Oasis House

Its so hard to design buildings in this region, other than the copy-paste plastic garbage that usually comes out here.  A major reason for that is the lack of context.  There is barely any historical urban fabric to build within.  Dubai for example has its very rich old city around the creek and thats it.  Theres nothing wrong with that, it just that there isnt much to work with when ones approach is to respond to context.

I had to create my own by going against the weak existing context.  Jumping to the end, the final result, the intention as the name suggests, was to provide a hidden oasis: a minimalistic yet compelling external shell, that held surprises on the inside.. Continue reading…

a02- ground floor plan


Mr. White

I just realized that I havent put up a post on this, just a page at the back of the site.  This is just some fun I had while giving myself another tonal study.  Really enjoyed getting those wrinkles in the head.  Makes me want to try my hand at Yoda.  So theres a regular study of the famous photograph, a tribute to Magritte the surrealist painter, and finally something I just had fun with.  A page with a WIP can be found here.

Heisenberg_WIP_04 Heisenberg_WIP_05 Heisenberg_WIP_06

Reef Residential

Its disappointing looking back on this one.  Along with the Business Towers I posted earlier,  this one also broke ground, the first step in constructing the building.  The crisis hit and suddenly the project was canned.  It may still go up one day though.

Anyway, what I’ve put up are the earlier renders, the ones I felt best conveyed the concept, as well as the color coded design breakdown.  Continue reading and view more images…

reef residential A01-Typical Basement Plan

Windtower Villa

This is an old one from when I first moved to Dubai.  One of my obsessions is trying to modernize traditional motifs and nothing is more iconic in the UAE than the wind towers of their old traditional homes.  This was my first take on modernizing the modern courtyard/windtower homes.  Continue reading with more images…

sultan villa A16- Perspective

252 Towers

mubarak A01- List-Schedule-Analysis

252 was typical of a project I would do as an architect.  It would start with an analysis of the site in comparison to the building code as well as the client’s needs, conceptual designs, 3D renderings, and presentation layouts.  We can only build so much up to such and such height, before we can continue having only certain types of functions after said height.  A site would be zoned for a general urban function within which the client would say “I want to have,” such as in this case, “a mixed-use office and residential tower.

mubarak A02- Mass Plan

Anyway without boring you with the full process, we simply got on with building within the 3 general constraints of client, code, and context: the 3 C’s if you will in this approach of architecture design.

Along the way we not only design the form of the building but we also suggested additional functions such as the health club, half way up the building.  The health club, which is indoor within the larger tower and outdoor on the roof of the smaller one, served two purposes.  First, it obviously added value to the property, giving both residents and office employees a place to unwind.  Second, it helped create a buffer between the private residential and the more public offices floors.

mubarak A03- Basement Plan

One of the things I enjoyed working on these presentation boards, was looking for creative ways to breakdown and explain the organization of the building spaces for client consumption.  Most of the time, done well enough, the design itself will do most of that work for you, and you need to help it along slightly, as in below.

Its clear the building is divided into two parts with one side containing all vertical circulation.  Its just up to me to take it a step further and say what every sub-space is.

Even back when working on these and other boards, I noticed the influence of gaming HUDs.  With HUDs its so important to pack in as much information with as little impact as possible on everything else you are seeing.  Its the same here.

mubarak A04- Ground Floor Plan

These are all images of the layout as you slice through the building as you go up.

mubarak A05- 1st Floor Offices [podium] mubarak A06- 2nd Floor Offices [podium] mubarak A07- 3rd Floor Offices [garden] mubarak A08- 4th Floor Offices[typical] mubarak A09- 6th Floor Offices [mech] mubarak A10- Health Club

Some numbers are introduced off to the side to compliment the apartments on the inside.  A little gradient scale fill extending from the building to the HUD brackets to the side help make the connection between the numbers and their respective spaces.

mubarak A11- Typical Apt Floor Plan mubarak A14- Section

Similar thing is is done in the section.  The role of the health club as a buffer is clear here and is emphasized.

mubarak A15- Elevations mubarak A16- Perspectives low res

My designs tend to have clean minimalist looks.  I also like to have one feature that is exhagerated and characterizes the building.  This building funnily enough is inspired by ice cream sandwiches: Two thick walls on either end and a band of grills going all the way up one side, over the roof and down the back.

mubarak A16- Perspectives

mubarak A17- Perspectives mubarak A18- Perspectives

4 Years Old Today

Its funny to think that I started on this map 4 years ago, 3 years before even joining this Gaming Academy program.   Knowing what I do now, its nice to see some things inherently understood in design and its also good to see where I could have, and still can really, go back to improve the design.

Mapping for TF2 was an creative outlet for me.  It satisfied me both creatively and analytically.  I soon learned it was just about making up some original locale and plopping everything down in Hammer.  TF2Maps.net and the VALVe Developer Wiki were great sources of information.  I must have printed loads and loads of pages to read at night.

I had a love for a particular map on TF2 called CP_STEEL.  It was a 5 control point [CP] map where each CP affected the map somehow.  Bridges were extended, doors closed and other passage opened, depending on what CPs your team controlled.  I wanted to do something similar, but I also wanted a more impressive change in the environment. Cue, CP_Dam.



That there is my first sketch of CP_Dam. Its older than my son.  The basic idea was that each team would start on either end of a dam and would fight for 4 control points within the dam itself before being able to win the round by taking the final point on the enemies’ riverbed.

Taking either CP_A or CP_B would change the water levels on either side: raising them 1 level on your side and lowering them one level on their side.  What this would do was decrease your spawn-to-conflict distance while increasing the enemies’.  See the figure above and assume you are playing as BLU.  Waterline 1 is what happens if you take EITHER A or B.  Take BOTH and the difference is further increased, therefore making a greater difference in running/swimming distance between the teams.



At this state, the enemies’ home CP was now exposed and prone to capture for the win.  It also pushes the conflict location closer and closer to the enemies’ home CP therefore making it easier to capture said CP.

Whats really cool is that this sketch below is what I would later learn was graphing theory complete with nodes, edges, and an expression of weight/value to them.  Here I was trying to abstract what was too complicated me to understand otherwise: how each the value of each CP according to the distances they were from spawn.



There was a little more to the idea even at this early conception stage, such as CP_C extending a bridge from spawn to the dam, which made travel distance even shorter.  Although C could be taken any time, the benefit could only be felt if A+B were already taken, therefore compounding the distance reductions.

CP_D was a similar idea except it was a speed tube [read glorified waterslide] that took you from spawn to the enemy river bed.  This reduced the distance even further still, but D was at the bottom floor of the dam and furthermore, tubing was useless unless you had brought down the water level on their end to 0. i.e. you drowned.

I would later learn to tamper with capture speeds of each point, spawn rates, and to a lesser extent, geometry, but more on that later.  Now, 1 year into my chosen new career path, I’m happy to look back and see where it all began.

The Late Show | Level Progression 2

Continuing off of my last update on the RLD, we now needed to quantify the aspects of level design in order to be able control the level progression.  One of the first steps as defining the ingredients, as seen in the previous post.  Next I took the street intersections and tried to assign values to different types of intersections.  The idea is that different types of intersections provide different egress possibilities for the contestants.  The more possibilities to escape the harder it is for the player.

Since the cameras are what allow the player to see the contestants, I compared the 2 camera types we have vs every intersection type.  Below you can see every intersection type ( remember there were 6 in total ) checked with the two camera sizes: 2-unit cams on the left column and 4-unit cams on the right column.  Its important to note that each grid is equal to a grid within Unity space.

Intersection Metrics

So above, you can see how many cameras of each camera size are required to cover those 6 different intersections ( white text ).  You can also see how many egresses each camera can cover ( in red ).  By doing this exercise we were able to lock down the street sizes as well as the camera radii.  From this information I was able to discern 3 general categories of intersections and therefore difficulty: those that required 1 cam, those that required 2 cams, and those that required 4.

RLD Spreadsheet_01

By providing a value to each of those 3 categories and counting how much of each type we have in a given level we can get an idea of the overall difficulty of a level.  Of course there will be other factor involved such as the numbers of items a player can place each level as well as the number of contestants running, but the bulk is weighed by the geometry.  At least that is the plan.

Below is an image of a placeholder level 7.  It started off as a block for block duplicate of the one in unity.  The same exercise can basically be done in both Unity and Photoshop.  Anyway,  in it I’ve identified all the intersections and colored them according to their type: yellow for “easy”, orange for “medium”, and [ although there are none in there right now ] red for “hard.”

Camera Coverage + Intersection Analysis

What was being done in parallel can be seen in the image below.  On the left is a first draft basic foundation of how we thought the levels should progress, what new features should be introduced and when.  Also in order for the player to have something familiar to hang on to from level to level, although new concepts are introduced, I made each level an expansion of the previous one.  It all culminates at level 7, then after that when everything has been learned, new levels are introduced unrelated to the previous 7.

Its important to note that these weren’t the final level layouts of course, but a roadmap that we could go by.  There are many questions we couldnt account for until we tested, such as whether too many features were introduced too fast.

Level Progression 1.0

As the RLD is evolving, the level layouts and the introduction of concepts are changing.  Its an organic process that we are still going through now.  Below is a snapshot of that evolution.  On the right you can see me checking the layouts against camera sizes and what percentage of the level’s surface area is covered by however many cameras we “provide” the player.  That in turn will affect how many of each type of intersection we want in there.

Level Progression 1+2 Level Progression 2.0