To start, this image only uses a shadow and line work 2D image export from Sketchup, and a clay rendering from V-Ray.

1. Setup the Shadows

In Photoshop, I brought in my Sketchup shadows and line work layers. I colorized the shadows to give them a blue tint since the overall tones of the illustration will be on the cool side.

2. Overlay the Clay Rendering

I wanted to see a little more detail in the forms, so I brought in the V-Ray clay model rendering and overlaid it on top. I lowered the opacity so that the clay rendering gradients remain subtle.

3. Bring in the Color

Once I had the base images setup, I started to lay in some color. I applied the color one zone at a time, starting with the sky, moving to the context, then ground plane, and finally the design. I typically try to get all of the different color zones masked in fast so that I have more time to make adjustments to the colors and experiment with which palettes work best. I almost never get the colors right the first time.

4. Sky Detail

I struggled with the sky but finally arrived at a hybrid style of real and illustrated clouds. I was looking for something that was subtle, added a little texture, but also worked with the minimal style of the image.

5. Street Life

Next was adding life to the public realm. I brought in trees, people, and some cars. All of these elements were placed into groups within Photoshop so that I could colorize everything at once by applying a layer style. I did this by selecting the group, choosing “Layer>Layer Style”. In the dialogue box, I chose “Color Overlay” and selected a color that worked with my composition. Colorizing all of these layers removed the detail giving everything a silhouetted abstract look.

6. Noise Overlay

The last thing that I did was add some noise to the image which softens the edges, helps hide little imperfections, and give a retro look to the image. To do this, I created a new layer and filled it with 50% grey. Next, go up to “Filter>Noise>Add Noise”. In the dialogue box, I adjust the “Amount” slider to around 60%, set “Distribution” to “Uniform”, and checked the “Monochromatic” box. Finally, I set the new noise layer to “Overlay” to apply the texture to the image. I also lowered the opacity a little to reduce the strength of the effect.

The Final Image

I typically like to start a project off with a brutally simple layout and large text signifying the start of a new project within the portfolio. One of my biggest pet peeves is looking through someone’s portfolio and not knowing where a project ends and another one begins.

I also think it is important to have a spread early on in the project describing the location and orienting the viewer.

From there, I introduced the form and a break down of how and why it was designed the way it was.

At this point, I though it would be interesting to jump out of diagram mode and into some vignettes at street level. The previous spread talks about the street engagement diagrammatically so these spreads will go into more depth on the subject.

I’m not into rendering out floor plans but instead prefer to keep them extremely diagrammatic and easy to read. I also like to combine floor plans with elevations and sections so that the viewer can better see the relationship between them. One thing that I need to fix with this spread is to size down the elevations to be the same scale as the floor plans and add graphic scales. This will help avoid any confusion and aid in the reading of the page.

The snow scene seemed like a good image to end the project on because it is memorable but also because it is a view of the entire design and shows its relationship to the surrounding context.

Time to move on to the next design. If you have any suggestions for what the next project should be, leave me a comment or send an email.

1. Base Rendering

The base rendering was pretty brutal. However, I was just looking for a rendering with soft shadows and some basic texture information.

2. Clay Rendering

To get the image closer to a winter scene, I desaturated the colors and added a clay rendering to the ground plane. This scene will have snow on the ground so using a clay rendering acts as a good base to Photoshop snowy textures on top of. A clay rendering is the same as a typical rendering except that all materials in the scene are replaced by a simple matte gray texture. I have a tutorial explaining how to create a clay rendering using Kerkythea, seen HERE.

3. Texturing the Ground Plane

As discussed in my last post, it’s all about the textures and that is what I focused on for this step. I combed through the internet looking for textures of tire tracks, foot prints in snow, and snow covered plants. I then slowly layered them on top of the clay rendered ground plane. Because they are black and white snow scenes, I didn’t need to cut anything out. I only needed the dark part of the textures so I just set the layer blend modes to “Multiply”.

4. Trees

Oddly enough, this step went fast. It is very easy to find winter trees with snowy white backgrounds. As with the above step, no cutting was needed. I only need the dark parts of the tree images so I set the layer blend mode to “Multiply” which gave me trees without the backgrounds. Nailed it.

5. Fog

This step is where the image makes a huge mood shift. The more I create visualization, the more I realize the power of fog/haze/atmosphere. It’s something that has taken lots of practice and not the easiest thing to explain in a tutorial because everything is so subtle. I have attempted a tutorial in the past found HERE  though I am probably due for an update soon. I also added a blue color overlay over the entire image.

6. Building Lights

Lighting up the architecture provided an opportunity to bring the focus of the image to my design. I therefore lit my design much brighter than the surrounding context bringing hierarchy and drama to the illustration. A tutorial for manually lighting a building can be found here.

7. Ground Plane Lights and Details

I needed to activate the ground plane so I inserted some people, car light streaks, and because this is the winter special, some tree lights. I was careful not to overdo it because I didn’t want to draw too much attention away from the architecture.

8. Color Editing

The illustration was coming off  too gray for my taste so I tweaked the colors to be more saturated and blue. I also added a little snow falling just to bring home the winter scene feel.

Base Rendering

Texturing in the 3D environment is important but it can be difficult to avoid “tiling” of textures. Often, my base renderings come out looking flat and “perfect” meaning the texture has no flaws. This is because there is only so much I can do with bump maps and texture attributes. At a certain point, time spent texturing in the 3D environment could be done much faster in Photoshop. After producing so many illustrations over the years, I am starting to get a sense of what that threshold is and when I need to get out of 3D and into 2D post processing. In many cases, I need both a good texture in the 3D model combined with a good texture in Photoshop to get the right look.

Texture Zones

For this illustration, there were four areas that I focused on: the roofs of the surrounding context, the textures of my design, the streets, and the sidewalks/landscape. All four of these areas feed off of one another and if one is left untouched, then the computer’s plastic feel is revealed and the effect is lost.

I used several images of asphalt roads to apply textures to my illustration in Photoshop. These have the right amount of grunge and imperfection and can be used in areas other than the streets. In this case, I used the asphalt textures for the roofs of the buildings and lowered the opacity and applied them to the concrete and brick sidewalks.

The Comparison

A side by side comparison reveals how powerful a little bit of texturing in Photoshop can be. It’s elements like roads and roof tops that I often see people ignoring in their illustrations that I have come to realize make a big difference in giving an image that human touch. Building in these little imperfections start to trick the eye into thinking this is not a computer generated image.

The Final Image

In this final image, I moved the colors to the warm side by using orange color overlays. As with a lot of my illustrations, I bumped up the detail and sharpness using plugins like Topaz Labs Adjust. For aerials, I typically turn off the perspective but in this case I left it on and increased the field of view in Sketchup to make the buildings feel taller. The perspective works here because it draws the eye to the center of the image and puts more focus on the design.

One of the very first tutorials I created for this site described a workflow that generated a night scene only using Photoshop. That tutorial used a very simple base image exported right out of Sketchup. The workflow is easy to implement and is especially useful if you are not comfortable setting up night scenes and lighting in an external render engine such as Kerkythea or V-Ray. However, that same workflow is equally useful with more complex and developed scenes and this latest post is my attempt prove it. As I was working on the previous post, I realized the daytime scene could easily be shifted into a night scene to generate a more compelling image. Everybody likes a good night illustration but they also tend to be the most intimidating. The steps below will walk through how I took the fully rendered daytime scene from the previous post and turned it into a night scene using a easy Photoshop workflow.

1. Turn Off Layers in the Original Illustration

The first step is to get rid of some layers in the original daytime scene that don’t belong in the night scene. These layers include the HDR effects, warm color overlays, and fog. I also turned off the cars and people layers.

2. Remove Sharp Shadows

Probably one of the more difficult steps is editing out all of the sharp shadows created from the sun. In most situations, I used the “Clone Stamp Tool” to edit these out. The tree shadows were on their own layer so I could simply turn those off. This is an important step because leaving these sharp shadows in the scene will confuse the viewer and give the appearance that the sun is up even though we are trying to say that this is a night scene and the sun has already gone down.

3. Darken the Image

This is the heart of the workflow. The image needs to get darker and an easy way to do this is by filling the canvas with blue paint and setting the layer blend mode to “Multiply”. Once this is done, I then go back and mask out part of the blue so that the sky and interiors show through. This quickly sets up a decent looking night scene base image.

4. New Sky

Sky is important in shifting the mood from day to night. You will notice the night sky isn’t much darker than the day time sky. The only difference is that the clouds darken a little. It’s fine to go with a darker sky but I like the contrast of the light sky with the dark buildings. I also have plans to drop in reflections on the road and a brighter sky will help increase the drama with this effect.

5. Lighting

Time needed to be spent manually painting in light throughout the scene. My focus is on punching up highlights underneath the building as well as giving the appearance of light in the large curtain wall in the center of the image. I also added accent lights on the large metal fins and painted the windows in the surrounding context buildings. I put together a tutorial a while back detailing how to light up windows, seen HERE.

6. Wet Ground

A move I like to do with night scenes is to give the appearance that it just rained. The workflow is simple and the effect is very dramatic. It also adds another layer of complexity to the illustration that better engages the viewer.  I put together a video tutorial a while back seen HERE explaining how this can be done though my technique has since changed a little. I may break this out into its own mini tutorial later if I have time.

7. People

To help bring home the idea that it just rained, I inserted a bunch of people with umbrellas and rain gear. I didn’t want the people in the foreground to be too distracting, but instead read more as silhouettes. I therefore desaturated and darkened them, and edited the umbrellas to all be black. This put the emphasis back on the architecture but still activates the street life.

8. Color Editing

For the last step, I did some color editing in Topaz.  My goal here was to pull out some more detail and contrast, darken the image a little more, and shift the entire image to a similar range of colors.

If you haven’t already, check out my “No-Render Night Illustration” for a more comprehensive breakdown of turning a daytime illustration into a night scene and to get a better sense of the principles at work .

Architecture visualization is notorious for producing complex and messy PSD files. Sometimes, it’s the little things that can make a big difference and in my opinion, layer management is one of those little things that many people don’t put enough emphasis on. Properly managing your layers in Photoshop can lead to better efficiency, a more editable file, and in many cases better looking illustrations. Over the years, I have moved toward a system of building up Photoshop files. This system may not be perfect for everyone but has streamlined my workflow and has lead to cleaner looking illustrations.

The problem with PSD files is that they can quickly get out of hand. Many of my Photoshop files that I work on easily exceed several hundred layers. Being careless about naming files or not properly grouping them may save a little time in the beginning. However, once you are several hours into Photoshop, the time wasted trying to find the right layers or make changes far exceeds the time saved at the beginning. Not only that, the image itself will start to get messy and unrefined since it becomes difficult to mask layers and control lighting.

I put together an illustration that takes into account a lot of the different types of groups that I typically use in my PSD files. Obviously, the specific groups change from illustration to illustration depending on the type of image, but for the most part, the overall structure remains consistent.

One other note, I am in no way saying this is the “right” way to organize PSD files. I am sure there are other ways and even better ways to do this. However, this system works well for me and provides a good amount of flexibility depending on the type of illustration I am working on.

Groups, Masks, and Adjustment Layers

Groups
One of the keys to a clean Photoshop file are groups. Groups are folders in the layers palette that contain one or more layers. Groups allow you to collapse many layers down to a single line allowing you to navigate between several hundred layers quickly. One more big advantage is that you can apply masks to groups which is useful for editing. A group can be created by clicking the folder icon at the bottom of the layers palette. Layers can then be dragged into the folder one at a time or several at a time.

Masks
A big part of working efficiently and keeping your files extremely editable is by using “nondestructive” methods to edit your layers. The idea is to edit the illustration in a way where changes can be reversed later on down the road.
If you’re not using them already, masks are essential for nondestructive editing. I have talked about masks on this site only a few times because they can be difficult to understand at first. However, they are a life saver when it comes to cutting out backgrounds or removing parts of layers. Masks are exactly as they sound allowing you to mask or hide parts of the layer similar to erasing. The difference being the mask can be removed later to bring back the area of the layer that was erased. I talked about masks more in-depth in my Hatch Patterns Post about midway through.

Adjustment Layers
Just like masks, adjustment layers are nondestructive and allow you to adjust things like levels, contrast, hue, etc. to several layers at once and also reverse the adjustment at any time later on down the road. Adjustment layers act like real layers allowing one to adjust the strength through opacity or even apply a mask to the adjustment. I don’t use these too much but have found them really useful from time to time.

Organizing the Layers

Above is a screenshot of the layers palette for this illustration. You will quickly notice that everything is grouped. Having everything grouped allows me to collapse all layers and move around the palette more easily. It also forces me to really think about the best place to locate layers so that finding them later won’t be so time consuming.

1. Base Files Group

The Photoshop process begins with several images exported from both Sketchup and a render engine such as V-Ray. These images often include the base rendering, a reflections export, refraction export, line work, ambient occlusion pass, z-depth, Alpha, among others. All of these layers are placed in the “BASE FILE” group. These layers represent the core layers from which everything else in the Photoshop file builds off of. Once these layers are compiled, a mask is applied to the group to remove the sky and background elements using the Alpha layer.

2. Sky Group (Below the Base Files Group)

After compiling the base files, I bring in the sky. I like to do this early on because the sky plays a big role in setting the tone of the image. It can impact the coloring and mood so figuring this out first gives the image good momentum. I will often combine several sky images to get the exact look I am going for. Because I masked out the BASE FILES group, I can simply drag the SKY group below the BASE FILES group and shift and scale the location until I get something that looks good.

3. Background Group (Below the Base Files Group and above the Sky Group)

Along with the sky, I bring in the background elements which includes the surrounding site buildings and landscape not modeled or rendered. The background information comes from many places such as the client, Google image search, or Google street view.

4. Ground Plane Group

The ground plane consists of road textures, sidewalk textures, and green areas such as grass. At this point, the entire image has been tweaked and a good base is in place. From here, there will be a lot of detailing with light, cars, people, and trees.

5. Interior Group

In the case of this illustration, there is a lot of glass and interior information. I like to keep this a separate group instead of making these adjustments with the BASE FILES group. The reason being that this group houses all things interior including people, furniture, and lights which can sometimes add up to a ton of layers. The advantage of keeping all of these layers in one group is that I can use one master mask to cut out mullions and define the windows.

6. Tree, People, and Car Groups

Now that there is a good base, sky, and ground plane, I begin inserting, trees, cars and people. Each category gets its own group folder. Within each one of these groups, there is a subgroup for shadows. I place all of the shadows in a subgroup so that I can control the opacity of all of the shadows at once instead of dealing with opacity individually. This is also important when two shadows overlap. Since it is the group controlling the opacity, shadows overlapping inside the group will blend together instead of overlaying on top of one another.

In many situations, I will have multiple groups of trees and people. Sometimes there are trees in the foreground covering everything and sometimes there are trees in the background behind buildings. This requires separate groups to avoid complicated masking placed in different areas of the layers palette.

7. Effects Group

This group is always at the top of the layers palette and contains layers such as color overlays, sun glare, fog, and overall adjustments. Introduction of this group varies,  it sometimes happens early on in the process or not until the very end. However, this group is one of the most important in giving the illustration character and atmosphere.

9. HDR/Plugins Group

Occasionally, I will tweak the final illustration using some plugins/filters like Topaz Labs Adjust to bring out some detail and refine the coloring. This involves copying and merging all of the layers (ctrl+alt+shift+e) to get a single layer that the plugin can be applied to. I keep this group at the top so that I can lower the opacity and lessen the effects of the plugin if needed.

To clarify, most of my illustrations follow this structure with some minor deviation depending on the illustration type. For the purpose of this tutorial, I carefully named every layer but in the heat of the moment it is not always possible to take the time to perfectly name everything. I do however make a point to thoroughly name groups in all caps no matter how bad the time crunch. The point is to have a solid system in place that ultimately leads to a more efficient workflow and that produces cleaner work.