Mr Robinson

Here to blog about my time In Creative Technologies at AUT.

[b-r-robinson]

Studio 5 - Coding the actual sea level to the potentiometer

I had 2 ways to go about showing how high sea levels get. The first way was to get a reading of how high the sea level will rise in 2150, and then just have the sea level rise on a constant average from the year 2000 to the year 2150.

The second way was to get readings from every year and actually show how high the sea level would rise over the course of the years up to 2150, and show how it would rapidly increase over time.

I wanted to do the second way because it was more realistic and showed the users that sea levels don’t just rise at a constant rate, that it is rapidly increasing as time goes on.

To do this, I needed a graph or something to go by to get readings for every year. I searched around and eventually found a report from Ministry for the Environment from the NZ Government (report here).

I found a graph that showed different variations of of sea levels rise over the next 200 years. The different variations showed the results from different tests, based on how much we reduce the effects of global warming. For my installation, I decided to go with the worst of the worst, so this would get a better reaction.

I took the graph, put it into Illustrator, and drew horizontal and vertical lines over it to get accurate readings for every 5 years. I then just filled in the middle numbers with logical guesses.

I thought that I would be able to get a formula that would create the line, and then be able to just add that to my program which would make the sea level rise appropriately, however, trying to get the formula by using Excel didn’t work. So I had to resort to just repeating the code over and over again, and changing the numbers to represent each year and what I wanted the sea level to do.

For example using the picture above:

howHigh represents the current year that the user has input from the clock, 1 being 2001, 2 being 2002 and so on.

HOW represents how high I want the video file, or the sea level, to rise.

The same code is repeated over and over for every year, until it reaches 150 (2150) at which the video file is at the top.

This accurately showed the rise of sea level, when projected onto a screen with the height at 1.8 meters. 

As stated in previous blogs, i’m no genius at coding and haven’t done it in years, so the fact that I was able to do this (and first time might I add) was quite surprising to me and was quite the achievement. It might be a very butchered way of doing it, but it works.

[b-r-robinson]

Studio 5 - Setup

So setup was a complete disaster, but it was kind of expected. Without being able to test my installation beforehand due to the size of it, I didn’t know what I was getting myself into, and was only presuming my sizes would be correct from simple maths.

I went into studio and started asking if it was possible to have my setup outside by the stairs, but was told that we weren’t aloud to during the mid semester showcase. However Charles asked me what I needed and once I told him I needed about a 7m x 7m size, he showed my a corner of the year ones studio that I could have. We worked out a plan, unfortunately having to get rid of my rear projection, that would work with 2 screens set up on a right angle.

However, once I neared final setup the following day (one day before open studio), Charles told me he didn’t realise I would need to projectors set up how they needed to be, and told me I couldn’t do that because they posed a tripping hazard. At this point I realised I needed to go down to just one screen, and wanted to just tear my installation down and be done with it. I felt like the whole immersive feel that user was meant to feel by being surrounded by rising water was just not there anymore and the whole point to my project was just gone.

However, after an hour of moping around, I got off my arse and just dealt with it. One advantage of going down to one screen was that I could get my rear projection back. I quickly set up and changed the video format to fit just one screen. My measurements that I thought I needed proved correct, and everything set up without a hitch after that.

I decided to create a measurement ruler on the side of the projection just so people could see how high it actually was, even though the point of the project was for people to measure it against themselves.

[b-r-robinson]

Studio 5 - Creating the Clock

To create the clock that people would be interacting with, I bought a cheap clock from the warehouse where I could take the glass front off and I could access the clocks hands and face.

I removed the mechanisms of the clock and put a potentiometer through the hole, securing it in place with just hot glue. I then created a clock face on Illustrator, that took into consideration the angle the potentiometer turned, and accurately putting every tenth degree in the correct place.

I had a couple of designs, and the numbers changed from going from the year 2000 - 2100, to the year 2000 - 2150.

Changing the colours from black to grey made it the outer edges more softer (in my opinion). I firstly made the numbers in a square shape to avoid have to angle the numbers, but eventually decided this just made it look worse, and went with the final design. I added the words “twist me” to the top so people would realise they could turn the handle.

[b-r-robinson]

Studio 5 - Posters

I created some posters to go along with my installation. I wanted people to see what was contributing to sea levels rise and what the effects would be to New Zealand in the future.

I wanted the posters to be simple so they were easily interpreted and anyone of all ages could understand it.

I think they came out well, some of the illustrations could have been a little better but overall I think they came out as I wanted them to, and are easily understood.

[b-r-robinson]

Studio 5 - Water Effect

After my first attempt at creating water, I attempted to follow my next tutorial (Tutorial here).

This one was meant to be more realistic by using plug-ins that I could get for free for a trial. I started following the tutorial and it started off well, but at one point he curves his water around, something that I couldn’t do because I needed it to fill the entire page. I think this made the water look unrealistic and I wasn’t happy with the outcome that I was getting, the water colour seemed off and the quality of it just wasn’t there, so I decided to try look for a third tutorial.

While looking on YouTube, I was searching after effects tutorials, presuming that would be the program best suited for the job, but I decided to try other methods. This came up good, as I saw a water effect tutorial using Blender (tutorial here).

I downloaded blender (a program i’ve used before but dislike) and started following the tutorial. It went well with the result turning out the same, but in this tutorial, the water came out in a small pipe. I needed the water to be like a wall, so I changed the shape that the water was coming out of. This worked in a way but when rendering, the water would clump together kind off at the bottom of the screen which left gaps within the water, something I didn’t like.

After some time fiddling around with it, I couldn’t get it how I wanted it, and started thinking of other ways to get my water animation. I thought about using footage of a waterfall, and started searching YouTube for footage, when I came across a video of water running down a glass pane. This looked exactly like what I was trying to achieve, so googled how to create a glass pane water effect. Eventually I came across another tutorial that showed you how to do this (tutorial here).

In this tutorial, he used multiple points where the water came out, not just one. So I followed his ways and created something similar, and it worked, getting the effect I needed. I added a basic shape that moved like a wave on the top, and changed the colour to a bluish colour and, after several renders and a few things changing in between, I got the output that I was happy with.

I uploaded the video to Premiere Pro and quickly learnt how to create a seamless looped video as the same video would be playing for 2-3 hours and I didn’t want to see any obvious restart of the video. I also slowed the video down by 30% as I thought it was a little fast. I wish I could have slowed it down more but unfortunately I couldn’t without effecting the quality of the video.

[b-r-robinson]

Studio 5 - Set up problems

After talking to Laurant the other day about my setup, it was decided to use less space, we would set up the projectors in the studio space, but then have my screen in the hallway by the stairs, which is separated by glass. This was fine by me, there were some vinyl dots stuck on the glass but it wouldn’t impact my projection too much. However, after thinking about it at home, I realised that my Arduino stuff (the clock) had to be on one side of the glass, and the projectors on the other side, which meant the wiring for this would be way too big and not possible. So I had to go into studio and negotiate some space inside the studio space.

However, I am still unsure about how much space I can get. I think I am going to have to scale back my plan. I’ve looked at some information on the internet about the projectors I will be using, and at what distance they need to be placed to get the height I need. I am looking at a setup (I think) of a 4m x 4m space, something I just don’t think is going to be acceptable, which stinks. This would allow me to have the sea level rise up to 1.8m, which would tower over most people, giving a huge impact to the viewers.

However I think I will have to scale this back, which means I will have to cut the number of years down. The information I have found from NIWA, says the worst possible prediction is by 2150 sea levels will rise by 1.8m, so I may just have to go to my earlier idea of going to just 2100 which goes to about 0.9m high, which will cut my setup space down considerably.

[b-r-robinson]

Studio 5 - Getting the Potentiometer to do what I want

The next step was to find a way to make the arduino do what I needed it to do. I searched up and learnt how to load and play a video in processing, and quickly merged it into the code for the potentiometer, replacing the coloured screen for the video file.

With my little knowledge in programming, I was able to quickly get the video file loaded and then have the potentiometer move that video file up and down on the screen, but at first, only at a slight amount, because the Arduino code was maping the values of the potentiometer input from 0 - 1023, to 0 - 255. I also had the problem where obviously the video file loaded in the center of the screen, whereas I needed it to start with the file being off screen on the bottom, so when you twisted the potentiometer, it would rise up.

It’s hard to tell in my above demonstration that the video is rising up and down given the nature of the video footage, but if you focus you can make it out.

I changed the video file to make it easier to see, but then encountered another problem. In the previous video, the footage had black bars on the top and bottom, but in the new footage, it didn’t. This meant that the background of the screen would glitch out when the video went down. 

There are two ways I can think of that would fix this: Put a very thin black edge on the video file, or maybe there is some code that would refresh the background every millisecond to keep it from glitching. I will probably go with the first option when it comes to doing it, just for the ease of execution.

To make the video start at the bottom of the screen, I just added code that added the height of the video file:

image(mov, 0, -howHigh, 1080, 720); to image(mov, 0, (-howHigh + 720), 1080, 720);

Image (load screen), mov (load my movie), 0 (movie plays at x = 0 on the screen), -howHigh (movie plays at y = potentiometer value), 1080, 720 (video aspect ratio).

I then needed to change the value that the Arduino code was mapping the potentiometer to. I changed this to 0 - 100, which is the difference in years that im basing my sea level rise from (2000 to 2100). This actually took me hours to do, pulling my hair out trying to figure out why it wasn’t working, only to realise that I was just saving the file rather than actually uploading it to the Arduino board... really am a rookie at this kind of stuff now.

I was also able to figure out which code was for the LED, which was removed from both the processing and Arduino code.

[b-r-robinson]

Studio 5 - Potentiometer Prototyping

So, i’d decided that I was going to go with a clock, with the clock hand attached to the potentiometer, but I still hadn’t actually tested a potentiometer, how to code it, and how to send the value from Arduino to Processing, so we could do something with it.

I looked online at how to set up a potentiometer with processing, and came across this article called ‘How to Make Arduino and Processing IDE Communicate’, that gave an example of changing the colours of a screen with the potentiometer. It also had coding that allowed the user to click the mouse once to turn an LED on, and then click again to turn it off. I quickly copied the code and the Arduino wiring, and quickly got a working prototype of it working:

So that works. I just need to now think about how I can make that relate to what I need it to do, making my sea level rise and decrease, and also get rid of the coding for the light bulb. That part isn’t urgent, if I can’t figure out which parts of the code I need to delete then I will just leave it in, the LED will be hidden from view anyways.

[b-r-robinson]

Studio 5 - Construction

I am at a point in my project where I need to focus on the construction of my physical piece where the projection will shine on. When I first started thinking and planning this, I thought the frame would be quite easy to do, just create like a cubed frame with a wall missing on one side for the curved fabric wall to sit inside:

However, since I’ve decided that I will be doing back projection, I have to think about what the back of the frame will look like so it doesn’t cast any shadows. In my previous design, in the back view, I had a supporting beam which would cast a shadow onto the wall. Getting rid of this beam may make the structure not great, so I have to take that into consideration.

I knew to make the curved wall, I was going to be using some flexible piping and bending it to make a curve, and wrapping the sheet around it. I thought about clamping the two ends of the piping to the movable T-walls we have in the studio space, and pushing the walls together to create enough force to create the curve. I quickly doodled some ways that the walls could be positioned to maximise the least amount of the walls I would take up, as these walls are quite highly sought after.

A lot of the ways would have used most of the walls up, but there was one way that looked hopeful.

The two T-walls would be positions on an angle, so the over spray of the projection light, which I need to fill the sheet due to the effect of the curved wall, will shine onto the back of the walls. That leaves the outside long while still usable by other students.

After talking to Laurant, it seems that it is too much to ask. He suggested I used some tripods that we have the are used for holding speakers, and clamp the piping to them. This would work and use up less real estate, as long as the tripods can withstand the pushing of the piping when its bent into shape. I would have to use sandbags to hold them down.

[b-r-robinson]

Studio 5 - Prototyping with a Projector

I wanted to test theory of using just a white sheet to do back projection. I hired a projector from uni (should of done this ages ago), and hooked it up to my computer at home. I then got a white flat bed sheet and put it in front of the projector, and sure enough the picture shone through the sheet onto the other side. Here is a picture of the back projection test, projector is in the background with the light shining through quite well.

I did do some research into other materials I could use, and I came across someone saying that frosted shower curtains are good to use, however this was just one person saying this and I decided to opt for what I could test without having to buy more materials.

I then created a model of what my physical screen would look like using paper, so I could see how the projector reacts to it. This is basically a re-test of the prototype I did earlier with a torch and cardboard box, but this time I created an accurate model of the curve, and the light would be shining on the opposite side this time due to changing to back projection.

This is my setup. The projector sitting level with the model sitting in front of it.

I shone a picture of a horizontal line in the middle, to see what happened to the line as it bent round the curve. (Colours are because of the camera lens reacting to the light projection). The line didn’t do anything, but in hindsight, thats because it’s in the middle. If I put the lines further up or down, then they would start to bend.

As you can see in this picture, the effect we had in the previous test with the torch where it tapers to the edges is the same in here, but opposite, so it gets bigger because its further away from the sides.

I could try making the water effect at this point with some degree of being able to predict how it changes around the edges with this prototype, however I believe it would be a waste of time, and I may as well wait until I get everything set up to get everything perfect.

[b-r-robinson]

Studio 5 - Talking Creates Confusion

While at lunch, I was talking to a fellow classmate about my project. While doing this, I had some thoughts about my project. My project will use light projection to shine the water onto a wall around people. However because a person will stand between the wall and the projector, there will be a shadow behind them, blocking the effect.

At first I thought this wasn’t that much of a problem because they still had the water shining on the sides, but if I wanted it to be good, I really didn’t want the shadow. I thought not using the projector at all, and using a screen instead. The TV I have measures big enough on its side to get the height of the water (900cm), but obviously this is flat and I wouldn’t be able to make it go around the person. I decided to talk to Laurant (my lecturer) about it. He suggested I could use both projection and the screen, the screen would cancel out the shadow being created and the projection would just go round the sides. However he then suggested back projection. I didn’t have a clue what this was, so Laurant explained and showed me a ‘video’.

Basically you put the projector behind the wall rather than in front, using a material that lets the light shine through onto the other side. This would eliminate any shadows. Laurant suggested I talk to the mechatronics lab technician we have to see if he knew what material would work best.

After talking with him (Sorry, forgot his name!), he suggested that I just start with a white sheet, as it should let the light shine through onto the other side while stopping the light going any further. I also asked him about attaching a clock hand to a potentiometer, which he said the longer the handle, the thicker the potentiometer has to be, to stop it from snapping, something I wouldn’t have thought of.

Time to start more prototyping.

[b-r-robinson]

Studio 5 - Water

I decided to try start making the water effects that would be projected onto the screen. I went onto YouTube and tried to find tutorials making water effects using Adobe After Effects. Surprisingly, there wasn’t many to follow.

I found one that I thought would do the job (click here for tutorial) and started following it.  After getting a bit of the way through, I realised that this tutorial was using the same method that I used to create the lava in last semesters project.

After realising this, I started to think that this tutorial wasn’t actually that good, and the effect it gave was quite substandard. So I decided to stop with this tutorial, but here is a rendered version of what I did. The effect is there but the colours are messed up as I stopped before finishing.

After this, I went back to YouTube to try find another tutorial. Eventually I found one that looked really good, however I needed to download some plugins to use. After looking at the website to get the plugins, I realised that I had to purchase them for an absurd amount, but I could get a free trial for them. I searched but couldn’t find how long the trials lasted, so decided not to get them yet because i don’t want them to run out before I get the project done. I still am not in a position to actually create anything for the projection, until I get the physical wall up so I can make the water as I am shining it on the projector, so I can get the curve correct.

This is the tutorial i will use to create the water for my project (click here).

[b-r-robinson]

Studio 5 - Projection Prototype

I decided to make a very basic prototype of a projection to try grasp how the light will change when going round a curved surface.

This picture shows my setup. I just shone a torch through a cardboard bow that was open on both ends, so I got the box shape that a projector shows. I would then hold a piece of paper in front of the light that represents the sheet I would project on. Camera on the right on a timer to take pictures.

This picture is showing the light hitting a flat piece of paper.

And then this picture shows me bending the piece of paper. As you can see, the light tapers towards the ends, because the light has less distance to travel, just like if I was to move the flat piece of paper backwards, the box would get bigger because its further away.

I have to take this into account with my project. As I see it, there are two ways to get around this. The first way is to use a program like Touchdesigner to map the picture around it. The second way, which in my opinion is easier at this point in time for me, when I create the GIFS of water, create them so the are curved, so when they are shone onto the curved surface, they straighten out.

[b-r-robinson]

Studio 5 - Prototyping Buttons

Finally my Arduino Leonardo arrived. I can now start prototyping how the Arduino board can control my keyboard input.

My idea was that I would have a potentiometer that would act as the left and right keys on a keyboard. Twisting it one way would activate left button, twisting it the other way would activate the right button. I would then have 100 GIFS that would show a wall of moving water, which would progressively get higher as the years go up (moving through the GIFS).

I wanted to start prototyping simple, so instead of going straight to a potentiometer, I decided to prototype 2 buttons replacing my left and right keys on the keyboard.  I searched the internet for a tutorial on how to do make an Arduino board replace keys on a keyboard, and came across this article:  Make an Arduino Leonardo Arcade controller for PC

Now this tutorial uses an analog stick, so I had to think about how to wire the buttons instead of the stick, but this was pretty simple and quite close to the stick. I then deleted parts of the code I didn’t need, for example the firing button.

It went pretty straight forward and everything worked fine. Here is an example of me using it to flick through random images on my computer. The left button cycles the pictures backwards, and the right button cycles the pictures forwards.

[b-r-robinson]

Studio 5 - Some notes

Below are some notes that I took while researching earlier in the semester on sea levels rising that will be of help when it comes to making information placards for people to read at my installation.

[b-r-robinson]

Studio 5 - Advice from Laurant

I had an idea for the interactive part to be something like a tap handle, where people twist it to turn the years up and effectively making the sea level rise (you turning the tap handle makes water come out - makes sea level rise, see what I did there ;) ). This then sparked a couple of ideas as to making the interactive part more gimmicky. Beforehand I thought of just some basic slider or twisty knob, this in my opinion, made it more memorable/fun to use.

However after talking to Laurant, he said that it could complicate things because if I made it something like a tap, people just wouldn’t turn it, because they can choose to do that, and then the sea levels effectively stop rising. Anything where the audience have a choice, they will take the option where the bad stops happening.

This made me think more about it, and really, the only thing that humans can’t stop is time. So I now have two options if I choose to go on the gimmicky route, a clock that has the years as the hours:

or have a physical calendar where each page represents the years.

The first option would be relatively easy, just replace the clocks mechanism with the Arduino board, the hand is attached to a potentiometer, and there you go. The second option would be harder to make and I’m not sure would work with the audience too well. Having someone having to turn pages would be slow and repetitive, and I would have to make it so there weren’t a lot of pages to turn, so each page would have to represent lets say 10 years, which wouldn’t give the effect I want.

[b-r-robinson]

Studio 5 - Arduino

I decided that which ever way I do the interaction part, whether its a knob, a slider etc, it would have to be done with a Arduino. Luckily I still have mine from the first year of uni, but I hadn’t used it since, over 2 years ago. I brought it out and started playing around with it again, doing some basic projects from the book that is included, just to familiarise myself with how it works, the coding and building the parts together. 

I then started to think about how I could make it connect with my laptop so I could control the laptop with the Arduino. After some research, I found out that I couldn’t do it with an ordinary Arduino Uno, because it lacked the tech to do it, unless I wanted to flash my laptop, something I wasn’t keen to do.

To do what I wanted/needed, I had to get a specific Arduino board, the Leonardo or the Micro, both of which I couldn’t buy from New Zealand *sigh*. Even buying one from overseas was a challenge. Many boards on sale were replicas, from different brands that claimed to be the exact same, but on forums that I searched, people had informed that they sometimes weren’t what they say. 

I didn’t want to risk buying another brand, and finally found an Arduino branded board, but it was going to take a month to get here *another sigh*.