Planning, Production & Practical

Arcade Primary Research

As I have time off over Easter break and am planning for my scene to be somewhat realistic, I decided to travel to an actual arcade and take reference pictures of the layout, walk space, CCTV cameras, Lighting, and general placement of the machines. I couldn't find any real-life derelict arcades that I could visit, so I had to settle for the next best thing, a present-day arcade. However, I have to keep in mind that the layout of arcades throughout the years hasn't all stayed the same. I will be taking inspiration from the arcade that I visited, however, I won't be simply basing my scene on the arcade that I visited.

In visiting this arcade, I realised that there is a lot more to making an arcade than I initially thought, as I didn't even consider any CCTV cameras, lighting, or walk space in my initial draft. Walk space and optimal pathing is an important aspects that I missed out on, as without these, some customers may feel cramped and may not know where to go or get to all of the machines which wouldn't be profitable for the arcade owners. There are going to be quite a lot of people walking through the isles which means the walk spaces should be big enough to accommodate these needs.

The optimal placement of change machines is imperative to an arcade's success. I went to two arcades over Easter break and the first arcade had small change machines which were scattered all over the place, allowing you to spend more money easier. The second arcade I went to had around 3-4 huge change machines, which you sometimes had to queue up for, and as there are only a few, you would have to leave your section to spend more money. Out of the two types of change machines I've seen, I'd say it's fairly obvious which one I'm choosing, the smaller change machines are what I'm going for. They'd actually be easier to model as well, as shown below they are quite a simple shape. I could, for the most part, grab a rectangle and slap a texture on it and call it a day.

Pictures

Gantt Chart

I made this Gantt chart to hopefully keep me on track To show how I can work against timescales. There's a well-known quote which I'm sure someone's told you at least once in your life which is, "failing to plan is planning to fail". Whether it be for a presentation, meeting a deadline, or just driving to work, planning is always beneficial. As long as I stick to the schedule, I should be fine.

Fallback Plan

Failing to finish a task in time is something nobody wants to think about, however, Creating a fallback plan can be a life saver if you do end up needing it. If your initial idea for one reason or another ends up not working out then you likely won't have any time to waste, you'd rather just crack on with your project, instead of thinking up an idea on the fly. If I do end up running out of time trying to make the scene, I could convert to an asset pack quite easily as I would have all of the assets completed, I could simply place them all next to each other and take a picture of them. I could go as far as to get a 3D turntable animation but as it is my backup plan, I probably won't have enough time to add anything unnecessary.

Asset list

- Pacman arcade machine

- Space Invaders arcade machine

- Asteroids arcade machine

- Change machine

- Coin pusher machine

- Claw machine

- Chair

- Pillar

-CCTV camera

-Lighting

-LED lights

Reference Boards

CCTV Cameras

Arcade Machines

Lighting

The main image that inspired this type of lighting was the bottom left picture. Seeing as my scene depends on lighting as the main focus, I want to make sure that I choose the right brightness, hue, and positioning as each option can drastically change the mood of the scene. For example, the bottom left picture uses more of an orange tint for the light colour, creating a cosy atmosphere, even though the scene is abandoned. Comparing this to the other pictures, they look a lot colder and in some cases sinister. I want to push the idea that my scene is abandoned and most abandoned places aren't the cosiest or nicest places to be, to do this my main reference will be the bottom middle picture. Having the light rays coming through the window gives off the idea that the room is dusty and that's exactly what my arcade scene would look like in real life.

2D Floor Plan

The software I used to create my 2D Floor Plan is called Dungeon Scrawl, which is a mapping tool primarily used for games like Dungeons & Dragons (D&D). D&D is a tabletop role-playing game in which "the goal is the exploration of the fictional world and the telling of the story", a game like that needs a world with endless possibilities and that's what Dungeon Scrawl offers. In my case, I'm using Dungeon Scrawl to make a world with a narrative behind it, but not to the scale of a D&D mapper. The map shown below is my first attempt and as I haven't currently made all of the assets the proportions or placement may be subject to change. If in the future it does change I'll show the updated version below this one.

Arcade Machine

Change Machine

Penny Slot Machine

Window

Pillar

Poster/Advertisement

After placing all of the assets into my world, I realised that the proportions were all off and there wasn't enough room for people to walk around in between the machines. To fix this I initially just moves the two rows of machines in the middle to the right by a few blocks, however, now there wasn't enough room on the right side, so I ultimately made the room bigger, and this time around I didn't restrict myself to the grid, in doing this, I could be more accurate with my placement of the machines

The size of the room in my first attempt was 15 x 19, and in the one below, it is 16 x 19. This by itself is barely anything to mention, however, you can tell that there is more room in the lower one. This is because I decided against making the room any bigger, in turn, I changed the size of the machines to a more reasonable size.

I quite like the idea of having light rays shooting through the windows to better show how dusty the arcade is (shown in the above picture).

As this scene is derelict, there won't be any working lights inside the arcade, so it needs to be bright enough outside to light most of the arcade up.

You'll notice that there are only two of the four walls present in the second version, this is because I want my scene to be more of a diorama, as opposed to an actual room. As of right now, I haven't tested whether only having two of the four walls would still work with the lighting that I'm trying to go for.

However, this may not work, I want natural light from outside to light up my whole scene, and to do that the lighting needs to be very bright to give more of a realistic effect. If it doesn't work out I can always use all four walls but I would prefer to have two.

Arcade Machine

Change Machine

Penny Slot Machine

Window

Claw Machine

Pillar

Investigating final product presentation techniques

Truck Shipment Diorama

Western Diorama

After finishing a model or scene, you want to be able to show it off in all its glory, and one of the best ways to do that is by having a high-resolution screenshot of your final piece as your front image that everyone sees first. First impressions are important and if you don't put the right picture as your first image, some people may click off before seeing the rest of your work, or not even click on it in the first place. The two buttons shown below are great examples of high-resolution screenshots that catch your eye.

Arcade Machine Diorama

Lighthouse Diorama

Another presentation technique is using a 3D Turntable Animation in which you can move around in to see the whole of the asset or scene yourself. The two buttons below will bring you to two ArtStation pages that use a 3D Turntable as their first image. Out of the two layouts, I will be using the arcade machine format as It kicks things off with a complete view of the asset, and below it shows things like the wireframe, all assets together, process/workflow, etc.

The lighthouse diorama kicks things off the same way, using a 3D Turntable, however, the rest of the work isn't shown as there are only two more pictures below it. Compared to the arcade diorama which has four pictures. Two pictures aren't that much more, however, one of the pictures shows the wireframe which is quite helpful in understanding how the model was made.

I think that both of these ArtStation pages are lacking in one thing. Progress pictures. Adding this helps the reader know how you did certain things. Sharing your knowledge on something that you're good at is a great way to help those around you grow as a community. I know that the PAC-MAN arcade diorama has helped me discover an easier and quicker way of making an arcade.

Making the arcade machines

I used the same format for the other two arcade machines so I'll show the final product, but I won't be showing the process. The reason why I'm showing this arcade machine over the others is that it was the most fun to make, it was the last of the three to be made so I could carry over my experience from the first two and overall it went smoothly. It's also my favourite looking out of the other arcade machines, something about the shape feels so nostalgic, even though I didn't grow up with arcades.

The link to the website that I got this template from is shown above. It's a super useful site that has helped me make these arcade machines as close to the original as I could. The way that I got the template onto Maya is by going to View, Image Plane, and Import Image. I've used this tool on previous projects and it has saved a lot of time. It's especially useful if you're making a Hard Surface or inorganic model like a car for example, as you want to be as accurate as possible.

To make the inside of the arcade I used these images below as a reference. I got these images from the same website as the template above. The arcades below are both fairly similar, but I decided to take most of the inspiration from the black one. As shown below the points highlighted are the main aspects that I copied over onto my design. Comparing #1 on the black one to the white one you can see the difference, as there is no downside to choosing either one, I chose the aspects that are easiest to make.

3D Floor Plan

1

2

3

1

2

3

After finishing the low poly models of my three arcades, I decided to use my previous floor plan and convert it into Maya to get a better look at the proportions and spacing. I'm glad I did this as there was clearly not enough room in between the arcades. I'd say there's just enough room for people to stand at the arcades, however, there isn't any room for people to walk past them.

Comparison

Deciding to make the walls and floor modular made it much easier to extend the room as instead of having to stretch the walls I could simply duplicate a wall piece and add it to the end. Stretching the wall would turn the square sections into rectangles, as shown below which doesn't change anything major, however, keeping all of the sections as squares is easier for me to edit the size and shape of the room and easier for the reader to understand.

One benefit to making the walls in pieces is that I can create a window in one section of the wall without messing with the rest of the wall.

As you can see to the side, all I needed to do to make the window was remove four squares from this wall piece, in doing this, I don't have to worry about fixing n-gons that appeared as a result of cutting a hole in a plane. This window is simply an example, the real window and how I made it is shown below.

Modular Window

Arcade diorama mockup

When I imported the claw machine into this scene, it was way too big and the reason for this is because instead of using a scale guy for reference, I scaled it to my reference pictures' measurements which ended up being too large for the rest of the scene. To make sure the proportions are still correct, I shrunk the claw machine down so that the control panel on the claw machine to the same level as the control panel on the rest of the machines.

After some while, I came up with this layout. I'd say this format is more realistic than my first iteration as there is more variety, but it is also grouped better. Before I had the arcade machines thrown in without any thought behind it, they were there simply to fill space, however, I'd say now it looks like more of an actual arcade. The placement of the windows was an ongoing struggle as I wanted the scene to be fully lit by the sun instead of having inside lights. This was because if this were a real derelict arcade, the lights probably wouldn't be working.

Making the window

The base shape of this window is a cube, after I stretched it to the size of my reference picture, I added three edge loops to mimic the structure of the window. These initial edge loops are placeholders which I'll remove later on, the next step was to add even more edge loops, placing them on either side of the placeholder edges which will later on act as the small pillars in between the glass. Lastly, I selected all of the side faces and extruded them to match the size of my reference.

In the reference photo that I'm following, the measurements are from 12" to 24" and from 24" to 84". For simplicity's sake, I've chosen the lowest number from both lengths. To give the effect shown in the reference, I started by adding an edge loop around the outer face to give some space between the glass and the outer support. I then selected all of the outer faces and extended them by 1 using the extrude tool.

In my eyes, everything here looks a little bit confusing and cluttered. And after puzzling on it for a while, I realised that some lines aren't needed. For example in between the edges of the window frame. After removing these inside edges the whole asset looks a lot nicer and easier to read.

Making things easier to read for people without any modelling knowledge has been something I've been working on for a while. To do this I've been showing my thought process and screenshot examples alongside my writing so that you can easier visualise what I'm saying.

When you remove a face from an asset in pretty much any modelling software, you will most likely be able to see inside the asset, having these holes open is a big problem as it will cause light bleeding to occur. In this case, light bleeding would cause light to travel through the asset in places where it shouldn't. This is especially important in my case as I'm making a window in which light will come through.

To cover up these holes, you need to use edge mode and select both sides of the gap, right-click and select 'bridge'. After doing this step once, instead of clicking on the 'bridge' tool, you can use the hotkey "g" to repeat the last command.

Looking back at my reference picture, I've only made one side of the window, so all I have to do now is duplicate the original window and place it next to it.

After finishing the window I ended up adding a plane with 50% opacity in between the window frame to act as glass.

Making the claw machine

The claw

Above is a first iteration of the claw which ended up being a failure. I spent around 15 minutes on this above claw and for that amount of time, it's barely any work done. I was struggling to get the claw portion made, and the main reason for this was that I was trying to make all three fingers of the claw within one shape but to no avail. This experience wasn't all loss, however, as it made me relearn something so simple, yet so effective, which was that I shouldn't be restricted to using one shape, I have unlimited shapes and I should be using that to my advantage.

To put this newfound experience into practice I started again but kept in mind that I could use multiple shapes instead of struggling with one shape. For the second iteration, I used primitive shapes without extruding or edge loops, all I did was resize and rotate. After I completed one of the three claws I then duplicated it and rotated it around the crane's centre by 120° to create the other two claws. The body of the claw is also made out of primitive shapes, being made entirely out of cylinders. To get the look of a bolt on the bottom cylinder, I removed every other edge around the side. In the reference picture, the bottom cylinder is an octagon and mine is a decagon, but as you can tell from the rest of the claw, I'm not going into too much detail as it is a side prop.

UV's

I Initially thought this process was going to be the hardest and without realising it, I was continuously avoiding it. On the contrary, it was arguably the easiest step, because of Automatic UV (referring to AUV for ease). AUV is exactly what it says on the tin, it automatically UVs your asset for you. AUV is pretty inconsistent as it can't cater to every asset. However, In my case, it worked well enough to give me a base to work with. The only reason that I can think of for why AUV worked in my situation is that all of my assets were clean, simple, and overall didn't have any problems with them. As I was using very primitive shapes the algorithm behind the AUV could handle my assets.

To tell if your UV mapping is correct you make sure all of the squares are 1, the same shape, 2, the same size, and 3, if you want to be picky you can line them up to carry on the pattern across your whole asset.

As you can see from where I've drawn arrows in the left picture, it is stretched and as previously stated, you can tell because the squares aren't squares. To fix this I need to select all of the edges around the area that I want to change, then go into the UV Editor and use the "Cut" tool to make it separate from the rest of the asset, the last step is to stretch the face until the rectangles turn into squares.

Colour palette for the machines

To make sure that the colours I've chosen are the right fit, I took a screenshot of my model in Maya, and brought it into photoshop to create a mock-up coloured version to get a preview before I go into Unreal. The colours that I've chosen aren't at random, they're from an actual arcade that someone made from scratch and the link to that is shown below. On the homemade claw machine, there were patterns on the sides, however, I decided to just take the base colours and use them instead.

The colours may look a bit diluted, but that's because, in Photoshop, I painted over the claw machine and if I left the opacity of the colours at 100%, then you wouldn't be able to see the claw machine underneath it. After a lot of trial and error, I ended up going with 70% opacity as it is the best mix of seeing the colours in as much opacity as possible while being able to make out the shape of the claw machine.

Homemeade Claw Machine

With two weeks left I have the ability now to stop modelling and fall back on my safety plan of making an asset pack, however, this is what my final grade is based on and what will represent all of the efforts that I've put in over the last two years and I don't want to be lazy now, it's not worth it. Half term is just around the corner anyway so I'm not worried about burnout.

Survey Results

The main consensus is that the arcade machines need textures to them, and I completely agree. Something felt off when I was looking at my scene but I couldn't put my finger on it. But after doing this form I realised that adding textures was exactly what I needed to take my scene to the next level.

I've said previously that I wanted to learn Substance Painter and this is my chance to try it out. If I end up not having enough time to texture all of my assets then I will leave it as it is, however, I'd at least like to try and add texture.

The majority voted for the scene to stay the same, however, there was still 40% that voted for the scene to be brighter. I've decided to have a look at the scene with it being brighter and see for myself whether it is better or not.

I would have to agree with the results here, you get the overall feeling that it was meant to be realistic, but, I'd say the fact that the arcade machines and walls aren't fully textured breaks the realism. Nobody responded with "Don't think so" or "Definitely not" which I'm happy with as it means that it at least somewhat represents a realistic scene, however, as the majority isn't certain whether or not it's realistic, I still have some work to do to push the idea that it's realistic.

Substance Painter texturing

The main reason for me to bring this arcade machine into substance painter was to add a layer of dust onto the tops of this machine to make it look like it's sat there for a few years. As it's my first time using Substance 3D Painter properly, I wanted to experiment with it and see what I could come up with. After a while of getting used to the controls, and adding some extra detail, I ended up with the product shown below. As it's my first attempt, I'm sure there are some things that I could've done differently, however, I'm quite happy with the results.

After making the first iteration above, I was super happy with it and I thought it was done. I didn't think I could add anything else to make it more realistic. I didn't realise how wrong I was. Comparing the two versions, I can't believe I thought the first version was good enough. The second version is so much more realistic, from a distance I might even mistake it for a real machine. With anything that you make, you can always see some spots that you could improve and my asset is no different. Even though area No.1 labelled below isn't that big of a problem, it still bugs me. The problem is that the resolution of this specific section is a lot lower in comparison to the rest of the asset. It may be small, however, it's still noticeable, at least it's noticeable to me. I do know how to fix this problem, however, I realised it too late. If I were to fix this problem I would have to restart all of my work in Substance 3D Painter, as I would have to take it back into Maya and redo the UVs.

Area No.2 isn't so much of a hindrance as the first problem. The line going through the back of the arcade machine appeared when I was adding the grease marks onto the back. The reason why it's not as much of a problem is that it doesn't look half bad. It might look better with that line there as it breaks up the space and adds character.

<-------- Area No.2

<-------- Area No.1

Issues & Solutions

The first problem is that whenever I zoom out away from my assets, the faces start flickering black. This issue had been bugging me for the longest time but who knew the fix was so easy. All I needed to do was go into the attribute editor and deselect double-sided faces.

The fix was easy, however, finding the solution to the problem was the time-consuming part. I was searching for at least a solid 20 minutes. All I needed to do was go into the attribute editor and deselect double-sided faces.

The second problem that I came across was a beam that was meant to be gold wasn't, I couldn't think of the answer to this problem for quite a while until I saw that there was a material that had been auto named to standardSurface1 as highlighted above, and as I intended to name all of the colours accordingly it means only one thing, I forgot to select that beam when I was selecting all the faces for the gold colour. After realising what the problem was it was an easy fix, however, it still took some time to correct it.

As soon as I saw this window I had a solution in mind, however, I knew it would take a while to execute. To fix it I had to make sure none of the faces are overlapping and it wasn't as easy as simply pulling them apart and putting them next to each other, as if I did that the cross going through the middle would be too wide to look like a proper window. What I had to do was redo most of the window, luckily I had been saving iterations as I went which allowed me to revert to a previous save where I only had a quarter of the window completed.

Working from a previous save was a little demoralising as it meant that I had lost a good amount of time making a window that didn't work. The images below show each quarter of the window that I had to make. To create the look that I was going for, I had to remove certain sides so that they would snugly fit together, almost as if they were puzzle pieces.

As always, it's never as easy as it looks. At first glance, it looks like all you need to do is select the faces on the sides, delete them, and bridge the gap with a new face. That is what you need to do, however, because of the way that I made this window, it was so much harder to complete than I thought it would. I'm glad that I took the time to fix this problem, even though this asset isn't the main focus, and I probably could've gotten away with leaving it. It made me realise all over again how important a clean mesh is, not only for the looks but the practicality of it. As having a clean mesh helps you easier identify and fix problems if they arise.

This next problem was a lot easier to fix than the last. This problem arose as a result of me forgetting to UV map the walls and ceiling, resulting in the surfaces not being able to work properly. The part that took the longest, however, was rebuilding the scene. As I had reimported the assets again, this meant that I had to remove my first iteration and make it again using the fixed assets.

The issue above led me to question all of my other assets, to make sure that I UV mapped them, I checked them all over. Verifying my suspicion, I found the side of my coin pusher machine's textures were all over the place. The fix was the same as the last problem, all I had to do was reimport the asset with the fixed UVs. This one was easier to execute as I had the experience from the last problem, and also there wasn't as much I had to rebuild.

Unreal Engine scene creation

Kemal Günel Youtube

The scene on the left was my first attempt at lighting my scene. I used spotlights at first which worked quite well, however, the lighting outside of my scene was still pitch black. To create the effect of it being daytime outside, I followed a series on YouTube by Kemal Günel on Lighting in Unreal. There are 25 episodes to this series with most of them being 5-10 minutes long, with some being an hour long. Watching all of these episodes would help me understand lighting a lot more, however, I don't have enough time to watch all of them, so, I watched the first 8 episodes which were all 5 minutes or under. Doing this taught me a lot of the basics when starting to create lighting. As of 2022, these videos are 4 years old so the layout of some things has changed, however, it was still helpful.

It may look a lot different to the last version of my scene, however, all I've done is texture the walls, fix the lighting, and start experimenting with some dust particles which I'll talk about later on as I haven't fully finished it yet. The floor texture is a tileable image that I found on Pinterest. I also found the wall textures on Pinterest, however, I couldn't find a brick wall texture with flakey paint, so I made one. By removing the background of the flakey paint in Photoshop, I could then layer it on top of the brick texture. As both textures are tileable, this pattern still fits.

In making the lighting, I realised that I'd need to add the other two walls to have the lighting working as I intended it to. The scene is way too bright for what I was wanting. Adding the other two walls made the scene a lot darker which was closer to my desired effect, however, now the scene was too dark. Making the outside light brighter did fix the issue, but in doing that, the outside light was unrealistically bright. To fix this I had to place a few lights scattered around the scene to brighten it up. Adding these lights was trickier than I thought it would be. After messing with each setting and trying to find which one worked best, I found out that aperture was the setting that I needed to change. Aperture is the range in which the light travels, enlarging this range and pushing it further away from the walls means that I could get more subtle lighting. The reason for adding this lighting was to brighten up my scene, however, it also acts as light scattering as in real life, if there was a window open in a room in the daytime, the whole room would be lit up.

Ben Cloward Youtube

gameDev Outpost Youtube

Creating the light shafts shown below came with its own set of problems. I did find a tutorial for creating light shafts by Ben Cloward on YouTube, however, some of the settings that Ben changed were for one reason or another unchangeable for me. Luckily for me though, this tutorial was part of a series which allowed me to see the process he went through. In doing this, I found out that the aspect I was missing was the volumetric and height fog. Which I then added and tweaked to my liking.

The dust was created by using a Niagra Emitter which I hadn't used before so I followed a tutorial on it by gameDev Outpost on YouTube. The tutorial taught me everything I needed to know and my dust looked just as I'd envisioned.

After getting to this point, I was almost going to leave it as is and submit it, however, the walls were bugging me. I spent a long time creating the walls in Photoshop so I was hesitant to change them out, but I felt it needed to be done. I was cutting it close to the deadline by adding the newly textured walls, however, the result was a big step up from my last attempt and I'm glad that I tried out the new materials for the walls.

Quixel Megascans research

Quixel Megascans is a huge library of thousands of materials and object scans. Quixel has famously been used in the creation of The Jungle Book which utilised Quixels vast array of plants and shrubbery. The creation of Quixel Megascans came about because of the demand for high-quality materials which wasn't on the market. The convenience of this library was the driving force for this software's success, as previously, anyone who wanted a high-quality material had to either make it themselves or find one elsewhere. Quixel promotes the ability to try something new and to test out using materials, without the hassle of creating them yourself.

At the time, I thought I finished the texturing of my walls and floor, however, looking back at it, and comparing it to other scenes on ArtStation, I realised that I desperately needed some proper textures. After searching for the best I remembered seeing Quixel Megascans in a topic I've done before and thought I'd give it a try.

I'm using vertex painting for creating the peeling plaster texture on top of a brick texture for the walls of my scene. Vertex painting is commonly used in combination with Quixel Megascans as they compliment each other quite nicely. This image shown below is from a tutorial I found on YouTube showing the basics of vertex painting. This tutorial was helpful in allowing me to paint, however, it doesn't give the exposed brick look that I'm going for. The transition between the two textures is too smooth as I want the plaster to look like it's flaking off.

Vertex Painting research

Basic Vertex painting tutorial

Advanced Vertex painting tutorial

As the first tutorial that I found didn't give the desired effect, I found another one, and this time it worked perfectly. The biggest difference was that the gradient between the two textures was a lot sharper which was perfect for my project. Another thing that became clear to me while watching this tutorial was that the more vertexes you have on your asset, the easier it is to paint upon. This should've been clear in the name, being called vertex painting, however, I'm glad that he touched upon this topic. To paint using this tool you'll need to hold down shift and click or drag click on vertexes to paint on your texture. This means that without these extra vertexes, you wouldn't be able to paint on them.

Comparing this version of the wall textures, to my initial textures, the difference is incredible. I'm so glad I decided to redo the textures as it brings a whole other level to the realism factor. From playing building games I learnt that when texturing walls, you shouldn't texture walls randomly, a lot of thought goes into it. I initially started texturing these walls by randomly removing plaster throughout, however, something about it didn't look realistic. After remembering what I learned while playing these games, I ended up removing most of the plaster at the bottom and gradually transitioned to plaster at the top. This created a more realistic pattern, as when a piece of plaster starts to flake off, usually, the surrounding plaster also starts to flake, resulting in big patches. I imagined that the most wear and tear would be at the bottom as people can interfere with it easier, either by scuffing up the walls with their feet or by moving things around.

Even with the texture added to the walls, I felt that they were still a tad empty, and having a space in your scene can drastically reduce the quality of it. To fix this issue, I decided to add some posters, and as my scene is based in the present-day, but has been abandoned since 2004, I thought it would fit nicely to add some games that came out in 2004. The three games that I chose to add were Grand Theft Auto: San Andreas, Halo 2, and Doom 3. To work around copyright issues, I scuffed up the posters to a point where they are almost unrecognisable. From seeing some of the letters and looking at the colour of each poster, you can tell what they are, which means you and I can tell what they are, but, it wouldn't count as copyright.