There are three main elements to this image that I was trying to portray: The ticket gates/entry portal leading down to the Red Line platform on the left, the “forest” of columns that visitors can meander through to access the commuter rail platform, and finally the floating commuter rail that places the train on display via large expanses of glass towards the top of the image.

I wanted this space to be red. It turns out it is somewhat difficult to work with this much red. Light areas kept getting blown out and I was struggling to find the right balance of saturation. Too much red saturation and detail gets lost. To little red saturation and the images quickly feels grey and faded.

I spent some time trying to separate geometry with light and build some depth in the image. Again, because of all of the geometry going on in this view, I wanted to avoid the space feeling too cluttered and confusing. I used the light to help define edges and push elements back and forth in the image in an attempt to help organize and clarify the composition.

Since it has been a while, below is the entire series of images for this project thus far.

First up, the commuter rail platform illustration was an important image in this series because it was the first illustration that really shows the new experience at the platform level. This meant rendering the train as accurately as possible. I had to make a decision early on to either go all 3D or try to handle it in Photoshop. While I was able to find lots of reference photos, nothing was at a high enough resolution or at the correct angle to simply cutout and Photoshop into the image. I also didn’t have time to fully model and detail out the car in 3D. I did, however, have a rough and low quality model of the train that I downloaded off of Sketchup Warehouse that gave a me a base to start with. I ended up taking a hybrid approach by modeling in some details and then stitching in the remaining details via Photoshop. The train cars had an abundance of subtle patina and aged qualities to it that I knew had to be approached through Photoshop. But, I needed reflections and high resolution that could only come from the 3D model.

I researched online for as many images of the train car as I could find so that I not only could build up a library of textures but also get a sense of how light was reflecting and reacting to the metal and painted areas.

For the 3D side of things, I focused my efforts on modeling some of the details such as window frames and roof corrugation as well as adding a slight ripple to the metal texture. Once in Photoshop, I added the doors, wheel systems, mechanical details, and lots of patina to get the cars feeling like they have been used for years.

Originally, I had the top of the image cropped with a focus on the platform. But switching to a vertical orientation and revealing the totality of the station amplified the drama of the image. The focal point is still the platform, with all of the activity of a Celtics game day.

Digital Mesh Display

One of the elements of the station design that I have hinting at is this idea of digital mesh display system turning the facade of the station into a digital billboard. This spread looks at several scales and display typologies to help explain the potential uses of the system.

There will most likely be quite a bit of text that will be added later on, but for now, this gets the idea out of my head. I am also realizing that I will probably need to create a perspective rendering of this system at some point too.

I was able to put this image together in a day, but so much of my time went into the details. Because there are so many elements to this image, I needed full control over the composition and positioning of the camera and light. This meant a lot had to be modeled in 3D. Luckily I could recycle elements from my other illustrations including the axon from the last post. Once the 3D was in place, most of my time went into Photoshop inserting textures and building up details.

1. Base Model and Rendering

For this image, most of the core elements were modeled including street infrastructure and context buildings. I didn’t have any site photos from this exact view angle nor did I take any at the correct time of day (busy week at work). This was fine though because the more I had in 3D, the more control I had over the placement of things like cars and street infrastructure.

2. Composition

In images that are pulled back like this, I love giving the sky a ton of real estate. If you cut the image into thirds, the skyline sits at the lower 1/3rd of the image. The horizon sits at the lower 1/6th of the page. All this means that the sky will become an important part of the image. Elements like street lights and bus cables can be dramatized by allowing them to break the skyline and cut across the page as you will see later on.

3. Sky and Texture

I spent a long time looking for the right sky and I am really happy with the one I found. The clouds are not overly dramatic, and there is a nice overall gradient from warm to cool. I also used Google Earth to collect street and sidewalk textures to help give things a little wear and tear.

4. Trees and Entourage

I went with fall foliage because I was worried to much green would steal some of the warmth of the early morning scene. Plus, the orange and yellows play well with the bright red of the new station design. For entourage, I placed just a few, but making sure they were all back-lit.

5. Cables

There are electric buses that run up and down Mass. Ave. which means cables everywhere. Power lines can sometime ruin the atmosphere of the image, but in this case, they add some really nice movement and texture. It took some trial and error, but I was finally able to Photoshop them in such a way that they work compositionally. It can be tricky to insert cables in Photoshop along with all of the attachments and fixtures, but they ended being my favorite part of the illustration.

6. Final Image

I didn’t do anything crazy with the toning. I just added a hint of light to the left, played up the glare, and increased the saturation. Some images, I will spend hours playing with tones. For this one, it happened relatively quickly.

For an image like this, most of my time went into prepping the 3D model. I downloaded a bunch of low poly cars and people and copied them all over the site. I also made sure to add in all of the street infrastructure such as street poles and traffic lights. Once I had the 3D model in a good place, I moved into Photoshop where I added the color, texture, and street markings. What I like about an image like this is that it reads well from both a distance and up close partly because it is still diagrammatic in nature while also containing lots of details. I can see this image being presented as a full image in a portfolio with additional zoomed in shots on corresponding pages allowing for detailed explanations of the different areas.

1. Base Images

Above are some of the base images rendered in V-Ray and exported from SketchUp. You can see in the model that a lot was done in 3D including the people and cars. As I mentioned above, these were really low poly models because I wanted to keep things diagrammatic and not draw too much attention. I am also showing the Material ID pass from V-Ray as this was really important to setting up this image. The Material ID allowed me to quickly and accurately make selections of different parts of the image in Photoshop to add color and texture.

2. Texture

Lots of textures were added in Photoshop such as street, sidewalk, and roof grain. This texture also started to setup some hierarchy in the image with the streets being the strongest element.

3. Ground Plane Detail

All of the striping and street markings were also added in Photoshop. There was a lot of experimenting with how I wanted this to look which is why so much of it happened in Photoshop vs texturing in 3D. However, because everything was organized carefully using groups and masks, the striping and road markings are all still very editable in the off chance that I want to explore other ideas down the road.

4. Color

As much as I love black and white images, I knew I needed to add color to define the project and better setup the reading of the axon. With that said, I kept color to a minimum and focused primarily around the site and not so much on the context.

5. Final

Finally, I used the Photoshop plugin “Topaz” to help punch up the textures and warm the image just a bit.

To see all of my past winter special images visit some of the links below:

Winter Special #7

Winter Special #6

Winter Special #5

Winter Special #4

Winter Special #3

Winter Special #2

Winter Special #1

1. Base Rendering Cleanup

A simple V-Ray clay rendering was generated from the Sketchup model. I then cleaned up that rendering in Photoshop by adding a white poché to the section cut as well as people for scale

2. Smudge

Adding “Smudges” is an easy way to break up the perfect gradients of the base rendering. Here, I am just painting in some black paint using a soft brush at a very low opacity. I am focusing the the smudges at corners and where shadows are collecting. You will also notice that there are moments where I masked the smudges for sharp edges to create an ambient occlusion effect to play up certain geometry planes. Finally, I added some tone to the sky to better define the horizon and actual section cut.

3. Micro Textures

I wanted to further break up the perfect gradients of the clay model rendering, so I added textures to specific moments around the image. These textures added a hint more detail when zoomed in close to the image.

4. Macro Textures

A texture was applied across the entire image. Since the textures in the last step were meant for a closeup view, the texture used here is meant to read from a distance and won’t be that perceivable up close. The effect is subtle, but helps further breakup the smooth gradients in the sky and over the entire image in general.

5. Noise

I’m in a “Noise” phase at the moment where I seem to be adding noise to every image I create. I like how it breaks up the super sharp edges of a rendering and adds a bit of abstraction. In the case of this image, I applied two layers of noise. The first layer had a larger scale so that it would read at a macro scale. The second layer of noise had a much finer grain to it and was meant to read better when zoomed in close to the image. Both layers had a lowered opacity so that they somewhat blended together.

6. Color

Finally, I added a hint of color to the image and darkened everything slightly. This section cut illustration with the desaturated tones and layers of texture is one of my favorite style of images to work on. From a technical standpoint they are easy to create. Even though it has an old school feel to it, the image still works well as a diagram and clearly conveys some complex information. I am planning to follow this up with an aerial axon illustration soon.