By Yiming Li
This project is divided into two semester sections for completion. Therefore this learning journal will also be broken down into two different Parts.
Part 1
Throughout the project, the first part was more like an attempt to try out a new approach, to broaden my horizons and abilities as well as my understanding of the subject of “Form”. The thinking that I did in the first part of the project was very much the foundation on which the whole work was shaped.
The inspiration for the project came from a visual type of work I saw from some people about flattening 2D by converting 3D models to point particles. This is a very interesting theory, because usually we can tell the dimension of an object by seeing if it has any undulation on the Z-axis, or in other words, if something is flat on the Z-axis, then we usually regard it as “two-dimensional”. This is not the case in the work we’ve just described, because neither the particle effect nor the reduction in the number of faces really changes the dimensionality of the object – even though they appear to. Reducing the number of surfaces can make an object look like it’s being squashed, and particle effects can make an object look like it’s being broken up so that it can only be viewed from a certain angle.
Some of the more interesting visual sources worth documenting are visual artworks about 2D images made with touchdesigner. On the contrary, these works discuss more about the techniques of transforming 2D images into 3D effects, which may be less relevant to my final output, but still provide some very important ideas in the opposite direction for the later part of my thinking.
Back to my own reflections on the project itself, as well as a discussion of the larger themes of the project as a whole. Form or shape to me is more of a physical meaning of the word, such as the state of an object, or its material and its characteristics. However, the laws of physics exist in reality, and for computers it is a string of data. I realised this when I tried to dismantle a cardboard box. I unfolded a cube and turned it into a piece of paper, did this turn it into a flat ‘square’, or just a squashed ‘ cube’? Of course all this behaviour is meaningless to a computer. Because whether it’s a square or a cube is just a string of data to it. What I want to think about is how to substitute this thought process and the feeling of deconstruction and visualise it in a way that can cause the viewer to think about the subject.
After the above process, I have a concrete idea of the whole project in my head. Is it possible to make a programme that can transform the nature of an object? I chose to use Touch Designer as the main production software for the whole project. There are many components to analyse 3D objects and apply different effects to them. First of all I dragged a complete OBJ model into it, and the SOP class of components can completely analyse and transform it from surface to point. During this process the whole nature of the model changes and I need to re-attach a material to it, as well as light it and set up the camera, a process that is a bit like 3D software. Finally, the model is re-rendered with TOP components, and at this point the model is completely transformed from a ‘face’ to a ‘point’ object.
Up to this point, the project still looks like a visual class piece. I wanted to add some interaction to it, as I mentioned the theory of ‘manipulation’ in my research. However, due to time constraints I didn’t come up with a particularly good way to do this, so I chose to first write a trigger called ‘button’ to show some changes. I will continue to expand on this in the second part of the project.
The final output is similar to a small demo, so this is the end of the first part. This part of the project has opened up some thinking and proved the feasibility of some methods for me. However, there are still a lot of issues that need to be solved, such as the choice of the object, how to deepen the presentation of the topic I want to discuss, and how to make the work more interactive. But the demo is still a good one in terms of the visual effects presented. When I press the button, the 3D stereo model looks as if it has been squashed into a particle with no volume.I will continue to work on it in the next part.
Part 2
Back at Design Domain after a few months! In this part, I will continue to follow up on the previous part of the study and continue to explore the theory and practice that shape and form lie in those interesting aspects of physics. In the previous part, I successfully used TouchDesigner to create some visually rich particle effects. This time I will introduce some new methods that I have never used in my previous projects. This time I will introduce some new methods that I have never used in my previous projects, so that I can improve the whole project.
As I said before, I think my choice of the whole object in the last project was lacking in consideration. I think this is important, so I will focus on changing it in this section. Likewise, the part about the interaction I will optimise to make the whole work more complete.
In fact, there are a lot of works similar to my last demo, Touch Designer gives artists many different ways to present their work. Particle effects can be converted to point clouds and point lattices; pictures can be converted to particles; objects can be converted to particles; particles can be re-clustered and dissipated, and so on. I’ve seen many presentations of this work and it’s very rewarding. It included a lot of visual presentation angles that I had never thought of before. One of my favourites was a project that used an airflow sensor in an Arduino in conjunction with TouchDesigner to make the objects on the screen change by blowing on them.
The choice of objects has been under consideration for some time. I discussed thinking about dimensionality in the first half of my research, but the subject was too broad and not conducive to expression, and I wanted to put more emphasis on the physical variations in form and material of the objects. During this process, I realised that candles in the home seemed like a good choice. A candle is an object that has wonderful properties in its form. A brand new candle is usually cylindrical in shape and changes as it burns under the influence of heat. The amount of heat, the height of the burning flame, the temperature of the room, the flow rate of the gases in the room will all unknowingly affect the shape of the whole candle. When you light a candle, you never know what shape it will take until you put it out. In this respect the object is related to what I have said before about human manipulation of form. People have been using candles for thousands of years and have known their properties for a long time, but there is still something uncontrollable about them. I think this object is related to my previous thoughts on ‘form’, so I chose it.
Of the candles in the house I chose one that had a larger feature in its shape, which I’ve always thought was pretty. In fact it was previously just a perfectly normal cylindrical candle bought at Ikea. The next thing to do was how to move this candle into the computer. I chose an approach that I had never used before, but it was interesting, I used an app called Polycam, which is similar to the LIDAR scanning software that comes with Apple, it quickly scans an object and transforms it and produces a one to one model.
In the process of using it I have accumulated some experiences and feelings about this software. It’s not really a ‘radar scanning’ software, in fact it requires you to take about 150 photos around an object and upload them to the cloud, where they are probably analysed by a server or Ai and a model is generated. This leads to some problems with the generated model, you need to manually adjust and polish it, including myself who dragged the model into blender first to remove some unnecessary points and surfaces, and adjusted some errors before subsequent use. But no doubt this is a powerful and portable software, and I will probably continue to use it for future projects.
There’s really not much to say about the Touch Designer part, thanks to my previous work I now just need to replace the new model with the old one I used for the demo before and make some changes to the detail data. The next thing to do is the interaction part of the design.
The approach I saw during the research above gave me some ideas. I don’t really need to use a gas sensor as a trigger device for the interaction. Because this component was actually originally called a liquid sensor, it was a little bit expensive, and it was used to measure the flow rate of a liquid. If I wanted it to measure gas changes I would have to change a lot of data in the Arduino, which would make this part of the workload huge. Also for monitoring gas variations, I thought of some alternatives: Touch Designer has an Audio Device In component, which was originally intended for sound visualisation or audio-visual interaction projects. I might be able to use a long-wire microphone, hidden inside a candle, so that changes in the sound detected by the microphone can be converted directly into data that can be read by Touch Designer through the maths and analyze components.
I wrote a new section of Audio Device In for Touch Designer as a trigger to replace the original virtual button. At the same time, I figured out how to hide the microphone in the candle. I drilled a hole in the middle of the candle to tuck the microphone in while the wire harness can come out from underneath.
After a series of tests, the results seem to be good. When I blew out the wick of a candle – as if I were actually blowing out a candle – the entire on-screen model of the candle dissipated into particles. After a while, the particles come back together to form the candle again. This is in keeping with my theme of artificially destroying, or ‘manipulating’, the form of objects. The act of blowing out the candle does not only destroy the candle, it also destroys the physical properties of the candle.
This is the end of this project. I am satisfied with the output of the project, but there is still a lot of room for improvement. On the day of the exhibition, many people came to experience the project and received a lot of feedback. Unfortunately the microphone in the installation had some problems in the afternoon, it didn’t seem to be sensitive anymore, probably because something was blocking the sound hole, so I had to change the data directly to show the change. But all in all I learnt a lot from this project, applying new approaches and thinking in different perspectives. There is a lot of new knowledge that I should take into the future.
