Blender New Innovative 2d Animation "GP Tools"

[herbert123]

Well, Herbert . . . looks like it "may" become quite useful and at the top of the heap.

But, unless you or someone else does that essential Python scripting that will get ALL the other stuff out of the way - ONLY exposing those panels and tools which are needed for Grease Pencil Animation - I have my doubts as to the adoption rate among 2D animation folks (mostly, I think they hate complexity that seems to exist for no apparent purpose - that is why they are not doing 3D animation).

And, right now, it is Blender's apparent complexity and unexplained functionality that keeps the majority of people gnashing their teeth and running away, frightened to the quick. Me included.

Since anyone who customizes Blender for a specific purpose can also charge for their modified application - there really could be monetary profit for someone producing a 2D animation only software package - complete with step-by-step tutorials. Who's gonna step up to the plate?

Thanks for the step-by-step instructions.

P.S. About Open Source project progress: using OpenToonz as an example - initially there was only the Dwango crew working on it - then, it seemed there was a flurry of outside programmer interest - including what looked like a serious commitment from one of the better known Blender contributors, Ideasman. It seemed he would be a leader among men for getting the daunting and necessary tasks finished. After a few commits - he bailed. He said that a lot of errors were popping up - and that he didn't have time to continue with his OpenToonz commitment. Since that time, the OpenToonz forum has been virtually inactive, for the most part. It could just be a coincidence.

@blurymind, can't redo either of those bugs, they both look to be OpenGL related.
Probably just using older graphics cards.

I've been fixing some other bugs, running into quite few errors in the code
(low level stuff - using uninitialized memory, use-after-freeing).
None of this is Linux spesific though - so not sure how they didn't cause problems on OSX/Win32.

Longer term I can't really spend all that much time looking into issues that require a lot of investigation.
This is a big project and really needs its own dedicated team to develop and support it.

Greg Smith

To be fair, it is already possible to create a dedicated workspace for grease pencil and the 2d animation stuff. I created a screen layout for GP animation today, which only shows the tools required to start animating 2d. Although the user cannot completely hide those superfluous side panels, at least we can close them, and only the header remains. And a screen layout in Blender remembers the state of those panels, so it works quite well.

As for Blender's complexity: well, that comes with the territory of 3d applications in general. They're complicated beasts, and with power comes more complexity. Compared to ToonBoom where the user must have Maya installed as well to work with 3d, I think Blender at least provides everything in one package. And I would be careful when coining "the majority of people [..] running away" - that statement is just not true. Blender is no more difficult to learn than Maya or Max or Houdini. Learning to use a 3d application just takes time and a little bit of effort. With some, a LOT of effort, while others pick it up quickly (I have seen this in my classes).

About Ideasman: he worked as one of the main contributors on Blender for almost two decades, and he decided he needed time for his own personal projects for a while. Ideasman worked on getting OpenTOonz to the Linux platform, and he was the only one spending time on this, and he did not just "bail" out. The guy earned my deepest respect seeing all the work he has done.

And others are happily contributing away, such as Turtletooth.

Here is my GP layout exposing just the tools for 2d animation:
highres: http://i67.tinypic.com/zxwrrp.jpg

[JaMike]

Blender is no more difficult to learn than Maya or Max or Houdini. Learning to use a 3d application just takes time and a little bit of effort.

I'm with Greg here - 2D artists don't WANT to learn something as complex as Maya or Max or Houdini.

Ideasman worked on getting OpenTOonz to the Linux platform

Looks like his effort would have been better suited to fixing some of the bugs first, THEN worrying about converting it to Linux.

What OpenToonz needs is some proper leadership. Greg, wanna step up? I'm only half joking. And Herbert can be head of marketing. I could be in charge of making tea.

[Psmith]

Mike:

Not in a million years - (which may be how long we will have to wait for OpenToonz to "get finished").

I wonder how many people there are in software purgatory?

Besides you and me, I mean.


Greg Smith

[herbert123]

Blender is no more difficult to learn than Maya or Max or Houdini. Learning to use a 3d application just takes time and a little bit of effort.

I'm with Greg here - 2D artists don't WANT to learn something as complex as Maya or Max or Houdini.

Absolutely agree! That is why you see so many 3d animators who can animate well in Maya, but have no idea how to model or render in it. They stick to the animation parts only, and will not touch the rest.

And I do not see any of my 2d animator friends / colleagues switching to Blender's 2d animation tools anytime soon. ToonBoom is too entrenched in the industry for that to happen, and Blender (any full-on 3d app) is just too much for them to handle in general. I am a bit of a rogue in that regard

That is why I hope for more base GUI configurability of Blender in order to hide all the superfluous stuff, and allow 2d animators to only focus on the core functionality to use the new 2d animation tools.

Ideasman worked on getting OpenTOonz to the Linux platform

Looks like his effort would have been better suited to fixing some of the bugs first, THEN worrying about converting it to Linux.

His effort was aimed at getting a professional 2d animation platform on Linux. Who are we to say his efforts are misplaced or not? It is easy to criticize. These are all volunteers, you know.

What OpenToonz needs is some proper leadership. Greg, wanna step up? I'm only half joking. And Herbert can be head of marketing. I could be in charge of making tea.

I agree - something similar to the Blender Foundation would be nice. On the other hand, development of OT is continuing at a good pace - just compare open source projects such as Gimp (which is glacial in comparison).

[herbert123]

Mike:

Not in a million years - (which may be how long we will have to wait for OpenToonz to "get finished").

I wonder how many people there are in software purgatory?


Greg Smith

Whether software is "finished" depends on your point of view, the work you (want?) to do with it, and so on.

Fact is that the first (buggy and feature regression-prone) version of OT was used in feature-length Ghibli productions - to them it was "finished" enough to do the job at hand. And I have seen work from hobbyist animators who are now working with OpenTOonz, and their work is quite nice - seems to work for them as well.

No software is perfect, eh? I wouldn't want to do extensive frame-by-frame animation in Moho. That is just not its focus. I feel Moho and OT complement each-other quite well.

[Greenlaw]

Software is finished only when it's abandoned.

[JaMike]

Not in a million years - (which may be how long we will have to wait for OpenToonz to "get finished").

I wonder how many people there are in software purgatory?

Lol!

Everybody with a computer, I reckon.

[JaMike]

Ideasman worked on getting OpenTOonz to the Linux platform

Looks like his effort would have been better suited to fixing some of the bugs first, THEN worrying about converting it to Linux.

His effort was aimed at getting a professional 2d animation platform on Linux. Who are we to say his efforts are misplaced or not? It is easy to criticize. These are all volunteers, you know.

I was being a bit snarky, there, I admit. I'm sick of half-assed open source software that gets abandoned, and I don't want OpenToonz to go the same way. Spending effort porting it to different platforms as an intellectual exercise (or as a promotion for the platform) doesn't mean a thing if it is unfinished and nobody wants to use it.

There would be a benefit if other Linux developers jumped on board because it was now on the platform, but it doesn't look like that's happened.

[JaMike]

Fact is that the first (buggy and feature regression-prone) version of OT was used in feature-length Ghibli productions - to them it was "finished" enough to do the job at hand.

Are you sure about that? I would have thought the splitting from the Premium version would have introduced all those bugs. At the very least they took out the software protection.

No way Miyazaki would have put up with software that crashed so much.

[Psmith]

Toonz worked for Studio Ghibli because they have hundreds of animation slaves working, largely, with a frame by frame workflow. The animation environment in Japan is quite unique and does depend on lots of underpaid laborers. I daresay that anything that absorbed at least some of the sweat, producing more output for less money, would have been sufficient.

The younger, Western animation enthusiasts have quite idolized everything Japanese - but details of actual practices and conditions within the animation industry in Japan need to be known. Quite literally, "you don't want to go there":

https://www.buzzfeed.com/danmeth/this-a ... .ppdxvxOlN

I wouldn't want to encourage any potential animator to aspire to working in any of the notable studios - experiencing life as an animator as this fellow has experienced. We need better, faster, more efficient ways to produce the highest quality animation (forget about emulating Anime) on a small scale. This was Mike Clifton's initial premise for Moho - and it should be the example used for the development of OpenToonz, as well.

Forget the studios!


Greg Smith

[herbert123]

Fact is that the first (buggy and feature regression-prone) version of OT was used in feature-length Ghibli productions - to them it was "finished" enough to do the job at hand.

Are you sure about that? I would have thought the splitting from the Premium version would have introduced all those bugs. At the very least they took out the software protection.

No way Miyazaki would have put up with software that crashed so much.

They were working with a very specific subset of functionality of TOonz. That is why the Japanese developer deactivated and borked pretty much all the paperless animation tools. I tested that part in the first version, and it worked fine. The other stuff, though, was in a severe broken state. Luckily, most of the regressions have been undone by now. It hardly crashes for me. I can work hours and hours in OpenTOonz without crashing it.

[herbert123]

Toonz worked for Studio Ghibli because they have hundreds of animation slaves working, largely, with a frame by frame workflow. The animation environment in Japan is quite unique and does depend on lots of underpaid laborers. I daresay that anything that absorbed at least some of the sweat, producing more output for less money, would have been sufficient.

The younger, Western animation enthusiasts have quite idolized everything Japanese - but details of actual practices and conditions within the animation industry in Japan need to be known. Quite literally, "you don't want to go there":

https://www.buzzfeed.com/danmeth/this-a ... .ppdxvxOlN

I wouldn't want to encourage any potential animator to aspire to working in any of the notable studios - experiencing life as an animator as this fellow has experienced. We need better, faster, more efficient ways to produce the highest quality animation (forget about emulating Anime) on a small scale. This was Mike Clifton's initial premise for Moho - and it should be the example used for the development of OpenToonz, as well.

Forget the studios!


Greg Smith

Any feature-length high-quality frame by frame animation takes a tremendous amount of man power. And this does not only take its toll on animators in Japan: an animator acquaintance of mine told me he had to produce at least 80 drawings a workday for Warner Brothers animation studio where he worked in his animator days. One drawing short: outta there. The work pressure, the expected unpaid overtime, and low wages made him decide to become a carpenter/painter, because he would get 25 dollars an hour for that work, and relaxed working hours.

A year or so ago I worked together with a friend of mine on a brief for a new Disney television show (animation studios compete for gigs), and inbetweens are often done in Asian countries. The quality we got back was pretty horrendous.

Good frame by frame animation, such as the ones done in the fourties/beginning fifties of the last century is just not really feasible anymore. It could be, but studio managers and producers have no interest in it at all. 2d broadcast animation nowadays is absolutely terrible in comparison - the model sheets are king, and it is all about tweening motion, mostly without any soul. I understand why that is - quick productions, no money, and graphic design rather than animation rules now.

The trouble is that 2d puppet animation is quick, cheap to produce, and can be made to look very stylish. But it has generally nothing to do with quality animation. Tweening is not animation. The classic Flintstone animators knew how to animate with bare bones assets, but it seems that craft and knowledge is lost in time. Instead our eyes are bomblasted by dreadful broadcast animation such as Family guy, Archer, and the most terrible 3d static looking kids animation ever.

I will admit these can be funny to watch, but quality animation it certainly is not.

I agree with you, Greg: good frame-by-frame is no longer an option. Not from a financial point of view, and even if the money would be available (and it actually is in studios, but you'd be amazed on what it is wasted), it would be incredibly tough finding animators who would be able to even come close to what was done in the fourties/fifites. The knowledge and experience is just not there anymore.

[Psmith]

Herbert:

The most salient point is that it is the Studios who are driving this endless downward spiral of non-animation for entertainment - so much so that it has been a conditioning exercise to adapt every successive generation to the new and lower standard.

Since we both agree that frame by frame animation is not practical for individuals who are attempting to complete a piece in their lifetime - and the chances are only slightly higher for a small group of animators working in a frame by frame fashion - what compromise might still produce a quality result - using tweening as the BASE structure for producing such an animation of quality?

Straight Cut-Out animation is basically "Silhouette" animation - a form that was the very first form presented on film:

https://en.wikipedia.org/wiki/Silhouette_animation

In straight Cut-Out animation, nothing really changes in perspective or proportion - and movements appear quite robotic - even in the most talented examples. When a hand or head or body needs to change its perceived viewing angle - an instant switch of parts takes place - quite unnaturally.

We saw this demonstrated in the earliest animated "Silhouette" films and it is rampant today, everywhere on T.V.

The entertainment is not coming from the animated content, but rather from the snarky dialog - (the acceptance of sarcasm as the only popular form of humor is the result of decades of crowd conditioning).

If I can use the example of the Pixar animation technique - keyframes (tween-frames) are put down in an approximate fashion to establish timing - and, gradually, more and more keyframes are added - which amount to in-betweens - until, as one interviewed animator put it, 'we end up with keyframes on almost every frame'.

Even if keyframes end up appearing on almost every frame of a high quality animation, how can this save time, you might ask? Well, Posing a character is far faster than drawing a character in every needed position, is my obvious answer.

That's the gist of it.

Model a character in several key positions, initially - then, through "Node Based", flexible 2D Mesh-Based Animation - create all primary poses between those Models - each Key Model having a specific Mesh and set of Nodes, allowing it to be aligned perfectly to each successive Model (representing every key pose) - thus eliminating all unnatural jerkiness. This is precisely the OpenToonz method of tween-based animation. The simplest and most practical example of the technique I've seen.

This method differs from standard Cut-Out animation in the way of total flexibility of Shape, Size, Perceived Viewing Angle and Perspective.

Once a character moves just as you would have it - with no robotic appearance - line detail can be added on another "Layer" - finalizing the believable 3D appearance and motion of the character.

Why not just use 3D, and be done with it? Because every 3D application that I know of is far more complicated and time consuming to use - just to set up a character. It takes even more time and sweat and frustration to animate it. And "Toon" rendering still does not produce a hand made look.

Using this method I've described actually gives an individual animator the ability to finish a project.


Greg Smith

[JaMike]

good frame-by-frame is no longer an option.

It is for Cartoon Saloon and Lupus Films. Song of the Sea? Ethel & Ernest? Both hand-drawn in TV Paint. Both beautiful, and cheaper productions than anything from North America.

[JaMike]

Why not just use 3D, and be done with it?

Also, 3D can't do the impossible things that 2D can do. Drawings can be "wrong", and not work in the 3rd dimension. That's why, IMHO, 2D is more emotional on an artistic level than 3D. Yes, Pixar can kill a character and make everyone in the cinema cry, but they struggle to get the same killer pose that they could in the storyboard that tells you what a character is feeling at that exact tragic moment.