KROB 2014

The 40th Annual Ken Roberts Memorial Delineation Competition is now accepting your submissions. This is one of my favorite competitions because of the wide range of styles and techniques that win each year. It is open to students and professionals and can be either a physical submission, digital illustration, or hybrid media.  I was lucky enough to be one of three jurors last year and was blown away by the art that I saw. I came away from the experience with a lot of inspiration and motivation to keep experimenting.

This year's competition includes a stellar list of jurors including Frank Ching (I own two of his books as do probably most architecture students and professionals), Thomas Sériès who is the founder of Labtop, and Clifford Welch who is the owner of Welch Architecture.

The submission deadline is October 27th, 2014 5pm CST. Winners will be announced Thursday, November 20th, 2014 6pm CST.

Below are some of the winners from last year. Be sure to also visit the KRob website and thumb through the past winners. There is some real beautiful work and inspiration hidden away in the archives. Good Luck!


Castaneda Eduardo, AIAS UT Arlington- President

UT Arlington

Best in Category - Student Digital/Mixed




Chris Cornelius 


Best in Category - Professional Digital/Mixed




Samara Hayes

Abedian School of Architecture, Bond University

Best in Category - Student Hand





Dustin Wheat

University of Texas - Arlington

Best in Category - Professional Hand




Changyeob Lee

Royal College of Art

Juror Citation



Gary Schuberth, AIA, NCARB, LEED AP

Juror Citation



Matthew Bohne

Rhode Island School of Design









Over the past several months, I have developed many illustrations for my conceptual wharf design. I wanted to start thinking about how these illustrations would be organized into a portfolio layout. One of my favorite parts of arch viz is page layout. Last year, I put together my Portfolio Vol. 3 which combined many different types of illustration styles and techniques. At first, I saw it as an experiment and not really sure how successful combining the wide range of images would be. However, once finished, I think the juxtaposition of the different styles generated a better portfolio.

Going into this new round of spreads, I want to continue this idea of mixing styles while still maintaining control and flow of the story. I will also be exploring portrait orientation. I have typically only designed portfolios in a landscape layout in the past. Portrait layout offers a lot more options and possibilities with the online printers which I plan to explore later on. Beyond printing options, I just want to see what other advantages and disadvantages there are with portrait vs. landscape. 

Below is the first round of spreads of the wharf project. The tone of the spreads was set with the first image of the foggy morning illustration. The layouts still need to be refined but the structure and coloring are in place. This early assembly of the pages will also help me to determine what information is still missing and which images need to be created next.




I also wanted to make a quick anouncement that the softcover version of Portfolio Vol. 3 is sold out. There are only about 50 copies of the hardcover left so if you had your eye on one, now is the time to get it. Thanks to everyone who has already purchased a copy and helped support this site.


Breaking Up Edges Using Masks

We have used this style a couple of times in the office and I am a big fan of it. It is also really simple to implement. The idea is to apply a mask over a full resolution illustration to break up the edges giving the illustration a collage-like feel. The great thing about this technique is that while you are getting a less formal collage style, the image still has all of its information, detail, and resolution where it counts.

I have chosen two new views of my wharf design to show how it works. The mask is best applied at the very end which means a full illustration edge to edge is still necessary. This also means if the mask isn't working, it is no big deal and can be removed without any harm to the illustration. 

Below is the full frame base rendering before masking (desaturated to better show the masking).


The blue represents what I masked off. The mask was made by creating a new layer and moving it to the top of the other layers. I then selected the "Polygonal Tool" and began selecting areas that I wanted to mask. Once the selection was made, I filled in the selection with a white paint. I went through many of iterations testing out what areas should be masked and what works best compositionally. I experimented with softening the edges or lowering the opacity, however,  I ultimately preferred the crisp edges and the sharp contrast between the white and rendered areas.


It is a simple move, but one that completely changes the tone of the image. 


The final illustrations were rendered with V-Ray at a resolution of 4500px X 2500px. I then tweaked colors, textures, and added people in Photoshop. Finally, I jumped into my Topaz Labs PS plugin to punch up the detail.



This was a fun image to work on and the breakdown is relatively straight forward. The big poetic moves such as fog and coloring come towards the end which is typically the case with images like this. There was a lot of prep in Photoshop such as refining textures, lighting, and adding entourage. But once those elements were in, some serious experimenting began to generate the final mood of the image. I have continued to tweak the image from the last post, focusing mostly on color. Because of the way this image was set up, the image is very flexible and could easily be manipulated into a dusk shot or even a night shot. However, for this breakdown, I will stay with the morning shot. 


1. Composition

An often overlooked element of an architecture illustration is composition. Going into this image, I wanted to capture the scope of the project as much as I could showing the length of the pier as well as its relationship to the water and city. I also wanted to express the topographical elements of the design so I brought the camera as close to the water as I could while still being able to see the deck lighting. The image is positioned so that the horizon sits right on the lower 1/3 of the image. I knew I would be recreating the sky in Photoshop so I simply added more height to the canvas in Photoshop to shift the horizon to the lower 1/3 of the page. The center of the foreground building's long facade is placed directly on the left hand 1/3 line of the image. After finishing the image, I realized the real focal point was actually the green wall in shadow. It would have been ideal to place this directly on the left hand 1/3 line instead of the other facade. This could have been accomplished by cropping the shot, but I wanted to keep as much of the docks and city towers as possible, so I left it as is. 


2. Base Rendering

I'm not going to spend too much time here but I thought I would give some basic settings that I used in V-Ray. Before I rendered, I edited the model to help with the reading. For example, little things like adding slight overhangs at each step to create shadow really helped  to define the step edge and topography of the wharf. I also added lights embedded into the pier by creating a material in Sketchup, painting that material in each location of the light, and then telling that material to emit light in V-Ray. However, running out of time to tweak the lights to my satisfaction, I abandoned the rendered lights and decided to paint in the lights using Photoshop instead. 

I used an HDRI for the sky and background seen below which was downloaded from Along with the HDRI sky and background, I also turned back on the V-Ray sky with an intensity lowered to 0.02 to generate stronger light hitting the surfaces. The biggest advantage of V-Ray over Kerkythea for me is the ambient occlusion settings. This really helps pull out the detail and depth in the base rendering. The AO settings used in this scene were: Amount- 1.0, Subdiv- 16, and Radius- 40. 

Other than that, I used the physical camera to control the exposure and set the output resolution to 4500px X 2520px.

Above, the final base rendering from V-Ray. 


3. Alpha

The first step in Photoshop often begins with separating out the background from the base file. Most rendering programs will render an alpha channel making the selection of the background much easier. Removing the background allows for better flexibility when adding the sky and background buildings.


4. Background Texture

Next, I searched Google for images of the surrounding context. I then begin sampling the buildings, desaturating the texture, and applied them to the base rendering. I used a range of layer blend modes, but "Multiply" seemed to work the best in most cases. I applied the textures one surface at a time allowing me to match perspectives for each facade.


5. Lights

Knowing that thick fog was going to be added later, I like the idea of the lights in the city bleeding through. I spent a little time painting in light and punching up the lights around the design. I go into greater detail explaining the process of painting in light in my PAINTING LIGHT tutorial .



The design calls for lights imbedded in the deck of the wharf. As mentioned above, I had originally rendered the lights in V-Ray, but didn't have time to refine the settings. I ended up painting in the light which allows me to hone in the look that I am going for. The process is very similar to adding the lights in windows described above. The important thing here is adding that little bit of glow to each one.


6. Entourage

People are important for understanding scale in a design. Therefore, it is crucial that they are shown at the correct size. I often throw in some 2d people in Sketchup at different locations of the site, then export a 2d graphic of the view, and bring that image into Photoshop. This provides a clear reference for proper scaling when I begin to insert high res images of people. Since the scene has low light, I darkened the people to have more of a silouette effect. This minimizes the destraction of bright colored clothing or improper lighting. It's also important to note the reflections on the ground. To some, this may be a subtle move, but this helps to "ground" the people and set them into the image. Shadows have the same effect in daytime scenes. Without shadows or reflections, the people will appear to float and seem out of place. Awhile ago, I put together a tutorial on adding people which can be seen HERE.


7. Sky

The more time I spend working on architectural illustrations, the more I realize the importance of the sky. This often will make or break the illustration. In this case, I was looking for something that would work well with a foggy scene but not distract  from the design. I found some images that gave me the flexibility to experiment with the softness of the reading. I ended up desaturating the sky textures and overlaying them on the background which then picked up the blue tones already in the base rendering. I then feathered out the edges so that the sharper texture of the clouds remained in the center of the image drawing attention back to the design.


8. Bring on the Fog

For many people, this part can be intimidating. In reality, there is nothing to it. The most important thing to remember is to slowly layer on the paint. This means setting the brush opacity low, below 10%. Rendering programs like V-Ray can render out a channel called "Z-Depth" which renders a gradient based on how far an object is from the camera. The problem I have with this channel is that it is too perfect. Especially for a foggy scene, the fog is inconsistent in density. I was looking for a loosness and softness that the Z-Depth channel couldn't provide. I therefore manually painted in the fog. I did two passes shown below. The first pass was to highlight the light coming from behind the city creating a glare look. The second pass represented a denser more encompassing fog. I switched the fog color to red to better explain where it was placed. In this case, the imperfection works to my advantage.  You can see a more in-depth explanation at my "Adding Depth Via Fog" tutorial .


9. Color Overlay

I almost always end with some kind of color overlay. The color overlay does two things. It takes the color range of all the different photoshopped images and brings them together into a more cohesive look. Color overlays also allow me to quickly cool or warm and images changing the entire feel of the illustration in a few seconds. In this case, I was going for an early morning atmosphere, so I looked at several color options. I tested a red overlay at a very low opacity first and then looked at a blue overlay which I ultimately ended up going with. I also really liked the image completely desaturated. The black and white seemed to pull out the texture of the sky and enhance the drama. For more on color overlays, visit THIS POST.



Moving forward, I plan to render out a few more views exploring some different graphical styles not yet addressed on this site. I'm extremely excited about picking apart this design and seeing what kind of new imagery can be extracted. More on this later...






Modeler: Sketchup

Renderer: V-Ray

Post Processing: Photoshop

Final Output Size: 4500 px x 2520 px


The Boston Long Wharf perspective illustrations are underway. This first illustration is one of the more important ones because it is describing the entire design and its relationship to the water and the city. An early morning fog scene has been on my to-do-list for a while and this view/project seemed like a perfect opportunity to give it a try. One of the toughest things I found working on an illustration like this was keeping the overall color saturation to a minimum. I have a tendency to amp up the color in past illustrations through lots of color overlays. With this image, I constantly was going back and desaturating layers to maintain that early morning mood.

This illustration took me a little longer than the others for a few reasons. I have been designing and building the 3D model as I go. Typically, I spend a few hours when I can after work during the week on the modeling. It got to the point on Friday where I had to buckle down and make some design decisions and stick with them for the final illustrations. The good thing was once these decisions were made, I was able to quickly brute force model and get things where I needed them to start setting up the rendering.

I have attached some of the screen shots of the Sketchup model below.


The other thing that absorbed some of my time was setting up the file to render. Anyone that follows this site knows I prefer to spend most of my time in Photoshop. However, with this shot, I decided to invest a little more time preparing the textures and lighting. V-Ray was used for this image instead of Kerkythea, a rare occurrence on this blog. Below are some screen shots of the raw rendering output from V-Ray.

I kept most of the surrounding context largely untextured. This area of Boston is highly photographed online and I knew I would be overlaying images of the background buildings on top of the base rendering. For that reason, I rendered the context gray allowing me to overlay the photography but still maintain the light and shadow of the base rendering.


You may have noticed the title says, "Part 1". I plan to salvage what I can of this 4th of July weekend and pick back up in the next post with an in-depth break down of the post processing in Photoshop. These types of illustrations are fun to work on and I think the workflow would be interesting to see. 




Favorite Architectural Cutout and Texture Resources

Architecture visualization can consume a lot of resources and time if you're not careful. With the tediousness of Photoshop and renderings calculating through the night, spending more that 20 minutes searching Google images for a specific texture or image can become a breaking point for many. Any way to save time in this field is a big advantage, especially in school. The thing is, there are a lot of free resources online that have already put in the hard work so that you don't have to. Many of these sites were life savers going through architecture school. 

Below are some of my favorites websites containing free cutouts and textures. Some of the resources made the list because of their clear organization of the information. Often, there are very specific demands of a project or client that require lots of time online searching. These sites offer smart labeling and clear categorizing that makes finding specific items painless.




Skalgubbar offers up a large library of high resolution cutout people in PNG format. The people are often in unique situations which don't work for all illustrations. At the same time, the unique positions can also help add a bit of character to some less exciting images. Either way, I enjoy the "tumblr-like" display allowing you to quickly browse through a lot of images.






Immediate Entourage is another great site to find a large collection of cutout people. The resolution of the cutouts are hit or miss, but the library is comprehensive. What is even better, I.E. also provides many cutouts of vegetation. Everything is well labeled so it is easy to find what you need. 





Gobotree is a large and growing library of free images and cutouts created by the people at Vyonyx. By using the categories drop down at the top of the page, you can find extensive libraries of cutout people and vegetation to quickly populate an illustration. I especially like their sky collection and visit it often. I'm impressed with how this site is progressing and hope they can keep it growing. You do need to register for a free account which limits downloads to 50 images a month.







Mr Cutout provides extremely high res cutouts from a descent size library. The problem is that once registered, you are limited to 2.5 mbs a day. Just a few downloads can consume the entire daily allowance. This can be limiting with only a few days to produce an illustration. However, the quality of the cutouts are some of the best and can be a good resource for some of the close-up foreground elements.







This site offers full cutout collections of people in specific situations. Examples include a collection of office scenes, coffee scenes, people scene from above, people scene from below, etc. Green screens are used meaning the cutouts are of the highest quality. XOIO also provides many other collections of 3d models and backgrounds that can be helpful as well.






Probably one of my all-time favorites, I visit this site every time I am working on a new illustration. CG Textures is a huge database of jpg textures, many of which are tileable. This works great for texturing in Sketchup as well as in Photoshop. What I appreciate the most about this site is its incredible organization and in-depth library. It's my go-to place when I need to find very specific and unique textures. The free account allows you to download 15 mbs a day.






Similar to CG Textures, Texturer does a good job of organizing the images. However, few of the textures are tileable. Though this limits my use, there are still many useful textures to be found once I'm post processing in Photoshop.






What I like about SWTexture is the size of the tileable textures. When used in a 3D model or in Photoshop, the tiling is barely noticeable. The blog focuses on commonly used materials so you will be hard pressed to find anything unique or specific. However, the textures that are provided are of great quality and worth checking out.






Not only a great blog, Matúš of FlyingArchitecture provides several V-Ray textures which include bump and displacement maps. Though the textures are created for V-Ray, you can use just the diffuse images in your 3d model and in Photoshop if needed. 






This site is a pleasure to browse and beautifully designed. Rendertextures is a spinoff of FlyingArchitecture so the images are of similar quality and also contain bump and displacement maps. Beyond that, the information is well organized and easy to navigate. This is not a large collection but a good suppliment to what you may already have.






While there are some free cutouts, I personally like Tony Textures because of their many high res texture images. Though most are not tileable, the viz images such as skies, ground planes, water, etc.  go a long way in Photoshop. I often explore this site if the other resources fail to deliver. 


I'm always looking for more resources so if there are any that I missed, send me an email or leave a comment. You can never have enough entourage.








In the previous wharf post, I discussed black and white texture studies that focused on pattern and ground plane elements throughout the wharf. The above image builds off of those studies introducing color and material. Site plans are extremely useful in the design process and I wanted this one to be somewhat malleable as I tweak things. 

The wharf is on a strong axis with Boston's Old State House. I wanted to be able to study this connection as well as the wharf's relationship to the dense city fabric. The site plan was therefore setup to include this information which meant creating one of the highest resolution images I have ever done. The above image tops out at 10,000 pixels tall. Typical illustrations for me are around 4,000 to 5,000 pixels on the longest edge. The high resolution was needed to be able to study different spaces and textures up close. However, this will also allow me to crop out several close-up images for presentations allowing me to get lots of mileage out of this one site plan.

To manage the high resolution, I rendered out two 5000x2300 images seen below. To speed things up, I didn't worry about adding texture or material in the Kerkythea rendering. I knew I would be experimenting with textures in PS and didn't want to have to worry about baked in materials. I only needed the shade and shadow so I rendered some clay models and stitched them together in Photoshop.



The final PSD file can be broken down into three main sections. I used an aerial image to get much of the existing texture information of the roads, buildings, etc. The aerial was overlayed on top of the clay model with a lowered opacity. I then took the black and white texture studies from the previous post and worked that information into the new site plan. Since all of the line work elements were separated and grouped in the previous PSD file, it was easy to make adjustments and extend the ground plane elements to the Old State House. Finally, I added texture, color, and vegetation on top of everything.


With the site plan far along, I'm ready to start diving into some perspectives. I hope to get going with those in the upcoming posts along with some portfolio spreads. I'm thinking about setting up a vertically oriented portfolio layout as this is a more common layout with the online printers. More to come on this later.




It was a little over a year ago that I started a visitor gallery tumblr page in an attempt to generate a another place on the internet for architecture visualization inspiration. It has been exciting to see so many unique styles and methods submitted week after week. I curate the gallery which means not every illustration submitted will get posted to the tumblr page. As a matter of fact, only about 1/3 of the images make it to the tumblr page. Total, there have been 260 illustrations posted since its inception.

I thought I would highlight some of my favorites in no particular order. These images were chosen for many different reasons: uniqueness, composition, color palette, style, technique and execution. Each of these illustrations combine those characteristics to create a mood and personality that draw in the viewer. Without knowing more about the projects, they are very engaging images on their own and a great source of inspiration. 


Rayal Arnand Yelamarthi


PRPSFL Studio:  Locis - Site sensitive train gantry design


Scott Tulay: Nests: Ink and graphite on illustration board.


RHINO, V-Ray, Photoshop, Exterior Perspective


 Original photography collage recreated in Illustrator and Photoshop


Jeffrey Bermudez: Bernard and Anne Spitzer School of Architecture: City College of New York


Ali Mustafa: Channel Orange: Photoshop, Sketchup


Jose Alvarez: Samana, Dominican Republic: Topographical Analysis: ArcGIS, Illustrator, Photoshop


Caleb Lowery: Green Wall St.: Adobe Photoshop: Photoshop collage created while working for Terreform ONE


Graduate Health Care Studio: Metal Health and Wellness Clinic: Sketchup, Podium, Photoshop


Minh Duc Le: Broken Vision: Sketchup, Vray, Photoshop


V & A, London 


Birds eye view: Site plan


I am looking forward to seeing what gets submitted next and how this gallery evolves. New images get posted every week, so be sure to check back often.  




The last few posts have been focused on diagramming the existing conditions of the site. In this latest post, I have shifted to studies for the new design. The overall concept and form of the Sketchup model is established, but now I am interested in exploring texture. I trid to keep things abstract before getting too specific with thoughts of actual materiality and color. The above black and white approach was one way for me to keep thinking conceptually without losing myself in the details. As much as this image looks "finished", the workflow was very much a loose and explorative process. Many iterations and ideas were tested. Changes in line weights, shadows, and hatches helped define hierarchy and texture qualities. 

From here, I expect the illustration to continue to evolve as the project is refined and more decisions on form and material are made. The process of creating an image like this in Photoshop is "sketchy" in nature which lends itself well to exploring ideas along side 3D modeling. Later on down the road, I may begin introducing color as ideas of materiality become more concrete.

Below is a quick breakdown of how the image was created. Everything was made in Photoshop except for the initial line work and shadows which were exported from my Sketchup model. For the most part, I used the Brush Tool to create most of the elements. I also applied drop shadows and strokes to the elements when needed.


1. Sketchup Line Work


2. Shadows Exported from Sketchup. 

I adjusted the levels to take the shadows from grey to black.


3. Water Poche

I darkened the water slightly to better clarify the relationship of water to land.


4. Profile Lines and Guide Lines 

Thick profile lines were added to punch up important elements of the design.


5. Abstract Boats

Simple rectangles with shadows were used to represent the boats in an abstract manner. These were then copied many times to imply movement.


6. Thick Dashed Line Work

The thick dashed lines represent an abstraction of several elements of the existing site. Density and location were determined by topographical changes and potential pedestrian traffic.


7. Thin Dashed Line Work

A second layer of thinner dashed lines were added. These are hierarchally weaker than the thick dashed lines and could represent a change in material among other things.


8. Poche

A final layer of shading was added to highlight potential locations of green space.


Though the workflow has been broken down into a very systematic explanation, the actual process of creating this image was not as clear cut. As mentioned above, it was a very iterative process where all of the element were slowly built up together through additive and subtractive means. There was a lot of copying, pasting, transforming, and scaling. Less focus was placed on craft and more on a relaxed and experimental process.







This post is also a follow up from many emails asking me to explain the diagonal line hatching used the site analysis diagrams. There are two methods in Photoshop that I know of that can create the diagonal line hatch seen in the image above. Both options use a relatively fast workflow and are easy to execute.

OPTION A: Google Image Overlay

A1. This option involves simply going online and finding a hatch pattern. In this case, I Google searched "diagonal lines" and found several images that would work. The images do not need to be large because I can copy the texture many times to create a much larger texture for what I need. 

If you do need to copy a small image many times, hold down the "Alt" key ("Option" on Mac) and drag the image using the "Move" tool. This quickly duplicates the layer. Once you have all of the layers aligned forming a larger image, they can be merged into one layer by selecting all of the individual layers in the layers palette, right click on one of the layers, and choose "Merge Layers". 

A2.  Next, the white background needs to be removed. This is easily done by selecting the new hatch layer, then choosing "Multiply" in the layer blend mode drop down in the layers palette.

OPTION B: Create a Custom Pattern

A second option to create a hatch is to define a custom pattern which can offer much more flexibility. This method will often generate cleaner line work than what can be found on the web. One other thing to note is that many PS patterns can be found online through websites like Brusheezy which can be imported into Photoshop.

B1. I want to create my own custom pattern. To do this, I first need to create a new document in Photoshop by going to "File>New". I set up the new document to be 12x12 pixels, however this size can change depending on your needs. Also, double check to make sure the background contents is set to "Transparent". 

B2. Next, select the "Rectangular Marquee Tool" and make a selection crossing the entire document. I made my selection 2 pixels high. The Rectangular Marquee Tool should snap to each pixel making it easy to determine how many pixels are being selected.

Choose the "Paint Bucket Tool" on the left and also select a black paint color. Then, fill the selection. 

B3. The pattern is ready, so now it needs to be saved into Photoshop as a pattern. Choose "Edit>Define Pattern". In the dialogue box, give the pattern a descriptive name and choose "OK". 

B4. To apply the pattern to the illustration, first create a new layer and move it to the top of the layers palette. Then, select the "Paint Bucket Tool". In the paint bucket options at the top, be sure to select "Pattern" in the drop down and set the Opacity to 100%.

To the right of the pattern drop down, there is another drop down which will show a thumbnail of all of the saved patterns. Choose the pattern created in the steps above.

With these options set, use the paint bucket to fill the page making sure it is on the new layer.

B5. Since this pattern was applied on its own layer, it can be rotated 45 degrees to create diagonal lines. Choose "Edit>Transform>Rotate" and hold down "Shift" when rotating to snap to 45 degrees.

Use Masks to Hatch Specific Elements

The hatch patterns are rasterized on their own layer meaning the hatch can simply be erased using the eraser tool. However, applying a mask will allow for editing later on down the road if changes need to be made. 

1. Apply Mask

To apply a mask, select the hatch layer and choose the "Add a Mask" icon at the bottom of the layers palette.

2. Make Selections 

Next, use the "Polygonal Lasso Tool" to select the areas of the illustration that will be filled with the hatch pattern. 

3. Inverse the Selection

I want the selection to be what I want erased, therefore I am going to inverse the selection. Choose "Select>Inverse" at the top.

4. Fill the Mask

Finally, choose the "Paint Bucket Tool", check that the opacity is set to 100%, and also that the drop down is set to "Foreground". 

The mask works by using gray tones to determine what parts of the image will appear or not appear. Black paint erases the image. White paint will reveal the image. I don't want the hatch to appear in the part of the image that is selected, therefore I want to choose black paint.

Double check that the layer mask is selected which is the white box next to the layer image in the layers palette. Using the paint bucket tool, fill the selection with black paint. This will "erase" away the hatch pattern where the selection was made. White paint can be used to bring back the hatch pattern later if changes are needed.

Below, the final result using patterns and masks to give a little more texture to the image and help define important areas of the diagram.