming only on 2.8, previously devs didn’t think it was useful or didn’t know about that.
That’s also something that is bad for blender, devs don’t want to look at other software, they want to reinvent the wheel again an again.
Like I already said numerous time, they must see what’s existing, take blender strength in the account and make something better without copying.
I think things change, Blender is more open to the outside world and I hope it will continue like that.
Blender can be blender even if it’s more open and in fact, for me it’s what will change people minds, having a great software open to the outside world!
2 Likes
Err, that very description is a note to the section listing “Animation Layer and/or Motion Cleanup” as one of the 10 big issues Blender faces, describing precisely what everyone else is saying: The NLA Editor kind of does what animation layers do, but in no way is it sufficient.
Though if you, in fact, know something we don’t - a very simple work around that would accommodate the animation layers workflow, I actually would like to know. Maybe start a thread or something because I have yet to come across a good workflow and I’m legitimately curious. It would also help out the many Blender artists wanting animation layers.
It’s very simple. Large teams of well paid people employed by wealthy companies specialized in solving very specific problems (companies making renderers, particle/fluid simulators, hair solutions or instancing/scattering systems for example) will always have way, way more resources than one or two guys hired by an open source foundation to accomplish the same tasks.
The leaders in the industry, the large studios rarely look at the price of the software. The only relevant factor is the quality to satisfy Hollywood AAA grade requirements. They need top of the line software and plugins to achieve that. As long as Blender is GPL, commercial plugins will always provide significant edge over anything Blender can ever offer in terms of output quality as overall DCC package.
If you take a look at the AAA CG production, most of the eyecandy you see can’t really be attributed to vanilla 3ds Max or Maya, but rather to specialized plugins like V-Ray, Thinking Particles, PhoenixFD/FumeFX, Ornatrix, RealFlow, ForestPack, etc… Those are usually the things that make the bleeding edge quality results possible.
If you were to compare vanilla 3ds Max without a single plugin to vanilla Blender, they would actually be surprisingly close in terms of capabilities, especially if you were to compare earlier versions of Max that came with Mental Ray instead of Arnold to Blender with Cycles.
Indeed, but the base of the software helps a lot.
For example, when learning Maya, you don’t have to use plugins to deal with it, you have everything you need.
Only after you will use plugins and only external tools to do what you need to do.
V-Ray, Thinking Particles, PhoenixFD/FumeFX, Ornatrix, RealFlow, ForestPack, etc…
Blender isn’t the same, 2.79 is a pain and honestly, without some addons, it’s not great.
This RMB is a pain, for people coming from other software.
2.8 will change most of people dislike from 2.79.
1 Like
Well, Blender defaults, especially pre-2.8 are terrible indeed 
But I excluded that from my post, because I wanted to talk just about raw capabilities. I mean these days, Maya out of the box is quite capable indeed, especially with Arnold. But few years ago, when Maya still had Mental Ray, Blender was pretty close in terms of out of the box capabilities, especially given the amount of addons bundled with vanilla Blender
My point is that defaults aside, what actually makes up most of the obstacle in a way of Blender to be considered powerful by industry standards (as asked in the thread title), are the 3rd party plugin capabilities. Those usually make most of the difference.
I mean, I can’t imagine using Max 2016 without any plugins. I could not render anything much usable because I’d be stuck with mental ray. I could not do any instancing at all because there’s no ForestPack and particle flow system doesn’t support instancing (the particle system in blender does). I could not do any fluid or smoke simulations, 3ds Max just doesn’t have any (Blender does), or I could not do any hair, because 3ds Max hair system is unusable and mental ray sucks at rendering it (Blender’s hair system actually works and Cycles renders it well).
Then the plugins come and change the tide: V-Ray is a lot more robust renderer than Cycles, making me 100% sure I can get any job done. ForestPack is powerful instancing solutions, allowing me to create large, foliage populated environments - Blender’s particle instancing fails at that. Thinking Particles are way superior to very basic particle system in Blender. PhoenixFD is a lot more detailed and advanced fluid simulator than Blender’s simple fluid and smoke simulator.
People here tend the really underestimate how incredibly much difference do commercial plugins make.
1 Like