All videos are used for citing. Copyright reserved.

There is an opinion that transitions are the most impressive part of motion design.

As a rule, when it comes to transitions creation, only the simplest ideas come to mind and they could hardly be called ‘impressive’.
Motion designers come across a creating transition problem every day. Like any other motion designer I had some difficulties. I thought every transition I’ve created wasn’t interesting enough.
The main problem of transitions is that there isn’t any information about possible variations. You can find tutorials on the Internet which describe some single examples, but they mostly depict the technical point and they have nothing about the art direction. And you’ll hardly find the explanation why a certain transition would or wouldn’t be suitable for a particular case.
It was the art direction that caught my interest. The questions ‘Why do I use one or another transition?’ and ‘How to make a reasoned explanation to a customer?’ made the main point of my research.
I started to pick out videos with interesting transitions. After a 2-year research I’ve come to a classification with 21 unique kind of transitions.

The main goal of all transitions is to make a narration visually interesting and catchy. Explainers use transitions to hold viewer’s attention during the whole video.

“Merit” video

The broadcast dynamics is achieved due to transitions; it makes animation look spectacular.

Comparing how explainers and broadcasts achieve dynamic, we can see interesting facts:

Explainers have the dynamics due to action scenes;
Broadcast scenes are very short, that’s why transitions add most of dynamics.

In order to understand all this information, it was time for me to bring all the selected transitions together on some basis and put them into groups to study.

I decided to begin with shutters.

Shutters.

Before I tell about shutters, I’d like to show you some points that were spotted during classificating.

The main feature that differs all shutters is their form.

Linear shutters are the simplest; they are formed of plain lines. More complicated shutters are made up of figures. And the most complex have figures forming compositions.

So, shutters can consist of:
lines;
figures;
compositions;
objects (an object overlays the scene like a shutter).

I’d like to draw your attention to object shutters.
The first object shutters were most probably used in cinema. Scenes full of action would lead stage directors and cameramen to the idea of using a bad frame with an actor or an object overlaying the view as a good way to change scenes.
And now, in the age of computers, alpha channels and masks, an object shutter has been improved to the kind we know it today.

If we compare an object shutter with the other ones, we will see that it has one interesting feature. Only the objects that have already been in the scene are used as an object shutter. It looks very inventive and allows to achieve maximum expression with minimum means.
Linear, figure and composition shutters need an extra object to make a transition, whereas object shutters use objects of the scene.

On the other hand, linear, figure and composition shutters are considered to be universal transitions: there’s no need to plan the scene thoroughly trying to find an object to make a transition. This is a great advantage, you can use these shutters at any moment, but they don’t look as impressive as object ones, though.

Some more interesting facts.

If a person or an object related to that person is used as a shutter, all attention is attracted to that very character.
If an object shutter is formed of an object of the scene, it can draw attention to the time flowing.
As a rule, for this purpose one can use scene objects that are far enough from the camera. At the moment the object overlays the scene, the camera movements make parallax. This is what allows viewers to consider the transition as a change of time and location.

Looking through a large number of shutters I noticed that an object of the frame has a tendency to copy a motion vector of a shutter after its passing. It gave me an idea to use a shutter as a part of a dynamic balance of the composition.
When designers speak of a composition balance, usually it’s a question of statics. But when we work with animation, the balance of the composition develops in time and is never static.
For more understanding I suggest to watch the example.

Why is the dynamic balance so important?

If objects are moving, the viewers’ attention will be concentrated on the very objects and on the side of the screen, where these objects have left the frame.
It is necessary to switch the viewers’ attention for more comfortable perception of the video. The simplest way is to make another object move on the opposite side of the screen; this can restore the balance and make your composition look more attractive.

The main feature of the shutter.

If we speak of the shutter from the art direction point, it’s main feature is the ability to detach one theme from another.
The shutter effect is called so due to the visual similarity, when shutters overlay the previous composition imitating covering a window with a curtain.
Also, after overlaying the previous composition, the shutter can be a brand new scene for another story.
There’s one more certain similarity – we can create a new story after turning over a page of a book.
For example, explainers are structured to reveal themes one after another, and one can use shutters to switch between them. It clearly shows that one theme has finished and another has started.
So, if you need a way to make a transition from one scene to another one, the shutter is the best decision with minimum time wasted.
It is necessary to understand that the more complicated the shutter is, the more time you need to create it. If we compare all the four types of shutters stated above, we will see that linear shutters are the fastest to create, and composition ones are the slowest.

Is there a difference between cinema transitions and motion design transitions?

The basic type of transitions in cinema is a montage transition. Cinema needs montage to create the illusion of seamless narration. Films are shot in many takes, but if all the rules of montage are kept, the film seems like a smooth story to the viewer.
If the rules of montage are broken, the viewer begins to pay attention to seams and becomes distracted from the movie.
If we compare the montage transition with any other one, we can see that perception of montage transitions needs the least of attention resources.
There is a huge number of montage transitions in movies, and we have enough power to keep our attention to the film. But there are rare cases of creative transitions in cinema.

Back to the question: what’s the difference between cinema transitions and motion design ones?

The answer is that cinema needs hold the viewer’s attention throughout the film. The viewer’s attention has been already focused, because we`ve already decided to watch or not to watch the film. That’s why the main thought that runs all through the montage is to make all seams less evident and less abstractive.
Of course, there are some special transitions in movies, but they are used to put an emphasis on something and to attract the viewer’s attention to the moment. This special transition looks more spectacular than an ordinary montage one.
It’s obvious that these special transitions are very rare in cinema, but there’s a quite opposite situation in motion design. Every single transition is desired to be unique and catchy.

Why is it so?

The motion design is all about attracting viewer’s attention to tell about a material;
The main task of cinema is to hold the viewer’s attention on a solid story.

That’s why motion design uses the most interesting transitions for attracting viewer’s attention and cinema creates the simplest.
Motion designers can’t use such simple effect as a montage patch, when the next scene has nothing in common with the previous. Because if the viewer can’t find an answer to a question ‘How are the scenes connected?’, he loses interest. And we lose the viewer.
The situation is quite normal in cinema, where scenes change each other having nothing in common at all; all these patches are bound by a movie plot.
In order to omit such difficulties of scenes correlation, motion designers have some tricks to hold the viewer’s attention on a story.
The trick is to transfer some elements from one scene to another to make it look like a continuation. You can transfer an object, a background, a motion vector by making a patch.

Montage transitions typical for motion design.

the same motion vector, the same objects
the same motion vector, different objects
different motion vector. the same objects
different motion vector, different objects

The more objects are transferred, the more dynamical patch we have.
This is because while watching a new composition the viewer sees some elements left from a previous one; this makes the entire scene easier to apprehend. As a result, active perception isn’t switched on, and we don’t have the necessary dynamics.
If we show a new scene without any connection with the previous one, the viewer needs to devote all his attention to the new scene, and that rises the whole perception dynamics.
I also noticed there’s a common usage of cutting at the moment when an object completes its action. We can call it ‘the completion expectation effect’.
When we watch a scene with an object acting, we know instinctively when this action will end. And we can make a transition at this very moment, because it will correspond with the viewer’s inner time; that’s why the transition will look appropriate and relevant. We can compare it with a music and video synchronization when making a montage. But here we have a picture instead of music that makes rhythm by complete actions.
Notice that this effect can be used with other types of transitions.

The Glitch

Then I noticed that such transitions as a glitch are very similar to montage ones, but they consist of intermediate frames also.

The Glitch

blurring
squares, rectangles
a lot of various intermediate frames

The power of this transition is in immediate switching of scenes, just like montage transitions. However glitching attracts attention. Microanimation lasts not very long, but it is enough for intensify attention.
At the moment of changing scenes we can observe some sort of magic; the viewer sees the next composition and it seems to him more beautiful than it actually is.
Our perception sees any moving object, no matter we want it or not. Microanimation gets our attention. As the viewer notices something’s going on, microanimation immediately ends, opening the next composition. Then comes the composition itself following the glitch. It should be interesting enough, otherwise the effect of perfected perception won’t happen.
The main aspect of the glitch from the art direction point is the expression of aggressiveness and technology.

‘NBC’s The Blacklist ‘Classified’ Trailer’ video

Object transformation

At first, I’d like to show you the variety of object transformation.

Looking through a huge amount of various transformations, I noticed that they can be divided into two big groups based on the way of working with a form.
One group uses forms for transformation, other uses the whole objects.
The whole objects transformations have such properties as ‘Position’, ‘Scale’ and ‘Rotation’.
And then I came to an idea to name this type of transitions as ‘the PSR’, so everybody could understand the way the object transformation goes.

And now we’ll have a little talk about variants that change their form while transforming.
As the transformation needs a form’s change, I decided to name it ‘the Path Transformation’.
A form’s change can happen on an outer parameter (‘Stroke’) and through the masses, for example, when some part of an object changes. Or there’s a mixed way when both variants are used simultaneously.

The PSR

Looking through all the variants with a whole object transformation, I’ve formed 6 groups.

The PSR

PSR scale
PSR rotation + scale
PSR rotation + one of the object sides serves as another object
PSR as a single frame
PSR rotation + as a single frame
PSR as a single frame + forms matching

As we see, both simple and rather complicated varieties exist among PSR variants. It is obvious that variants with forms matching are the most spectacular. Such transformations cause a ‘wow’ effect, because they emphasize common features between objects.
During a work with objects within a script, it is rather difficult to see these features, as objects can vary so greatly, and there’s a big work to find something common in their forms.
And since finding matching is a rare case, we can see transitions with a forms correspondence not so often.
Other PSR transformations are simpler, and we can use them almost in every combination; that makes these transformations universal.
The strong side of the whole object (the PSR) transformations is connected with the fact that we see the B object almost immediately without intermediate morphing frames. This means the object appears suddenly, that’s why we should imagine the B object clearly.
If a new object is interesting enough to the viewer (it touches particular themes), then such transformation gives a lot of emotions even being simple enough.
If an object is rather boring, such transition will look too simple.

The Path

A change of a form can be reached by different ways: by smooth flowing or a form fragmentation.

The Path

the Path change of a form (smooth)
the Path change of a form (smooth) + form matching
the Path fragmentation of a form

If we compare the PSR and the Path, the latter has much more variants of transformations. It has the same object manipulations just like the PSR does, but there are some more variants of form change added, that’s why the Path has a huge number of final combinations.
And it is obvious that making a Path transformation needs more time than making the PSR variants. We should take it into account while planning our future project.
When making a Path transformation, we should notice what an illustrative object we use. If the object is simple enough visually, we won’t get good dynamics, because there isn’t enough elements for rich animation.
When we work with a Path transformation, it is better to use compound objects (illustrations).
All the beauty of the Path is concentrated in a morphing process and there’s no matter what emotions can the B object produce because all visual sensations depend on a morphing process itself.
Also the moment of perception overloading is rather important when there are a bit more elements moving from A to B during a morphing process, than the viewer can follow; so the viewer has an impression of rich animation.

Where do we use the Path?

If we speak about the Path usage from the stylistic point, we see that it appeal to our imagination. We can’t change a form of an object the way we want it in real life, this is possible only in our imagination.
So the Path is the best way to depict a digital or a fancy sphere.
The Path usage is justified when the object consists of a great number of elements. A lot of elements make the dynamics active, so the composition looks more attractive.
There’s another case of the Path usage, when the B object is too simple, and any of the PSR variants can’t make a transition interesting enough. Every object can be prettified to be catchy just by using a morphing or a defragmentation process.

When do we use the PSR?

From the art direction point, the PSR transitions are best suitable for making a restrained look.
Unlike the Path, the PSR transitions allow to transform objects without the hustle that causes a morphing process when using the Path. Or when the B object is so good that it doesn’t need a morphing to make a transition ‘awesome’.
That’s why the PSR can be disguised as a strict and restrained and maybe deliberately simple design.

How many of them?

As there are more and more transitions found, I came to a question ‘How many of them do exist?’. Having all possible transitions available, we can choose a perfect one for a particular case, suitable by the sense, dynamics and time.
Sorting out the transitions, I’ve marked out so called ‘clear’ transitions that have one plain feature. This was made to simplify the understanding of the transitions structure. But the vast majority consists of combined transitions.

When we create a transition, we don’t think about it’s properties. We do it only using our empirical experience and realization.
It is hard enough to understand at once the concept of combined transition, that’s why, at first, it is necessary to understand simple transitions and then it will be easy to sort out parts of combined ones.
To answer the question ‘How many transitions are there?’ we should proceed from simple transitions to combined ones.
As yet I’ve found 21 kinds of transitions that can make more than 100 combinations of various difficulty. I didn’t have enough technical feasibility or practical need to check.
But I’m sure there’re some really tremendous transitions among them.

If you want to repeat any combined transition, you should have the same set of elements as in an example to create it.
We can’t always replicate every transition just because we don’t always have this kind of objects or that background or we just don’t need such motion, etc.
The point is that all combined transitions need a prep work. You need to plan all illustrations and compositions in the scenes, having these combined transitions in mind.
In order to study combined transitions in details, we may make up a list of aspects that we should take into account while creating a combined transition.
Having this wish list, the motion work can be much more effective; this allows to plan the scenes composition, illustrations, object motion, etc.
Of course, not every transition needs a prep work. For example, if we compare any complicated combined transition to a shutter, we’ll see that the shutter appears as a universal transition. The idea is that such a transition is possible at any moment and doesn’t need any special conditions.
Also we can check the list to sort universal transitions out. Maybe it will be useful for those who is limited in work time. When it’s not enough time to think over, to look through,to twist and spin any variants, we can use our list of universal transitions.

The Pipeline

Concluding the transitions review, I’d like to tell you that to get a set of 21 (~100) transitions is quite good.
In fact, a more interesting thing is that knowledge of transitions helps to think over a work pipeline.
Having studied the transitions more carefully, we can sort out the moments where and how we can use this or that transition. Being able to understand it, we can plan from the very beginning what transition to use.
Any script is a set of thoughts that are configured to flow one after another. They can last from one scene to another and they can initiate new ideas. Some transitions help visually continue the thought, and others represent its ending.
For example, shutters end thoughts, and an object transformation tells about a continuation of a plot.

What’s the awesomeness of this method?

At first, having already started a project with a set of transformations we’ve chosen, we can see what dynamics will be in a video.
Secondly, our colleagues and we can discuss our future decision about transitions and find a perfect variant suitable for both sides.
And thirdly, planning transitions at the beginning we can time our animation production; this gets an extra control over a project.
I haven’t seen any case with such a preparation to a motion design project before…
I wonder what the community thinks about it…

I also realize this isn’t a complete classification of transitions.