final + evaluation

So here is the final video and the code for the animation:




Things that I like:

• I'm happy with the overall aesthetic and style of the animation combined with the 2d graphics
• The motions has a nice organic quality, and the movement is fluid and interesting.
• The forms relate nicely to both the movement and to the underwater context of my final video. 
• I like the pace of the video, and the way the movement in combination with the music creates a positive mood, where the forms seem to be happily roaming about together.

Things that could be improved: 

• I'm not totally happy with the final image quality in some parts, which is a result of slightly enlarging my code image sequence in after effects for the close up shots. While I know this could have been avoided by not creating the close ups, I felt they were necessary for making the overall video more interesting and appealing (one shot for the whole thing would have been boring).
• This problem also could have been fixed by creating different variations of my coded file, where the forms were bigger for the close up shots and smaller for the distant shots. 
• I think that the effect of the layer mode I used to composite the pyp file slightly took away some of the colour from my forms, but I didn't want to do colour correction and adding colour in after effects because i wanted to keep my coding and the decisions i made there in tact. 
• I think my forms could have been taken further with the code, and made more unique, but I also think the simple look works well with the 2D graphics in creating a simple, but interesting composition. 

sound

I felt it because i've essentially created a full video with the animation, it really needed some sort of sound or else it feels a bit strange to watch.

I didn't want to apply a specific sound effect to my forms movement, because the entire animation is essentially the same movement and it would be a weird repeated sound over and over.

So i instead opted for a music track based idea, and tried to make sure the track i choose was not too overbearing, and that it also has the right mood for the animation.

I choose a royalty free track by kevin macleod, and it has a really nice, flowing, positive tone, which reflects the fluid and energetic movements of my forms.

I paired this with an underlying ambient track, which is an eerie, ocean like sound composition, taken from free sound.org. (Both of these are credited in my video on vimeo). This sound effect reflects the eerie, somewhat scary environment of the deep sea. It then contrasts with the positive nature of the music track, in the same way that the colour/brightness of the forms contrasts the darkness of the environment.

lighting and masks

I created a mask around a duplicate layer of my foreground, and then feathered it, to create the area where the light from my create will "bounce" off.

I then tracked the path of my forms using a null object, which I then parented to a couple of feathered circles, which mimic the colour of my forms. This creates two coloured ellipses that follow my forms.

So that these ellipses dont effect anything else, I used a track matte in after effects, to have them appear only on the alpha channel of the fore ground layer.

I also wanted the composition to be more dynamic, so I added in some light layers in the background, to mimic the way light rays sometimes travel through the ocean from above.

In the final animation, they also drift slowly to the left/right, to add some subtle dynamic movement to the whole composition.

2d graphics

2d graphics that made up the foreground, mid ground and background of the final. Also the anemone like create/plant for the added colour in the scene (I adjusted the colour later in after effects).

context 3

After my test, I then created a large 1280x720 version, as this is the size i want for my final video. I'm choosing this size, because I want to have a cinematic feel, and a 16:9 ratio helps with that. It also means it will look really good when uploaded to vimeo, and will also match my other video projects if i was to put it into a show reel.

Here is the test:
(You'll have to ignore the play button, I just took a screen shot of the quicktime movie.)

I showed this to Elle and came out with a few things to work on, which included:

• - the brightness and saturation of the form is too different from the context, so it's too apparent that they are just separate layers composted on top of each other 
• - we weren't sure if crabs lived underwater, and Elle suggested something like sea anemone's instead, because they could also be coloured similarly to my forms and help to integrated the form more effectively into the environment through the colour similarities.
• - it would make sense, in a deep sea environment, for the glow of the forms to effect the environment
• - the 2d stuff is a bit static

context cont

So, i've decided to run with the same 2d style that i tried out in the beginning, but apply it to a deep sea context.

I think the simple 2d style of imagery for the ocean compliments my forms well. It doesn't distract your attention away, but it creates a frame and context which gives the forms so much more meaning and interest.

I'm working on it so that there are rock like structures in the foreground and background, so that my animated forms can move in the middle ground, and move dynamically behind the foreground layer. 

Here is an initial composting test screenshot:

The layers are probably a little dark at the moment, but this was just a quick test to see if it would work.

adding a context

Now that I've started to decide on exactly what i want with my form, I've started to think about how I can put it into a context.

I don't really like the idea of just having them moving around on a black/plain background.

So i've gone back to my initial desire to create a combination of animation and 2d graphics (like in my northern lights test at the start).

Because my form has become a little more creature like, it doesn't quite fit as well into the context of the night sky (in my opinion), as much as something that looked like the northern lights would.

However, I felt like they do look a lot like some of the creatures at the depths of the ocean:

summing up so far

So after all the experimenting, i've decided on a few things:

- I want to keep it simple and not go overboard just because it's easy to with the code
- I want to focus on the relationship between two forms, rather than having 3 or more
- I want the colour scheme to remain subtle, in particular in the difference between my two forms

I also want to make sure that the movement of my forms is organic and smooth, and that they appear to move in relationship to each other.

I guess I have gone round in circles a lot with my experimentation, but I think it is helping me to define what I want, and what doesn't work.

colours 2

The reason for trying the green/red colours scheme was i guess, kind of simple and basic. The general thought was that if i was to have to forms interacting with each other, i should try and distinguish between them, and the obvious way to do that is to have them as complimentary colours. It's something used everywhere, so i thought i'd try it out here.

I'm not that keen though on the way the colours look when they come together, which i mentioned earlier. So i tried an experiment where i had a third white form in between the two colour forms, to try and separate them out a little. 

Test 1:

However by this point, I've started feeling as though the blue-green/red contrast is too strong and blunt, and i want it to be more subtle. So i tried substituting the red for a blue-green colour that was very similar to the other blue-green, but different enough to distinguish it (much like i tried out earlier in my colour tests).

Note: these animations are still only involved the 3 forms, i've just taken 2 screen shots for each and composed them together.

focusing back on 2 forms

After the experiments with 3 or more forms, I decided i prefered to have only 2, and to work on the relationship between them. I want them to move about the screen together, in a way that suggests a playfulness between them.

I decided to also look at how i could use hierarchy to make this more interesting, so I tried having one form larger than the other. I wasn't a huge fan of the way the larger form takes over the other. It has the potential to be visually interesting, but in my tests so far it just seems a little messy to me.

You can see i also returned to the green/red colour scheme. 

3 forms at once + hierachy

I also began looking the prospect of having 3 or more forms at once. However I found that it made the code lag a lot, and i couldn't figure out how to improve it.

I think there is inevitably problems with my code not being the most efficient it could be, but it's hard for me to troubleshoot what could be better, but i am trying!

Here are a few different attempts at having 3 forms of the same size (each image is two different attempts at the code, i've just put them together in photoshop):

And in this trial, i tried to play with the hierarchy, and  have to forms the same size, and one larger. It didn't really work out to be very well composed or balanced.

bodyless forms

My form so far has been mostly consisting of two parts - the lines on the outside, and the body of ellipses in the middle. This is what has been making it appear like a creature of sorts.

So i decided to go back to what I had been trying earlier, and do some more experiments of the form without the ellipses to see what it would look like.

I guess i was trying to return back to my initial concept based on the northern lights.

However I found these to not be as interesting visually as the forms with a 'body'. I prefer the relationship between the ellipses moving directly along the path, in contrast to the lines which are moving dynamically up and down.

Here are a few trials anyway:

simplified code

I found that in a lot of my experiments, I wasn't getting a whole lot of trail behind my form. This trail is being created by the repeating background having a low opacity (actually its a rectangle the size of the background) and I already had it done really low.

I found out that my simplifying down the code i've been using for the motion of the lines in my form, it actually allows for a slightly longer, more dynamic trail. I don't know the real reason for this, but I'm thinking perhaps it's because the complex code was lagging the program? I'm not sure.. but with it simplified (i.e less sin waves inside of sin waves etc) it seems to be more smooth and have longer trails. :)

There is also a more dynamic and fluid curve when the form turns corners, which I like.

In code, my functions went from:

r = (200. * math.cos(1000*6.28*z+phaseX) +  5. * math.sin(3. * 6.28 * z))  * math.sin(3.14*z)
 q = (200.* math.sin(1000*6.28*z+phaseY)  + 5. * math.sin(3. * 6.28 * z))  * math.sin(3.14*z)

To:

r = ((scale) * math.cos(1000*6.28*z+rphase)) * math.sin(3.14*z) #path of form
q = (scale * math.sin(1000*6.28*z+qphase)) 


I guess it also shows that all of the code in the initial functions wasn't having a hugely beneficial impact on the form i was going for.

colour

I've also started to look at the colour in more detail. The initial colours i've been using have been based off my precedents (I literally used the colour picker in photoshop to take the values of one of the northern lights images). Now, because i've got multiple forms, I'm experimenting with how the forms might differ from each other, and how the colours might work together. There is an interesting (and difficult) part to consider, where the forms cross over eachother, as the colours really have to work together and look good together at that point.

So here are some examples of the colour trials. Some are subtle differences in blue green, and others are more dramatic contrasts between complimentary colours. I've not yet decided which direction to go with.

Reviewing my concept

In reviewing my concept, I feel it is changing slightly. I'm still using the forms, colour and movement of my northern light examples to inspire my form, i'm also now considering the posibility that these forms are not going to be some grand, large scale phenomenon, like the northern lights and that, rather, they will be forms on a smaller scale, like little creatures moving about the screen.

multiple forms

With help from Richard, i've now began to experiment with have multiple forms, rather than just one. When I say multiple, it's really only been two at the moment, because i'm struggling to create smaller forms which stay maintain the look that i want (When i shrink the size of my form, it loses a lot of the defined shape). If i can figure that out, i want to look at create some dynamic hierarchy between the forms.

So here are some examples:

You can also see that i'm continually adjusting the details of the form, such as the amplitude of the lines, the opacity and spacing of the ellipse and so on, as I look for a combination i am happy with. 

I've also started fiddling with the colours to look for a combination i am happy with for the 'body' and 'wing' parts.

2d path and form updates

I've  switched to using 2d shapes as the path (like the exercise example above), so that the form would move in a loop of sorts, but still maintaining a dynamic flowing path.

I also then tried out using variations of the other curves from the exercise as the paths, such as the one create for the golden ratio, as these allow the form to move around the screen in a way that looks slightly more random and spontaneous.

The path also effects how the form moves, here are some screen shots of the form as it is moving through the 2d paths.

As you can see, i've started leaning towards the forms which incorporate a lot of lines, rather than small groups of circles or blurred objects. This is to reference some of the look of my precedent images of the northern lights.

Just to try and be more clear for the work below, each of the little forms was following a curve like the ones in exercise 1. I tried using both a simple sin wave, and waves like the one above, to create dynamic paths similiar to how the northern lights move.

Initial form tests

These are some of the forms that i've created by trying different combinations of sin curves etc, and also the use of ellipses combined with lines.  Sorry they're quite small, they are small on the actually animation.

Progress1

Ok so i've been moving on the idea of some sort of surreal pattern moving through the sky. So what i've done so far is make a simple sin curve as  path across the window, and then i've been trying to find a way to have a form move along that path, while having a path of its own (eg more sin curves).

Basically figured out how it works with this kind of thing:

def draw():
  
    for i in range(50):
       
        noStroke()
        fill(0,0,0,10)
        rect(0,0,600,600)
        fill (200,255,255,5)
        x = frameCount*5 + i
        y = thefunction(x) + 200
        z = (20 * math.pow(1/2. * math.sin (20*x) + 1/2., 2) + 20*math.pow(1/2. * math.sin (20*x) + 1/2., 2)) * math.sin(5*6.28*x)
        q = (20 * math.pow(1/2. * math.sin (20*x) + 1/2., 2) + 20*math.pow(1/2. * math.sin (20*x) + 1/2., 2))  * math.sin(5*6.28*x)

        y1 = y + z
        y2 = y + q
        for i in range (10):
            y1 = y1 + 2
            y2 = y2 + 2
            ellipse(x,y1,50,50)





          

This is a really quick test of an idea that just popped into my mind. What I really want to do, is create a series of 2d silhouttes of interesting landscapes, which will then appear in front of a coded night sky, which will have a series of interesting forms moving around.

The video is entirely a test of concept, and not what the coding will look like at all. There's no real forms or anything, just some 'stars'. It's also spinning illogically in relation to the foreground, but it's not important, I was just testing some things, like the composting.

It's also horrible horrible quality (but uploaded really fast :P) 

On a similar note, i'm interested in the phenomenon that occur with long exposure and timelapse photography. So here are examples of that: 

image sources:
http://www.myknownbuzz.com/wp-content/uploads/2011/06/Star-trails-3.jpg
http://fc06.deviantart.net/fs43/f/2009/152/8/2/8293f9633d18ea4b2a28db62b0f67edd.jpg
http://webdesignledger.com/inspiration/30-stunning-examples-of-long-exposure-photography

P1 Inspiration

After today's lecture, I thought about the idea of composting my python animation with an image. I was also thinking of how I could relate this project to things I actually care about or am interested in. One idea was to do something to do with photography and natural phenomenons, and one idea was to look at the northern lights. I actually was Norway for a brief period over the holidays, and although I didn't actually see them, that's where the thought came from! So here is some inspiration images:

image sources:
http://www.hd-wallpapers.com/download/aurora-borealis-northern-lights_1280x1024_159-standard.jpg
http://apod.nasa.gov/apod/image/0407/moussette_aur16jul1_full.jpg
http://www.zastavki.com/pictures/2560x1600/2010/Nature_Clouds_Northern_Lights_Aurora_Borealis_019637_.jpg

and a video:

https://vimeo.com/21294655

More having a play around now than anything else. 

I've found that by altering the scale of things, the experience and image is changed dramatically.

These are the functions btw, it's all pretty simple.

def xfunction(z):
    return 600. * math.cos(300*(0.05*frameCount)*6.28*z) +  15. * math.sin(3. * 6.28 * z)
   
def yfunction(z):
    return 600.* math.sin(300*(0.05*frameCount)*6.28*z)  + 15. * math.sin(3. * 6.28 * z) 

A also noticed that it made some really cool 3d looking patterns, so I took out the frameCount and manually adjusted the frequency of one of the curves until I created some interested patterns like this.

Just for laughs, I added in a frameCount to one of my cos curves, like this :
math.cos(100*(0.05*frameCount)*6.28*z)

and ended up with a pretty cool animation that continually loops without the need for any extra code!

These are my experiments so far for exercise 1. I know it's more than we need to do, but it's so quick and easy to make changes, I figure why not!

This is a mixture of 1d and 2d functions, but obviously when I choose the 8 for the hand in i'll arrange them in their respective groups. 

A second test post

Second test post, to see if tags work.

Tag Trial

A test post.