Pencil2D development continued: 0.5.5 is on it’s way!

Pencil2D development continued: 0.5.5 is on it’s way!

Hi all Pencil2D fans. This is Matt, I’m back.

First of all, I have to apologize to you for not being so active in Pencil2D development in the past year. 2014 is a tough year for me, my work occupied nearly all my time and energies. And what is worse is to see a “Development ceased?” post on the forum ;(

After the rush has just finished, I can finally come back to my lovely Pencil and start to contribute.

I think the most important task for now is to release a stable version of Pencil2D. There are already a lot of new features in development branch, but it’s buggy, I know.

I will start browsing all bugs on Github issue tracker and try to fix them as soon as possible. You can report bugs you’ve found using issue tracker, it’s easier for us to follow and manage than post on forum.

Cheers.

Newsletter #1

Pencil2D has taken off !

Hi everyone! A few months ago we launched the website www.pencil2d.org to host Pencil2D, a fork of the original Pencil software, written by Pascal Naidon and Patrick Corrieri, which was no longer being maintained.

While it has taken some time to start this new project, it has now taken off. Matt Chang is our new developer and while he is hard at work.

Matt began by creating Git repositories on Github and Bitbucket, to  host the source code for the project. He quickly fixed some bugs and released the first version of Pencil2D.

We started a dialogue about Matt’s work and his mission to fork Pencil, and people on the Internet soon started to to take notice. (For a history of Pencil, see this article; another one hereis about 2D animation softwares on Linux)

After that, Matt got in touch with Konstantin from the Morevna fork, and they decided to merge their repositories so as to have a single, active repo that would contain all history of commits, including Oluseyi and Morevna, as well as Matt’s branch. In the past 30 days, 136 commits have been made — among others, Matt has implemented custom shortcuts in Pencil2D.

We would like Pencil2D to become a tool that allows people to make animation sketches (or more complex animations) with a very simple, responsive software, where you can focus on animating.

Pencil2D is just starting out, but there is a feeling around here that things are moving forward. There is still a lot to do, but we are sure there is a place in the animation world for Pencil2D — while there is a lot of software available for vector animation and sketching, both open and closed source, there isn’t much available for raster animation with the simplicity that Pencil2D could offer.

We would love you to join our Pencil2D community. We also need you to test Pencil2D, spread the word about it and report bugs, as well as eventually contribute to its code: Matt is actually the only developer working on it, and he surely needs some help !

You are receiving this e-mail because you are registered on the Pencil2D website. As this is our first newsletter, we have sent it to all our members. In the future, however, only those who subscribe to the newsletter will receive it, and there will only be a few newsletters per year. Please make sure that you have signed up (the subscription form is in the right sidebar of the website) !

The Pencil2D team

PS : thanks to Kathryn for having reviewed this text !

Repository moved!

Hi all,

Pencil’s source code repository has moved to new place.

Thanks to Robin, he contacted Konstantin (morevna) and me after he saw Pencil, Pencil, and Pencil, a report about Pencil’s current situation on LWN.net.

After some discussion, we decide to create ONE merged repository that contains all commit history include Oluseyi, morevna, and chchwy branch. I believe users and potential contributors would benefit with an active Pencil repo.

Old repositories are no longer maintained.

 

Pencil, Pencil, and Pencil

This article was originally written by Nathan Willis on LWN.net. Thanks to him !

An unfortunate drawback to the scratch-your-own-itch development model on which many free software projects depend is that creators can lose interest. Without a maintainer, code gets stale and users are either stranded or simply jump ship to a competing project. If the community is lucky, new developers pick up where the old ones left off, and a project may be revived or even driven to entirely new levels of success. On the other hand, it is also possible for multiple people to start their own forks of the code base, which can muddy the waters in a hurry—as appears to be happening at the moment with the 2D animation tool Pencil. Plenty of people want to see it survive, which has resulted in a slew of individual forks.

Pencil, for those unfamiliar with it, is a “cell animation” application, which means that it implements old-fashioned animation drawn frame by frame (although obviously the software helps out considerably compared to literally drawing each frame from scratch). In contrast, the other popular open source animation tools Tupi and Synfig are vector-based, where motion comes from interpolating and transforming vector objects over a timeline. Despite its old-fashioned ambiance, though, Pencil has proven itself to be a popular tool, particularly for fast prototyping and storyboarding, even when the animator may create the final work in a vector application.

Original Pencil maintainer Pascal Naidon drifted away from the project by 2009. At that time, the latest release was version 0.4.4, but there were newer, unpackaged updates in the Subversion source repository. Version 0.4.4 eventually started showing signs of bit-rot, however, particularly as newer versions of the Qt framework (against which Pencil is built) came out. Users of the application, however, have continued to maintain a community on the official site’s discussion forum.

A box of forks

Understandably, there were never a ton of Pencil users, at least as compared to a general-purpose desktop application. But the dormant project picked up a dedicated follower when the Morevna Project, an open source anime movie project, adopted it for its workflow. Morevna’s Konstatin Dmitriev began packaging his own fork of Pencil in late 2009, based on the latest official Subversion code. He added keybindings for commands, command-line options to integrate Pencil with Morevna’s scripted rendering system, and fixed a number of bugs. Subsequently, he began adding new features as well, adding user-selectable frame rates, some new editing tools, and support for multiple-layer “onion skinning.” Onion skinning in animation is the UI technique of overlaying several (usually translucent) frames onto the current drawing, so the animator can visualize motion. But there are also a lot of bug fixes in the Morevna fork that deal with audio/video import and export, since the team uses Pencil to generate fill-in sequences for unfinished shots. Since Morevna is a Linux-based effort, only Linux packages are available, and they are still built against Qt 4.6.

In contrast, the Pencil2D fork started by Chris Share eschewed new features and focused squarely on fixing up the abandoned Subversion code for all three major desktop OSes. Share’s fork is hosted at SourceForge. One of the fixes he applied was updating the code for Qt 5, but that decision caused major problems when Qt 5 dropped support for pressure-sensitive Wacom pen tablets, which are a critical tool for animators. In early June 2013, Matt Chang started his own fork also at the Pencil2D site, using Qt 4.8.4. Whether Share’s fork hit a brick wall with the Qt 5 port or has simply stagnated for other reasons, Chang’s is still active, to the point where he has posted a roadmap on the Pencil2D forum and is taking feature suggestions. Chang has only released binaries for Windows, but he believes the code will run on Linux and OS X as well, and maintains it for all three.

Both of the forks at Pencil2D headed off on their own, rather than working with Dmitriev’s Morevna fork. More to the point, Chang’s roadmap includes a different set of drawing tools and a separate implementation of function keybindings. Luckily, the two forks’ editing tool additions do not conflict; Morevna’s adds a “duplicate this frame” button and adds controls for moving layers, while Chang’s include object transformations and canvas rotation.

In contrast to the other Pencil forks, the Institute for New Media Art Technology (Numediart) at the University of Mons took its fork in an entirely different direction as part of its “Eye-nimation” project. Eye-nimation is used to produce stop-motion animation. Numediart’s Thierry Ravet integrated support for importing images directly from a USB video camera into Pencil, where the images can be traced or otherwise edited. It uses the OpenCV library to grab live input from the camera, and adds image filters to reduce the input to black and white (bi-level, not grayscale) and smooth out pixelation artifacts. Ravet spoke about the project at Libre Graphics Meeting in April. The work is cross-platform, although it is built on top of an earlier release of the original Pencil code, 0.4.3.

As if three concurrent forks were not enough, many Linux distributions still package the final official release from the original project, 0.4.4. And there are several independent Pencil users who maintain their own builds of the unreleased Subversion code, some of which refer to it as version 0.5.

Sharpening up

On the off chance that one might lose count, the total currently stands at five versions of Pencil: the final release from the original maintainer (0.4.4), the unreleased Subversion update, the Morevna fork, Chang’s Pencil2D fork, and Numediart’s Eye-nimation. The situation is a source of frustration for fans of the program, but how to resolve it is still quite up in the air. Dmitriev maintains the Morevna fork for utilitarian reasons (to get things done for Morevna); his preference is to work on Synfig and he does not have time to devote to maintaining Pencil in the long run, too. Chang does seem interested in working on Pencil and in maintaining his fork as open project that is accessible to outside contributors.

But combining the efforts could be a substantial headache. The Morevna fork is considerably further along, but Chang has already refactored his fork enough that merging the two (in either direction) would be non-trivial, to say the least. And it is not clear whether the Eye-nimation feature set is something that other Pencil users want; Dmitriev expressed some interest in it in his post-LGM blog report, though he was concerned that Numediart had not based its work on the Morevna fork.

The primary competition for Pencil is the prospect that cell-animation support will get added to another program. Krita has a Google Summer of Code (GSoC) student working on the feature (in addition to the partial support already added), while Dmitriev said in a private email that he hopes one day to implement cell-animation features in Synfig. If either effort bears fruit, that would be a positive development, but in the near term news of something like a GSoC project can sap energy from existing efforts, yet still might ultimately fall short.

It is a fairly common problem in the free software community for a well-liked project to fizzle out because the maintainers can no longer spend the time required to develop it and no one else steps up. It is rarer for multiple parties to independently take up the mantle and produce competing derivatives—especially in an obscure niche like traditional animation software. But when that does happen, the surplus energy, if it remains divided, can still end up doing little to revitalize a project that many users want to see make a return.

Pencil : new Chchwy branch !

There is a great news on the forums :

Dear all,

I’m a pencil lover, and I start a new pencil fork last month : https://bitbucket.org/chchwy/pencil2d

Currently I’m trying to simplified the program structure, the old code base is not well-organized, I believe it stop the growth of pencil, code is too complicated to add new features.

I will be appreciated if somebody could give me some feedback (features, bugs, etc..)

Cheers

Matthew

Pencil2D (Pencil Development Fork)

Hi Everyone,

I’ve decided to create a fork of Pencil. The fork is called Pencil2D.

The goal of the fork is to

  1. produce working versions of Pencil2D for OSX, Windows and Linux,
  2. fix existing bugs and problems in the codebase, and
  3. if time permits, look at adding new features.

Currently, Pencil has not had any code commits for over a year so I’m hoping that the fork will reinvigorate the development process. A major problem that Pencil is currently facing is that software that Pencil uses (such as Qt, FFMPEG, etc.) has moved on and changed. This means that Pencil no longer builds.

In addition to this, the current Pencil repository needs a spring clean. It contains several unfinished branches, the status of which is unclear.

Anyone is welcome to join the Pencil2D project at whatever level they wish.

Those interested in Pencil2D can use the discussion forums for any information they require.

If, at a later stage, the original Pencil project gets moving again I’d be happy to consider working together.

Cheers,

Chris Share

The vision for Pencil ( by Pascal Naidon)

PDF by Pascal Naidon, created: 2010/05/07 modified: 2010/06/16

Abstract: The document is an attempt at defining the concept behind the Pencil project. It is designed to serve as a reference and guideline for contributors of the Pencil project. I will revise it as it becomes necessary during the evolution of the project and following discussions with contributors and users of Pencil. It is currently based on my original vision and discussions with Oluseyi Sonaiya and Ian Moore.

1. Introduction: some history

Pencil was originally an application created by Patrick Corrieri by the end of 2005 under the name “Pencil Planner”. It was a simple yet effective pencil test (a.k.a. line test) programme. It included a drawing area, a single-track timeline where keyframes could be added, previous and next onion skins. The drawing tools were a pencil and an eraser, along with a colour chooser. The animation could be played at a specified frame rate, and in addition it was possible to add a background image and a background sound. It was developed under Qt 3 and made available on Macintosh and then Windows platforms under GPL license. The Mac version could output QuickTime files. There were two releases v0.1 and v0.2. This is a screenshot of v0.2:

screenshot of v0.2

When I became aware of this programme, I realised that it could be evolved into an application for making traditional animations rather than just pencil tests. Obviously what was needed was more drawing tools to fill shapes with colours, and some selection tools to move things around. I got in touch with Patrick, and started implementing those functions, along with cleaning the original code.
I also implemented tablet support. This became v0.3 of Pencil, released in October 2006. Pleased with the result, I went on with more ambitious goals. On the one hand, the basics of traditional animation is to sketch pencil tests, and then draw clean images on top of those sketches.
Another fundamental of traditional animation is to be able to organise separate elements (background, character, or parts of it) into different superimposed layers. This definitely called for the possibility of superimposing layers. On the other hand, I also convinced myself that vector graphics rather than raster (bitmap) graphics were useful for making animated drawings. Vector graphics can be rendered at any resolution, can easily be deformed and resized, colours can be readjusted, etc. They are quite well suited to render the traditional animation style: contours filled with some colours – inherited from the “cell(uloid)” style. Therefore I decided to introduce not only raster layers, but vector layers as well. Yet I wanted vector drawing to have the same artistic feel as raster drawing, so I implemented the same tools for both layers – although some of them such as the pencil and brush behaved differently.This lead to v0.4 released in January 2007. Below is a screenshot:

screenshot of v0.4

It was a huge jump from the previous versions. A lot of the original software was rewritten at this stage. Although incomplete, buggy, and still very unsatisfactory at many levels as an animation programme, it had the capacity to produce decent animations. At that time, Patrick told me that he was pleased to follow these developments but did not have the time or capacity to get further involved in them, so he kindly agreed to grant me administration of the project. This was perfect timing as I had already invested a lot of efforts (in particular on the vector side) and started developing a personal vision of what the application should be in its final version. I considered creating a new project with a new name, but failing to find anything better than “Pencil”, I kept the existing sourceforge project. I supplemented it with a newly designed website, a forum, and later and wiki to get more feedback and public attention.
From then, I brought many refinements (some of them being quite unknown such as curved gradients) and got some occasional help from people such as Benjamin Raverdy, and Mark Schmelzenbach, although I remained mostly the only developer. The latest version v0.4.4 to date (released in early 2008) looks like this:

screenshot of v0.4.4

As of May 2010, this version is downloaded about 1000 times a day, according to the Sourceforge statistics.
After this version, I had to make a pause for several reasons. Firstly I could not dedicate as much time to Pencil, and secondly I had long realised that the code had grown organically and inherited many poorly designed elements and architectures which needed some cleaning. So I set myself to this task. Some time has passed. Fortunately, then came along Oluseyi Sonaiya who by and large shares the same vision of Pencil as me, and also decided to take advantage of his experience as a developer to clean and extend Pencil to the next level.

2. The spirit of this project

When I started to publicise Pencil, my intention was to make everyone benefit from my ongoing effort to design an animation application. However, while being an open-source project, I believe that an application like Pencil should be designed with some personal vision and goal to remain effective. I hope other developers will share my vision and join this effort, and that Pencil will become useful for most people interested in animation. As general rules,

  • feedback and suggestions are very welcome and will get due consideration
  • any kind of contribution is very welcome if it indeed integrates into the vision and concept of the project, and will get consideration for official release
  • people interested in making substantial changes inconsistent with or departing from the concept of the project are free to do so, provided they create their own distinct project, or possibly subproject (optional extension, etc).

It does not mean

  • that Pencil will necessarily comply to anyone’s needs, habits, desires.
  • that the Pencil project is everyone’s, that anyone can do whatever they want with it, claim new version, or dismiss credits of people who contributed to it.

These points, in my opinion, follow common sense and the usual acceptance of the term open-source, consistent with the GPL license of Pencil. (If it is not the case, the license should be updated accordingly).
So far I have developed Pencil mainly by myself and for myself. Now that several people are interested in participating in the project in the long term, it has become necessary to provide some information about my original vision and concept of Pencil. This is the goal of this document. It was kindly suggested to me by Oluseyi.

3. The concept

3.1 Why this project?

As I once wrote on the Pencil website, I have always liked drawing and the art of creating motion from a series of drawings. As a child, I used to make animation by drawing pictures on each page of a notebook. It was fun and simple, but not so convenient and quite limited (compared to a real animation film featuring colours, sound, etc).

pascal-vision-d-003Diplaying pictures and playing sound is an ability that computers quickly acquired. As a result, a lot of animation software naturally emerged over time. However the usual input systems (keyboard and mouse) used by computers, along with conventional software design, have kept those programmes remote from the original drawing experience of traditional animation. Fortunately, pen tablets and later tablet PC appeared and made it possible to get an experience much closer to actual drawing. Still I found that existing animation programmes were either

  • not taking advantage of the tablet
  • not recreating the drawing experience of traditional animation, or not completely satisfactory in their design
  • oriented towards other goals than traditional animation
  • not available / affordable to the general class of users

So I decided to develop an application which would be simple, available to anyone, and designed to recreate the drawing experience of traditional animation, along with some other benefits coming from using a computer. In a sense, it is an attempt at conciling the traditional pencil approach advocated by people such as Hayao Miyazaki[ref]Hayao Miyazaki: drawn to perfection[/ref] and the advantages offered by today’s computers, in particular pen tablets. This is the guiding line of the project.

3.2 What Pencil is not:

  • Pencil is not and will never be a clone of some existing software
  • Pencil is not and will never be defined as the “free of charge” alternative of some existing commercial software
  • While Pencil might get inspiration from existing software, it is not required to follow closely the interface or design of some existing software just because it meets some users’ natural expectation. Original design/paradigm might prevail over previously established ones.
  • For practical reasons, and unless some essential function makes it necessary, Pencil in its internal structure will not be bound to comply to some predefined format or protocol, other than those defined by the Qt framework. In particular, Pencil is not a SVG editor or Flash editor, etc. It should be however be able to import and export various useful formats with some reasonable accuracy/compliance to those formats’ specifications.

3.3 What Pencil is:

The primary goal of Pencil is to create traditional animation, which consists in drawing several pictures to create the illusion of motion.
Because of its potential, it also appeared that it may be used in the future for related usages such as

  • drawing, painting, editing pictures (without animating)
  • drawing storyboards or comic strips
  • taking notes/sketches/doodles and organising them
  • editing videos.

However, these are not the primary goal of Pencil at present.
Finally, Pencil might be the basis for an integrated solution for animation project management, often referred to as “animation studio”.
Therefore, the anticipated development of Pencil is as follows:

anticipated development of Pencil

In order to allow the secondary usages, Pencil will be designed in a very modular way from v0.5
and beyond.

3.4 The general interface

3.4.1 Appearance

Visually, Pencil should be dominated by the drawing/viewing area since it is focused on drawing.
Tools, options, and other technical elements should appear around this area in a non-obstrusive way and kept to a minimum. Their graphical representation should also assume a simple (but pleasing) appearance to let the user work with some “peace of mind”[ref]In that respect, the current Windows version has some unnecessary embossing lines and approximate layout which leaves an uncomfortable impression. The Mac version displays unnecessary boxes around each tool.pascal-vision-d-006pascal-vision-d-005Drawing area[/ref].

The platforms of reference will be pen tablet laptop computers and single-screen desktop/laptop computers, although we might also consider multi-screen usage. To satisfy different users, the interface will be customisable to a large degree, although predefined, thoroughly thought interface environments will be provided by default for each usage and platform.
Ideally, Pencil should be a multidocument application; open documents could be indicated by a series of tabs at the top of the screen.

3.4.2 Multiple usages/environments

The current and only usage of Pencil is creating animations. When (or if) secondary usages become available, it is expected that Pencil will have some option to switch between those environments. It may appear as a menu in Pencil menu bar or as a series of buttons at the top of the screen (easier for tablet users), or even as a splash screen option.
The idea behind the different usage environments is to display (or possibly alter) only the elements which are relevant for a given usage.
For example, the current interface of Pencil is dedicated to drawing animations. It has the following elements : A layer/timeline palette displaying both layers and their associated tracks, a colour palette, a tool palette, a parameter/option palette, and a display palette. A menu bar (or a function bar with buttons when using a tablet) contains the basic file/edit functions (open, save, import, export, undo, redo) and some more specific functions.

pascal-vision-d-008

Switching to a purely drawing/picture editing usage will make the timeline irrelevant. However, it will still be necessarily to access layers. The interface would switch to something like this:

pascal-vision-d-009

Switching to a purely video editing usage will make the drawing tools irrelevant. The interface would switch to something like this:

pascal-vision-d-010

Switching to a note-taking environment might look like this:

pascal-vision-d-011

This versatility is quite interesting, although it raises many questions.
Obviously the same kind of information cannot always be displayed in different environments (an animation is not quite a note which is not quite a picture). This would call for creating variants of the Pencil format for each usage, which would effectively correspond to having a suite of different applications just like OpenOffice. Another, possibly more fruitful approach is to allow the Pencil document format to have a section for each kind of environment. For example, the same file could contain a section of animation and a section of notes of handwritten ideas, instructions, or sketches which could be copied and pasted in the animation section. Of course, if a file contains only one section, it would open by default in the corresponding environment. In the case of creating a new document or opening a documents with several sections, a splash screen would ask the user which environment to load first.
Can this be done in practice? The danger of this approach is that Pencil become an overweight application. While it might be acceptable to wait several seconds to load a full-featured animation programme, it might be less so in the case of a quick image-editing programme or note-taking application. One possible solution would be a light-weight core that could dynamically load only those modules which are required for a given usage. The effectiveness of this approach should be investigated.
Other practical questions arise about the interface. For example, the onion-layer buttons allow to display/hide previous and next images in semi-transparency, which is a traditional useful trick to draw animation. It may be seen first as a display feature. However, it does not make sense in usages which do not involve time and the notion of previous and next image. In that sense, it should be included in the timeline (which disappears when switching to non-time related usages). On the other hand, it would hardly be useful in, say, video-editing mode. Defining where and when the elements appear obviously requires some thinking.

4. The animating environment

The following describes specifically the animation environment, which is Pencil’s current main goal and prority.

4.1 Drawing

4.1.1 Raster and Vector Layers

Pencil has two drawing modes: raster and vector. Although it is advisable for the user to know the difference between the two, it is essentially a technical difference about how drawings made by the user are encoded and stored in the programme. In most applications, the choice of one of these two technical implementations heavily influences the way the user can input and interact with the drawing. As a result, raster graphics programmes and vector graphics programmes usually have quite different interfaces and tools. In Pencil, I want to break with this tradition and offer a common interface which is close and to the natural drawing experience. Other conventional tools more specific to raster or vector modes might be added as a special set for more advanced or conservative users.
There are many styles of 2D animations, from the conventional American celluloid style to the Russian sketch/paint style, and Pencil will give some freedom in that respect. Yet the celluloid style is the most common one and will be taken as a reference. We will assume the following steps:

  • sketch/pencil/line tests
  • draw clean lines and temporary lines
  • fill areas with colours

pascal-vision-d-012

The basic drawing tools should be able to perform these basic tasks. These tasks are in fact quite separate, and it is convenient to perform them in different layers. For example, pencil tests are initial drawings which are later used to draw clean lines, and eventually hidden from the final view.
Temporary lines (dotted lines above) might also be used to define boundaries of colour areas, and then hidden from the final view. In raster mode, for clean results, it is necessary to put colours in a separate layer so that they do not mess up with the line art. These are some of the reasons why Pencil introduced layers.
Layers are like transparent sheets of paper on which one can draw different elements that can be shown or hidden by placing or removing the sheets. For convenience these sheets do not have a definite size, it is possible to draw anywhere, as far as one wishes. Because of their different nature, both raster and vector graphics could have been contained in a single layer, but I decided to create separate kinds of layers for raster and vector graphics.
While it makes sense to have different characters or backgrounds on separate layers, it is also clear that for a given character, line tests, clean lines and colours are just successive steps of a single entity. It might become confusing to have so many layers for each step. I considered two solutions:

  • Probably the most generic solution to this problem is the ability to group several layers into a single one, visualised as a single entity (for example, a character) and whose internal layers can be displayed when necessary.
  • A second solution is to define within a layer several “sublayers” corresponding to each step, namely: sketch, clean, fill.

The first solution is very generic. It gives a lot of freedom to the user to organise layers the way they want, but I am afraid this solution is not intuitive enough for everyone, and might still be a hassle to deal with (having to create and group layers by hand, select the appropriate layer at the appropriate time many times, etc).
I therefore advocate the second solution which is slightly more preemptive, but will probably simplify the user’s workflow. It is explained in the next section. That being said, the idea of grouping layers should also be implemented in the long run as it is needed for other purposes,
namely making different layers behave in a similar way. I will detail this idea in section 4.2.4.

4.1.2 Sublayers

The idea is that for any drawn element (such as a character or part of it), there is a single layer, which contains 4 sublayers:

  • sublayer for line tests (sketch)
  • sublayer for clean lines (clean)
  • sublayer for temporary lines (temp)
  • sublayer for colour fills (fill)

In terms of rendering, sublayers are ordered as follow: sketch, fill, temp, clean, with the clean sublayer being the foremost sublayer. (Note: these names are provisional).
A GUI element (in the display options or on the drawing canvas) will indicate and let one choose which sublayer one is working on. Selecting one drawing tool will often select the appropriate sublayer. In other words, most drawing tools keep memory of which sublayer they are acting on.
While this can be changed by the user, drawing tools will by default act on the most useful sublayer(s), thus a beginner will not need to pay attention about sublayers at first, and will then learn their meaning as they use the tools. This should also save the user some time.

The GUI element may look something like this:

pascal-vision-d-013

where the triangle indicates which sublayer is active, and the sublayer buttons can be used to show or hide a particular sublayer. In this example, the fill sublayer is empty, the sketch sublayer is hidden, and the temp and clean sublayers are visible. The clean sublayer is active. Hovering the mouse below one of the non-active sublayers will make a grey triangle appear. (This interface is inspired from the channel selection in MidiSwing). Text labels could be replaced by icons if they are suggestive enough.
When a camera layer is active, sketch sublayers of all layers will be hidden automatically unless clean and fill sublayers are empty. This should save time for the user as they do not have to hide and show each sublayers manually.
Note of implementation: in vector mode, the sublayer separation “naturally” exists because all different vector elements are distinct entities. On the contrary, different elements of a raster graphics are parts of the same of pixel matrix and there is no way to distinguish them, apart from storing them in different matrices (raster). The notion of sublayer is implemented differently for raster and vector graphics but should appear essentially the same for end users.

4.1.3 Drawing tools

The number of basic drawing tools is intentionally limited for simplicity. These tools should behave more or less the same, whether in raster or vector layers. The precise behaviour might not be exactly the same, but it is important that the goal and meaning of the tools are the same. Ideally the vector tools should change as little as possible the input from the user (i.e. a stroke should not change shape just after being drawn), to recreate the sensation of a faithful drawing tool.
Other tools more specific to raster or vector graphics may be available as an advanced mode. Tools often found in other programmes such as geometric figures (rectangles, ellipses, polygons) and text are intentionally absent from the set of basic tools, as they provide little benefit to traditional animation. They may be provided as optional advanced tools (or as basic tools in other usage environments such as drawing and note-taking).

Pencil

The pencil tool is defaulted to the sketch sublayer.
In the sketch and clean sublayers, it operates as a usual pencil. By default, it has constant width and opacity depending on the stylus pressure. Some extra options, such as a slightly shaky appearance due to the simulated graininess of the paper, would be nice (plugins?).
Pencil
In the temp sublayer, it draws dotted lines which are displayed only when the fill sublayer is active.
These lines are used to draw invisible contours of areas to be filled with some colour.

Pen

The pen tool is defaulted to the clean sublayer.
It is used to draw clean lines. By default it has variable width and full opacity with stylus pressure.
Pen
Some extra options, such as a slightly shaky appearance due to the simulated graininess of the paper, would be nice (plugins?).

Lines

The line tool is defaulted to the clean sublayer.
It is used to draw either straight lines (when two points are clicked), or curved lines (when more than two points are entered). This is particularly convenient to draw clean lines when a tablet is not available, and only the mouse can be used.
Lines
The first two points make a line. Next points make a smooth curve, unless there are sharp angles.
Double-clicking stops the line.
It should have the same options as the pen tool.

Brush

The brush line is defaulted to the fill sublayer.
Its aim is to colourise large or small areas. In the current version 4.4 of Pencil it is shaped as a disc with adjustable border “feather” (fading gradient). A larger selection of brush shapes could be added as optional plugins.
Brush
Apart from the shape options, two drawing options are available to help colourise areas using the brush tool.:

  • stop at contours: the effect of the brush does not exceed the limits given by nearby lines (clean or temp).
  • preserve alpha: the brush does not affect empty areas.

Note of implementation: this tool can be implemented in vector mode. Vector brush strokes can be painted either by recreating exactly the drawing sequence that is used in raster mode (a sequence of “brush touches” separated by a constant distance) or by painting a regular vector pen stroke and adding a blur corresponding to the feather radius. As for the “preserve alpha” option in vector mode,
it can be done by associating to the brush stroke the areas located under that stroke, and clipping the rendering of the stroke inside those areas. Implementing the “stop at contours” option in vector mode looks more tricky. It may be temporarily replaced by the contour fill tool.

Flood Fill

The flood fill tool is defaulted to the fill sublayer.
It fills an area with the active colour starting from the clicked point. The flood fill stops when it hits
the lines drawn in the clean and temp sublayers (if visible).
Flood Fill
One big problem with this tool is that the flood fill might go outside the intended area if there is a gap in the line art. While it may be possible to introduce some gap tolerance, this is complicated to implement and not always satisfactory. A common solution for the user, is to fill the gap manually with temporary lines. Another solution is the contour fill tool.

Contour Fill

The contour fill tool is defaulted to the fill sublayer.
To my knowledge, this tool does not exist in other programmes. The way it behaves is that when the user draws a closed curve with it, the curve tries to match the line art (clean and temp curves) if it is close enough, and then the area delimited by the curve is filled with the active colour.
Contour Fill
Currently (as of version 4.4) it is only implemented in vector mode. The raster mode version might be slightly different, for example like the brush with the stop at contour option enabled.(Is there a way to combine this tool with the usual brush tool?)

Colour Pick

The colour pick tool is defaulted to the fill sublayer.
It just picks the colour of a particular point on the drawing area. In raster mode, it picks the actual
pixel colour. In vector mode, it picks the colour reference of the pointed object.

Finger

The finger tool is defaulted to the clean sublayer.
It deforms the graphics according to the pressure and motion of the cursor. In raster mode, this can be used to blend areas (standard smudge tool). In vector mode, this can be used to adjust the shape of a curve. Note: although the current version (v0.4.4) displays vertices (control points) for debugging reasons, this is not the intended goal of this tool. It should be essentially an intuitive tool for deforming shape by dragging in portions of lines (with some kind of fading magnet attraction).
Editing of the control points is to be provided within optional advanced vector tools.

Eraser

The eraser tool does not remember the sublayer ; it operates on the currently active sublayer.
It is used to erase any drawn element. Ideally, it should erase just what lies under its shape. The basic shape is a disc of adjustable width. Eraser shapes may also be extended with plugins, similarly to the brush tool.
The behaviour of the eraser tool in vector mode is tricky. Ideally, it should be the same as in raster mode. Just the part of a line which is affected by the eraser should be removed. However what to do if that line is the contour of a colour area? If some part of an area is erased, it means that a new invisible contour should be created to reflected the new shape of the area. This is somewhat complicated and not so useful in most cases, so it is better to delete the area altogether (as is done in current version 4.4).

Arrow: Select/Move

The arrow tool does not remember the sublayer ; it operates on the currently active sublayer.
The arrow tool is a multipurpose tool to select and move parts of a drawing. It operates slightly differently in raster and vector modes.
When one clicks on the drawing area, in raster mode, or if there is no vector object at that point, the tool acts a kind of lasso tool which makes a selection by drawing a closed curve. On the other hand, if there is a vector object at that point, the object is selected.
Once a selection is made, it appears as rectangle with handles. The handles enable to either rescale of rotate the selection. Clicking and dragging inside the selection enables to move the selection and its content.
Click-and-release outside the selection removes the selection rectangle without altering the selected content.

4.1.4 Colours

Colours should be available as collections of colours with their tag names, which are called palettes.
Each element of a palette is a colour reference.
This system is needed to be able to quickly reuse the same colour. In vector mode, this also makes it possible to readjust colours which were already painted on a drawing.
Colour reference can be added by pressing a plus button. A suitable (customisable?) interface is provided for readjusting an existing colour reference (RGB and alpha).
Picking a colour with the colour pick tool in raster mode, makes a new colour available temporarily in raster mode for the active tool, without altering the currently active colour reference. To add this colour to the palette, the plus button should be pressed.

4.1.5 Colour gradients

A first and obvious method to create colour gradients is to use the brush tool with some large enough feather.
In vector mode, there are two other methods.
One method is to select a temporary line setting the boundary of a colour area, and assign a feather to it.
Colour gradients Method 1
The second method is to select an existing colour area and assign a feather to it.
Colour gradients Method 2
Ideally, the gradient should be centred on the temporary line. However, in the current version
(v0.4.4), the gradient actually starts from the temporary line outwards. It is implemented by blurring
the colour area and clipping its rendering to the area behind it (if any).

4.1.6 Operations

In addition to the drawing tools, there should be a set of operations on the drawing or selected
drawing.

  • Copy
  • Cut
  • Paste
  • Clear

(this used to be in the tool box, but should be removed as it is more an operation than a tool)
These operations appear either in the menu bar (mouse usage) or function bar (tablet usage).

4.1.7 Move/Zoom/Rotate view

In the current version (v0.4.4), a hand tool is present with the other tools, and lets the user move, scale and rotate the view. It is not quite satisfactory. It is not really a drawing tool, so it might be better to put it elsewhere. Also, it currently relies on the keyboard or right-button for scaling and rotation.
It might be more logical to have separate move/zoom/rotate view functions available somewhere inside the drawing view. Different ways might be considered so that it is easy to use either with a stylus, keyboard or a mouse. The ideal situation would be to use fingers for move/zoom/rotate the canvas while using the stylus to draw, in the case of dual pen/touch screen devices – although such devices are only emergent at this time. Some tablet devices also feature a special button for zooming.
To be meditated.

4.2 Animating

4.2.1 Timeline

The make animation we need to arrange drawings along a timeline. Since we have different drawing layers (raster or vector) their names are displayed on the left side of the timeline. The right-side is the timeline itself, displaying properties as a function of time. Playback controls are located on the top-right side of the timeline window.
Timeline
In addition to drawing layers, one can also insert other kinds of content such as sound layers, video layers, possibly 3D object layers. We will refer to all layers that can be seen in the view window (raster, vector, video, 3D) as visual layers. A special kind of layer called camera layer makes it possible to have a continuously changing view of the animation. When a camera layer is selected, it
shows the view corresponding to that camera at the currently selected time.
(In the long term, especially if 3D object layers are introduced, it could be interesting to have 3D Camera Layers letting change the view in a 3D space. In that case, positions of visual layers should be assigned not only in 2D but also in 3D with some extra depth parameter z. This may require a lot of revisions of the current implementaion of rendering the view. This obviously needs some deep
thinking.)

4.2.2 Playback controls

The playback controls are very standard: play/stop button, rewind, next image, previous image, loop on/off button, sound on/off button, and a frame rate selector.
Playback controls

A time/frame indicator is welcome too. Next to the value, a drop-down menu showing the time unit (sec, min, frame, etc) could be used to change the time unit. This should dynamically change the display of time in the timeline. If units of frame are chosen, the display will depend on the frame rate specified in the playback controls.

4.2.3 Layer tracks

Each layer may have several properties, whose time dependence is represented by a track. A track is line containing elements indicating changes of a given properties. A dropdown menu in the right part of the timeline window enables to choose which kind of track, i.e. which kind of track, is to be displayed in the timeline. If a layer does not have the chosen kind of property/track, it is simply greyed out. Possible properties/tracks are:

  • Content
  • Position
  • Opacity
  • Volume

For example: for sound layers, opacity, and volume are the same ; position might be undefined (or maybe corresponds to stereo balance). For camera layers, content and position are the same; opacity and volume are not defined.
Each track has its own representation and interactive system. There are three main kinds of representations:

  1. Numerical value property: dots indicate the value of the property at a given time, and lines continuously interpolate between these points. Examples of this kind of tracks: volume, opacity.
    pascal-vision-d-024
  2. Non-numerical value property, or set of several numerical values: small rectangles, usually referred to as a keys, indicate new changes of properties at certain times. Example of this kind of track: content of vector and raster layers, position of vector/raster/video/camera layers. For some tracks, double-clicking a key brings a dialog window where the properties at that particular key can be edited.
    pascal-vision-d-025
  3. Media content property: long rectangles indicate the clip start and end of some media (sound, video)
    pascal-vision-d-026
4.2.3.1 Position tracks

Position tracks are important because they allow to make basic animation by simply moving a visual layer across the view. When the position track is selected in the track menu of the timeline, all visual layers are shown with some kind of celluloid/plastic rectangles showing their boundaries.
The selected layer should be shown differently to indicate it is selected. Each rectangle can be moved freely (translation, scale, rotation) using the same handles used in drawing mode when moving or resizing an object.
Whenever a visual layer is moved, if no key exists at the currently selected time, a new key is created. Positions between keys are always interpolated. Similarly, positions of a camera layer can be changed by changing the view at a particular time, when that camera layer is selected.

4.2.3.2 Effect tracks

In addition to the basic tracks, all kinds of post-rendering effects could also be added (as plugins) to the list of possible tracks. Typical effects would be: blur, colour adjustments, etc. Effects can be set at different times in a track and would usually be interpolated.

4.2.4 Layer hierarchy

The layer name section is a standard tree list. One can rearrange the order of the layers, as well as
making some layer a child of another layer by drag-and-dropping it on top of that layer. Children of
a layer can be hidden and shown by clicking the icon of the layer.
 Layer hierarchy
When a layer becomes a child of another layer, its positions at different times (if a position track exists) are recalculated with respect to the parent layer, so that the child layer then moves together with the parent layer. This way a parent layer and its children behave as a single entity.

4.2.5 Onion skin

4.2.6 Interpolation/In-betweening

A good animation is achieved by drawing many pictures. Usually some key pictures are drawn and in-between pictures are then drawn to complete the motion. Drawing these in-between pictures is quite tedious. To some extent this task may be done automatically, but the result may not be as good as if it were done manually. Indeed, in-betweening between two pictures requires some understanding of the picture, which a computer usually does not have. Nevertheless, automatic in-betweening may be useful and save some time.
Implementing in-betweening looks a bit tricky and requires some thinking. As the computer does not understand the drawings, the user should help it by indicating which parts are related in the initial and final pictures. Typically, I imagine some interface whereby the user would draw some line or contour, both in the initial and final images, to establish a correspondence. The more the user draws such correspondence contours, the more accurate the in-between frames become. Eventually retracing the whole drawing would lead to a perfect in-betweening, but would amount to almost twice the effort of drawing a in-between picture manually. Thus, a trade-off between interpolating quality and effort may be achieved.
In raster mode, this could be implemened by morphing algorithms. One inevitable side-effect of this approach is that the interpolating pictures often look like a superposition of the initial and final drawing. The vector mode should allow better interpolation since the line art information is available, but the algorithm is yet to be determined.