• 

• 

• 

For SuperCrit my approach was to have as much to show as possible, I tried to present all my ideas so I could see what might work and what may not for the final outputs. I wanted to present a mockup for a book and a model as well but I didn’t have time – however I am still quite happy with the amount that I managed to get done for this moment, it was a good push.

Overall the feedback was quite positive – people loved seeing the back and forth of the ai/human and found the whole concept of the project very interesting and exciting.

The video was quite successful, it seemed to explain the process well and was engaging to watch. The only problem was just in terms of focus – it was hard to stay focused on two things happening on the screen at once, there needed to be more of a split.

Notes from feedback (before tutor advising session)

• Give AI more rules/constraints? Narrow down to more of a theme?
• VR Experience?
• How can we tell that it is a collaboration from 10m away before – make it easily and explicitly clear.
• Performance aspect to the end display? drawing during etc, showing it in action.
• Installation could be more interactive/less still
• Is the relationship symbiotic or parasitic, that changes the relationship
• Hard to focus on both parts of the video, didn’t realise notice/pay attention to the text in the top left.

I thought these were all quite good points and things that I need to work on/consider. I especially like the idea of considering how I can make the collaboration more clear – I would like it to be very obvious before even reading anything that the artwork was done by two ‘artists’.

Creating a mockup of the installation aspect of the project – for supercrit I decided on digital painting as it was the easiest and quickest way to go through the collaboration process and makes sense in the context of the project.

Below is a proposed layout for one panel of the installation

This will present:

• Input/name used for the world
• Shape generated
• Map generated and mapped onto a globe
• Collateral – Creatures, props, vehicles, plants, textures, etc

Full installation – large format prints with each section layout repeated and displaying a different variation.

Creating a test output for supercrit. The generated name chosen for this one was ‘Hithladreg’ , and below is what was created after a few rounds through the program

• 

• 

• 

• 

• 

I then used my map-making program to create a map based on the input

And mapped it to a globe

Next, I generated a line version of the same shape, to use as a prompt for the extra stuff – characters, vehicles, props etc.

Exploring creating a few different things within this shape

I mostly just created whatever I saw in the shapes but tried to give myself some different goals eg character, plant, vehicle.

Mocking up the concept for a video as an output which would show the process behind the collaboration, and bring the final artwork to life.

Concept/storyboard for the video – I am imagining a kind of ‘conversation’ between human and the computer, showing which is doing each part of the process.

Inspiration:

This video shows the process behind the creation of an art installation WDCH dreams – I like the highly technical feel of the video and the way everything is presented.

I then created a simple video which shows the process of creating a landscape and globe – it shows the program generating and me painting over. The end shot is of the final painting, which I thought may be animated to give the world a bit of extra life. It would also include all the extra things that I have designed using the shape, potentially these would be animated also.

I began testing a few different methods of creating the landscape within the gauGAN program. I wanted the process to go back and forth if possible as I thought this created interesting results when I had tried it previously and it also seems a more engaging collaboration with AI.

First I tested feeding the generated image straight back in to the machine with no input from myself

• 

• 

• 

• 

• 

• 

• 

• 

• 

• 

• 

• 

• 

• 

• 

• 

• 

• 

Definitely produces some fascinating results, however I found that it often seems to loose definition over time and become quite non-descript/undefined.

Next I tried painting over-top each time

• 

• 

• 

• 

• 

• 

• 

• 

• 

• 

• 

• 

• 

• 

More defined using this method

And lastly tried repainting the scene each time

• 

• 

• 

• 

• 

• 

• 

This method seemed to produce the most unique result

I would be curious to know how these would be interpreted had they been painted much more realistically, spending a lot more time on each piece. This however could be quite a length process.

Some other potential ways to test are –

• Photobashing – this is typically how concept art pieces are created, this could yield a more realistic looking piece
• Physcial painting
• The program has the option for style filters – I could input some concept art into this to get more variance, so that the program isn’t limited to making landscape images that work within our planet (ie it thinks grass must be green sky must be blue etc

I would like the initial generations to be based on something rather than just completely random and different each time.

I looked back at my old coding for creative practise project in first year – In this project I created random maps based on an input typed into the program. It did this by transforming the word into numbers and adding them together to create the random ‘seed’. The generation created was unique to the input text and would always be the same.

In order to throw the machine into the collaboration as much as possible I turned to random name generators in order to generate a random fantasy name for the planet in order to use as the input.

• 

• 

• 

For supercrit I decided that I wanted to try and have a mockup of as many different aspects of my project as I could. I have had a number of different ideas for the final outputs and have been struggling to decide exactly which one would work best.

Ideas –

• Large format print/original artwork as a long strip showing different iterations
• Concept art book
• Video showing the process/animated final artwork
• Model of characters

Mocking up some ideas for the installation. This is the one output that I am quite set on currently – I want to have some very nice large paintings at the end, whether that be physical or digital I am not sure.

One idea is that the landscapes could link together, where one ends the other starts but each would like quite different still.

I also thought that each seperate ‘world’ could have a shape that determines all of the content eg landscape, creatures, vehicles etc. This information could be located underneath the artwork.

• 

• 

Shantelle Martin was one of the speakers at semi-permanent that I found quite inspiration and quite relevant to some of the concepts that I have been exploring within my project. Her work has quite a raw/surreal quality to it that I enjoy – it reminds me of some of the exquisite corpse and automatic drawing work I have looked at previously by the surrealists.

Shantelle talked about being spontaneous while drawing – if you overthink, you have time to hesitate and feel insecure about your decisions. Spontaneous is more ‘you’

She also discussed children’s drawings and how they have a level of innate creativity – drawing without limits until taken away as an adult through expectations and restrictions eg told how something should look.

This ties in with what I was exploring around Pablo Picasso and other artists that believed there was great power in children’s drawings and it was beneficial to attempt to return to that raw creativity. I could see the AI renderings as these ‘children’s drawings’ which may help an illustrator back into that state of uninhibited imagination and play.

Day two kicked off with a discussion from Elastic – The design animation studio behind title sequences for shows such as Game of Thrones and Westworld.

One thing I got out of this talk was a big focus on the power of collaboration. This is something that I saw throughout semi-permanent – it was never ‘I did this’ it was always ‘we’. I found it interesting that there was such a focus on collaboration and yet our major project is a solo affair for most people.

Mostly what I got from listening to elastic was a very strong desire to bring some animation into my project – I have always enjoyed doing animation work and would like to explore it further if there would be room to do so

Panel Discussion between the talkers from day 1

Again a big focus on ethics and thinking towards a sustainable future –
LIFE CENTERED DESIGN – Designing with life at the center, not people.

Semi-Permanent reflection

Overall it was a very rewarding and eye opening experience. What I got from the conference overall was the idea of really considering making good, ethical decisions in my work. Even for my project I need to think about the implications that could come from it – who benefits, and who looses from this new way of illustrating? What is the future that I am envisioning, could it help the future of humanity?

The second day of Semi-Permanent was highly inspirational and I found that I was able to relate and draw something from pretty much every speaker, regardless of their discipline and focus.

Some relevant notes

Kyle Mcdonald – Artist working with code

Kyle talked about his work using code to create art and the relationship between man and machine – the good and the bad.

• Computers imitating human – just distilling creativity?
• Machine learning allows a different perspective – like talking to an alien
• Algorithms have a bias, they learn from human mistakes

Shantelle Martin – Quite relevant so I will dedicate a seperate post for her.

Eric Brandt – Graphic designer and professor

• Eric talked about working within limitations and how that can be beneficial. He spoke about mastering limitations and how that is a great test of an artist.