Pudding Lane: Recreating Seventeenth-Century London
Joe Dempsey, Daniel Hargreaves, Daniel Peacock, Chelsea Lindsey, Dominic Bell, Luc Fontenoy and Heather Williams
In early 2013, Crytek, GameCity, and The British Library teamed up to launch a competition that challenged students to create interactive environments using maps from The British Library and Crytek’s CryENGINE 3, a suite of tools used to develop video games. The six of us, final-year students studying Game Art Design at De Montfort University, Leicester, were invited to take part in the competition, titled Off the Map.
We established Pudding Lane Productions as part of a second year group project that would be entered in the competition. We chose to take part in Off the Map because we knew that it was an opportunity to develop our 3D art and concept design skills, as well as demonstrate how video game technology can be used in a variety of ways.
The aim of the competition was to recreate one of three historical landmarks. The initial choices were seventeenth century London, Stonehenge, and The Pyramids of Giza. We chose to recreate London because we felt that it would be the most interesting option and would provide a challenge.
Our goal was to develop a recreation of the area of London surrounding Pudding Lane in 3D. We chose this specific area because it’s famously where the Great Fire of London started in 1666. London was a large place, even in the seventeenth century. It was overcrowded, dense, and run down. Recreating one street and the surrounding area allowed us to focus our efforts on quality over quantity.
Before we began to create concepts of the different streets that would make up the environment, we wrote up an art guide that we could use to ensure that the style and visual tone of the assets would remain consistent. The atmosphere of the level would be dark and unsettling; it would portray just how unpleasant it would have been to live in London at that time. The colours used would—for the most part—be dark and cold, and we would use particle effects such as fog and smoke to add to the sinister atmosphere.
Seventeenth-century London was cramped and unhygienic; something that needed to be reflected, though the layout of the level would be determined by the maps of London provided by the British Library. However, there were other things we could include that aren’t shown in maps. These things included sewage in the streets, plague-ridden corpses, meat hung outside butcher shops, market stalls, and other assets that would make the area look lived in, as well as add to the claustrophobic, cramped appearance.
The buildings we designed and modeled would need to appear old and structurally unsound, as most of the buildings in London around the time of the seventeenth century had already existed for a long time and were in need of rebuilding. As such, we would focus on making the buildings appear wonky and misshapen; the beams would bend and bow, and the windows and doors would look slanted. Details such as these helped create a realistic, virtual portrayal of seventeenth-century London.
One key issue caused by following the source material so closely was that a lot of seventeenth-century London looked very similar. The best way to add variety to the different areas in our level was to give each individual area its own colour scheme in our art guide. Each area had a separate colour palette to help the different areas stand out and portray a different feeling to the viewer.
Since Pudding Lane would have been full of butcher shops, the sensible colour scheme for that area would be predominantly red. This scheme also helped accentuate Pudding Lane as one of the more significant landmarks in our level, and symbolised the fact that Pudding Lane is widely believed to be the origin of the Great Fire. Dark, muted blues seemed to be an obvious choice for the colour palette of Fish Street. Naturally, there would have been lots of fish mongers in that area, and the fresh fish filling the market stalls as well as the discarded fish scattered across the pathway would reflect a green-bluish hue. The industrial area would have a predominantly brown colour palette, as it would have been full of old wooden warehouses and wharves. In our art guide, the sunset would be visible from the industrial area, however, we were able to block it out to some extent with the use of particle effects. As a result, the colour palette was muted and had a more sombre appearance.
We thought it would be a good idea to give the docks a warm, orange-red colour palette, so that it contrasted with the dank, dingy streets and the ominous blue of the open area. This would give the player a moment of relief from the significantly depressing tone portrayed by the other sections of the level, which was important, as we were not trying to create a map of horrors, but a historically accurate portrayal of seventeenth-century London.
It might seem like an impossible task to create varying colour schemes in an outdoor environment, considering that the main light source is the sky, which, of course, casts predominantly one colour of light. In some areas, such as the streets, we used chimney smoke and other particle effects to block out the sky light, so that we could then add artificial light sources without interference. We also disabled the day/night cycle, as this way we were able to have more control over the lighting cast by the sky. The time of day was late afternoon so that the sun was beginning to set, which was visible from both the docks and the industrial area. In areas in which the horizon would be blocked on all sides by buildings, the visible portion of the sky would cast blue light.
The art guide also included technical specifications that we would use to ensure that the level would run at an optimum level, which ideally would be at least twenty-five frames per second.
Research and Early Concepts
We wanted the level to appear as authentic as possible to the era in which it was set. As such, we took a trip to the British Library in order to collect photographic references from their seventeenth-century London maps. These maps were very useful as they showed images of the city from various angles and in very intricate detail. We spent a great deal of time scanning and surveying the maps and gathering details about the scale, architecture and layout of the area in which we were interested. Some of the maps offered us a new perspective and inspired great ideas from each of us.
One point of discussion was the level of detail we would include of the area. The most sensible solution was to prioritise detail in a sphere of influence starting at Pudding Lane and working outwards. To elaborate, the streets of London were the highest concentration of our efforts, as this area was the most important for telling our interpretation of the history. Lower priority would be the bridge and the tower, for which we would make low-detail versions made to be viewed as interesting features on the horizon.
Using the references gained from our trip to the British Library, we each began to concept the individual streets that would make up the level. We tried to give each street its own visual personality, as we wanted to keep the environment from appearing too monotonous.
Once all of the concepting had been completed, we began to produce whiteboxes for the level. These were basic blockouts that would give us an idea of the scale and overall layout of the environment. We built these whiteboxes on top of the map images we shot at the British Library to ensure that the layout was as historically accurate as possible.
Once we had created a total of three whiteboxes, along with the art guide and concepts, we presented our progress to our tutors. The feedback we were given was that the streets needed to be narrower, as the whiteboxes weren’t achieving the cramped look specified in the art guide. We were also told that we needed more first-hand reference, so we decided to take a trip to York to visit the Shambles.
The trip to those twisting, narrow lanes in the centre of York, with its overhanging timber-framed buildings sheltering the cobbled street below, meant the group began week five with lots of new ideas and inspiration. One of the main points I think we all took away from visiting York was just how tightly packed the streets of seventeenth-century London would have been. At some points on the streets of the Shambles, you could stretch out your arms and touch buildings on opposite sides of the street, and the tops of buildings were only inches from each other.
We were able to collect plenty of photographic reference of old seventeenth-century buildings from which we could determine things such as building materials and the size and crude shapes of the houses.
We also took a trip to the Rutland Museum, because we discovered that they had a set of New Drop Gallows on display, as well as a vast collection of farming equipment from various past centuries. The surrounding village was also full of old buildings and landmarks, which proved to be useful references.
From all of this research, combined with our concepts for the individual streets, we started to build our level. We used one of the bird’s eye perspective maps we gathered from the British Library to make sure the placements for the streets and landmarks were as accurate as possible. We also produced some very primitive textures and 3D models to get a rough idea as to how everything would look.
Design and Production of Pudding Lane
As we moved into production of Pudding Lane, we continued to develop ideas for how to ensure that individual street concepts, including taverns, tailors, butcher shops, fishmongers, and grocers, were unique. This would also mean that the group would again have to look back into some historical references in order maintain historical accuracy. One resource suggested for this was the seventeenth-century diary of Samuel Pepys, in which he describes visiting specific places and buildings, and gives the names of things, such as taverns.
Planning to produce a wide variety of buildings that would make the level varied enough for the user was a challenge, but we devised a way that would allow us to produce a lot of different buildings in a short amount of time. We did this by creating modular pieces that we could arrange to form buildings and other structures, very similar to how a child might build a Lego house. This method allowed us to put together the buildings of the level in a very short amount of time.
After building the base houses, we needed to create a series of texture sets that would be used throughout the buildings and environment. We produced several different versions of each type of material (i.e. different wood grain) to produce an even greater variety of buildings. We also created a range of doors and windows.
Now that we had textures and models to put together we started to import them into Cryengine. Cryengine Software Development Kit is the set of tools we used to place, edit and light our world, as well as add dynamic effect. We experimented with the many parameters for the lighting that Cryengine allows in order to get a more atmospheric feeling for the level.
Our building placement and quantity were becoming finalized by this time, but because we only had the buildings placed in the level, it just looked lifeless, without the hustle and bustle which would be going on in London at the time. This was because we had no props to litter the environment. So we created a large list of props we could place within the level, such as crates, sacks, barrels, carts, and hay bales. These could be used throughout the level to make it seem more lived in, as well as to close off certain streets to create a final perimeter.
The level was growing significantly in scale and in quality, but after we placed all the props and played the level, we felt everything all looked the same and very artificial/man-made. There were multiple suggestions for trying to combat this issue, such as colour grading specific areas of the level and the inclusion of smaller prop assets to help break areas up. We decided to produce some trees and vegetation to break it all up and add some colour to the level. We also decided to model a selection of unique buildings.
A selection of tavern buildings were built to be placed around the level, as well as various unique signs based on actual documented building signs from various sources, including the diary of Samuel Pepys. An important task was building the model for Farriner’s (or Faynor) Bakery, which is located on Pudding Lane. The bakery bears some significance to our project, as it was located at the center of the street from which our level expanded, and from which we took our team’s name, Pudding Lane Productions. In addition, Thomas Farriner was the King’s appointed baker at the time and, most importantly, his Pudding Lane bakery is the location at which the Great Fire of London is said to have started.
The team wanted the bakery to be unique, but not stand out so much that it looked out of place in our level. With that in mind, we designed the bakery using the materials and textures we had already been using. We also wanted the building to have an interesting (although not unimaginable) shape, and took into consideration the inclusion of some important elements such as chimneys and large display windows, as well as where the shop sign would be placed.
Lighting was a major factor in the look and feel of our level. Cryengine has a wonderful daylight system; adjusting the sky-colour gradient, fog levels, and adding pools of warm lamplight helped to making our level look a lot more atmospheric and a lot less flat.
As you can see, it really brings all the individual elements together and sets a very grimy, heavy atmospheric feel to the level. A custom skybox was also added to further push the mood. We also created an epic array of dark clouds and golden light, with an added cityscape to suggest the greater area of London in the distance.
During the final weeks of the project we spent most of our time polishing the level with lighting and texture tweaks. We used a decal system to add dirt, grime, damage, and puddles to the streets to make it feel that much more lived in and to break up the repetition of tiling textures. We created large bricks, small red bricks, rough stone, and roof tiles, as well as rough stone, zig-zag red bricks, 4 types of wood, and plaster. We had to stay in constant contact with one another to make sure that our textures stayed consistent.
Other finishing details included non-tiling textures, such as the doors and windows. We produced 8 different doors, allowing us to have a variety to play with. Along with the doors we produced 4 different windows, with enough room between the windows and doors to create some general textures, such as concrete and metal. We produced a wide variety of hanging sign concepts, mixing several types of metal frames with several wooden signs. This allowed us to mix and match between all of these assets to create various unique signs. To be as authentic as possible, we went through Samuel Pepys’s diary to get a list of actual businesses that existed around this time period. We also created posters that we placed on the sides of buildings to give a greater sense of realism.
Presentation and Results
We presented our level at GameCity in Nottingham. We were all fairly nervous, as the other contenders had produced some very interesting levels. We were also the only team that came in matching shirts, which was slightly embarrassing. It was both a surprise and a relief when we were revealed as the winners of the competition. We all put a lot of effort into the project and it was good to see that it paid off. The after party was also a lot of fun. It was great to be able to talk to industry professionals who gave us lots of valuable advice. We also made some useful contacts.
The overall experience was great. The project was a lot of fun and it taught all us all about working as a team, and on the professional pipeline, and I think that we have all improved vastly in terms of our artistic skills.
Since winning the award last October, we’ve been overwhelmed by the feedback and response from the general public and the mainstream media, something that none of us were expecting when we began working on the project. The team would like to thank everybody for the incredible amount of positive feedback and support we have received over the last few months, and we’re overwhelmed that our project has spread so far and been of interest to such a wide and varied audience. From historians to primary school teachers using our work in their classrooms, we are all extremely happy that our work has made such an impact.
Originally published by Pudding Lane Productions Spring 2013. Revised for Journal of Digital Humanities March 2014.
We’d like to thank the British Library and in particular Tom Harper and Stella Wisdom for the access we were given to their incredible resources at the library, as well as Crytek for the use of CryEngine, which truly is an artist’s tool.