top of page
Writer's pictureJoe McLoughlin

Creating a Historically Accurate Recreation of 1547 Hampton Court Palace in Unity

Updated: Aug 17, 2023

Written by Joe McLoughlin

Introduction

This is an excerpt from my dissertation thesis regarding the methods used to recreate historical settings, following research of how developers go about the same challenge. Sometimes faithful recreations aren't indicative of quality gameplay, so it's always a dilemma on how to remain accurate, yet not hinder game enjoyment. This article covers the methods used to recreate 1547 Hampton Court Palace and the considerations made to tackle this issue.


Building the Palace in Unity

...Whilst AAA games can take years recreating historical landmarks, involving hundreds of artists and technology such as photogrammetry, this project is far more limited in scope. Despite this, inspiration for the methods used in other games can be utilised. A very well-known example is Assassin’s Creed Unity’s (2014) incredible recreation of Notre Dame. Senior level artist, Caroline Miousse, worked with a historian and researched the look of every individual section of the cathedral during the French Revolution, making it as faithfully as possible [21]. Miousse also notes that Notre Dame wasn’t built as a gaming space, so some liberties had to be taken with item placement to create paths [21]. This was less of an issue for this project, as it wouldn’t have the climbing aspect of Assassin’s Creed. Nevertheless, this does highlight the fact that the main two areas of study for this project were likely to have times of incompatibility with one another. Whilst Ubisoft would favour gameplay, the aim of this project is, first and foremost, authentic reproduction.

This project was ambitious in scope. Time management was key to seeing the completion of all desired aspects. Long term and short- term goals were needed to be set and priorities were made. As first enunciated by Northcote Parkinson in 1955, "Work expands to fill the time available for its completion” [22]. Following this, early deadlines were set to improve efficiency and allowed extra time for troubleshooting.

Carrying out preliminary searches on the asset store, it was clear that period accurate assets were unlikely to be found. There were very few, if any, Tudor NPC skins and only very generic ‘medieval’ items and furniture. Whilst some simple models could be made in Blender, time and ability would limit this solution. Some research was done into the software RealityScan, a free app which creates 3D models using scans from a smartphone, a primitive version of photogrammetry technology used by games companies. This would have been an effective solution and perhaps an area to explore for future development but for now, generic NPCs and props from the asset store are to be used to offer the best representation possible in the timescale available.


The procedure decided upon to recreate Hampton Court in Unity was firstly to produce the footprint or foundations of the palace to scale, then use this to build upwards and develop the second-floor footprint, creating the shell of the palace, before finally moving onto the interiors to construct a realistic appearance.


Figure 7a: Shows the complete outer wall footprint.

The first step used the diagrams shown in figure 4 along with a tape measure and a ratio calculator to create a scale model at a ratio of 1.1cm equaling 10m. This was a time-consuming task, taking longer than originally expected. However, it was beneficial in the long run as it avoided any misjudgments of scale and size that would have required time to fix. It also added to the realism and accuracy, as the player would ultimately experience a true to life feeling of its magnitude.


Figure 7b: The footprint, including some first-floor and stairs additions.

To start with, the entire outermost wall was drawn out with simple cubes representing walls. Once this was complete, interior walls were added, creating rooms and courtyards The first floor was much quicker to add, as the footprint of the rooms and walls on the first floor often matched that of the ground. Thus, a copy and paste approach could be adopted.

Figure 7 shows the progression of the footprint over its lifetime. Once the shell was completed, windows were added. These are visible on the diagrams in figure 4, so using this, along with videos and images of modern-day Hampton court and contemporary drawings such as that shown in figure 8, it was possible to produce historically accurate window locations, sizes and appearances. The windows were self-built as prefabs within Unity, using elements of a model from Sketchfab. This allowed for the relatively quick creation of a variety of window sizes and styles, including stained-glass. The sheer number of windows in the palace was underestimated and, again, this took longer to complete than expected. However, it was at this point that it began to start looking like a real building. Figure 9 shows some early window additions on the east front based on the drawing in figure 10.

It was at this point that it was realised that the reconstruction needed the external elements of the second floor, or risk appearing incomplete and damaging the player’s immersion when viewing from an outdoor perspective. However, other than taking time to implement and the limited nature of historical sources for the second floor, it was not too problematic an addition and benefitted the project in the long run.

Building the interiors of the palace was the next step. This would rely heavily on the appropriateness and prevalence of suitable models and textures and time was spent on Sketchfab, the Unity Asset Store and Free3D finding them. When suitable models could not be found, they were created by hand. A major example of this was the reconstruction of the Great Hall ceiling, seen in figure 16. No models came close to the suitability required for this and as it was such a vital component of the Great Hall, the time taken to build it was a necessary sacrifice. On a smaller scale, models such as the fireplaces and spiral staircases were also self-built but, again, their prevalence and importance to the palace made this necessary.

Focus for interiors was first put on the Great Kitchens. This was chosen as the starting point due to its mostly unchanged present-day appearance. The manner in which it is presented for visitors to Hampton Court today is that of its Tudor appearance, based upon work done by the curators. It is therefore an accurate depiction of the kitchens of the sixteenth century and thus this project could largely imitate it, saving time on research. The beer cellar, Great Watching Chamber and Great Hall followed.

Building Hampton Court Palace to the standards demanded in Unity was not a small task. Whilst the initial plan was to recreate the entire ground and first floor, this was later truncated to a section of the palace upon realization of the ambitiousness of this target. It was decided that this smaller game area would cover the Great Gatehouse, Base Court (including two accessible double-room lodgings), the beer cellar in the undercroft, the Great Hall, the King’s state apartments up to the Privy Chamber and a part of the Great Kitchens.


Figure 8: A view of the no longer existing east front of Hampton Court, drawn by Hendrick Danckerts in 1665-7. Paintings like these were heavily relied on for external recreations.

These areas were selected for a few reasons. The first being their locations, these were all areas close to or accessible from the main gatehouse, where it was decided the player would start, which made sense from a logical and gameplay point of view. The second was that many of these rooms are still in existence and thus saved time on the research required for their historical recreations. Whilst their modern-day appearances do not exactly match how they were in 1547, it was still a major help drawing on the existing rooms. The rooms that do not exist in any form of their original appearance required more effort to reproduce.

Figure 9: An early view of the in-game east front, showing the addition of windows based on the drawing of the old palace in 1680.

However, a major motivation for this project was to allow the player to experience areas of the palace that no longer exist and because of this, it was important to see this through, even at reduced scale. This final game area offered a good balance between extant and extinct rooms that would fulfil motivations but also aid in time saving.

The general order of precedence when reproducing a room that no longer exists was firstly to use drawings and written sources that specifically referred to that room in Hampton Court. Should this not be available, the next tactic employed was to use evidence of interiors of the same room, but from different palaces of the time. This was the method used for the recreation of the presence chamber (see figure 10). Again, failing the availability of this, non-specific evidence was used, perhaps descriptions of royal apartments of the period in general. Finally, any loose evidence at all was used in conjunction with an appropriate appearance. This was the case for the gallery outside of the player’s lodgings.


Figure 10: The recreated in-game presence chamber. As this room no longer exists, a combination of surviving evidence had to be used, described in 5.1.1.

Having settled on a smaller game area, focus could concentrate on producing high quality reproductions. Referring to historical sources including paintings, drawings and written descriptions, described in 4.2.3, along with the catalogue of models and textures that had been found or created, all interiors were finished to a very adequate level. Once the interiors were finished, improvements were made to exteriors that would be visible to the player. This included the roofs, vaults, crenelations, smoking chimneys and fires and details that would finalise the game environment. To define the playable area, walkways and doors that would lead the player into incomplete regions of the palace were blocked off. For future development, the blocks remain fully functional. The UI displays the correct name for the rooms into which the doors lead, should the player approach or interact with them, and they are also fully unlockable for when the corresponding keys are added to the game.


Figure 13: A comparison between the in-game Great Hall and the real-life Great Hall. Notice differences such as the colours of the ceiling and floor, which correspond to the evidence backed original appearance.

Figure 14: A comparison between the in-game Great Watching Chamber and a drawing made suggesting the possible appearance it had in the 1500s.

Successes and Shortfalls

A primary motive behind this project was to provide an enjoyable platform on which players could experience the original appearance of Hampton Court. Attention to detail and

accuracy was therefore of utmost importance if it was to successfully achieve this goal. As a reconstruction with considerable evidence-backing, the end product of this project was both pleasing to the eye and historically accurate.


Figure 15: Queen Elizabeth Receiving Dutch Ambassadors, c.1585. The walls, floor, seat of authority and bird cages were referenced from this painting for the presence chamber reconstruction, seen in figure 12.

Successes

Almost every design choice made when developing Hampton Court in Unity was supported by evidence. Very little conjecture was required, and gameplay also had negligible impact on accuracy.


Figure 16: 'The Family of Henry VIII', c.1545. The columns, frieze and throne ensemble referred to this painting which is set in the presence chamber at Whitehall Palace.

As explained in 4.2.2 many of the rooms reconstructed are still in existence today and this greatly aided in accurate depictions. Whilst even these rooms have changed considerably since 1547, they offered a glimpse of how they would have appeared, which stood as an advantageous starting point. From this the use of drawings and written sources provided excellent material for the recreations. Evaluation for this field is based on the comparisons between the recreated environments in the game and the paintings, drawings and writings that were applied. The well-achieved outcomes can be seen clearly in figures 13 and 14. These were all locations with direct, room-specific backing. When considering the success of non-existent rooms with little to no direct evidence, analysis must be taken into smaller details within the room. The presence chamber, for example, was created using a combination of two Tudor period paintings: ‘The Family of Henry VIII’, c.1545 and ‘Queen Elizabeth receiving Dutch Emissaries’, c.1585. Despite the latter having been painted forty years after the setting of this project, Simon Thurley argued its case as a good representation of a Henrician interior, as only the windows are of an Elizabethan fashion [28], see figure 15 and 16.


Figure 17: The in-game Privy closet, which took inspiration from three sources of evidence including the painting in figure 20.

A final example of successful reconstruction is for that of the privy closet. Two paintings were again used, alongside extra descriptions and an extant closet in Hampton Court. These paintings were ‘Margaret Beaufort in her Privy Closet’, C16th and ‘Henry VIII in his private prayer closet’, C16th. As can be seen in the comparison between figure 17 and 18, there are no elements present that do not have support from either the painting or the existing architecture. To accentuate the detail and care with which the palace was recreated focus is placed on the meticulous locating of windows. Whilst copying the current appearance would have been easier, extra research was done to ensure true 1547-specific window locations and sizes. Many of the windows have been replaced since the Tudor period and so care had to be taken not to fall into the trap of copying the modern-day appearance. These figures offer an example of the accuracies to which this reconstruction adhered and as such can be seen as an overall great success for the purposes of this project.


Figure 18: Henry VIII in his private prayer closet, C15th. The canopy, leaning desk, book and altar are followed from this painting.

Shortfalls The full extent of the palace was not recreated; however, this was a small sacrifice to ensure high quality reproduction of the areas that were reproduced. The primary shortfalls of this field were the limited availability of period-appropriate models and textures. A lot of time was spent finding them and, in the end, often generic ‘medieval’ models were used. This issue was restricted to use only for timeless objects such as plates, tables and chairs. Other than this, the lack of evidence for every room was the only limiting factor and some areas did require a ‘best guess’ solution, using patched together evidence, such as the Base Court gallery, mentioned in 4.2.2. This was always going to be an issue for a historical piece of work due to the natural loss of resources over time.



References

[21] Brett Makedonski. 2019. One dev spent two years making the Notre Dame in Assassin’s Creed Unity. Retrieved 28 June 2022 from https://www.destructoid.com/one-dev-spent-two-years- making-the-notre-dame-in-assassins-creed-unity


[22] Jenny Steward. 2015. The persistence of Parkinson’s law. Quadrant, 59(9), 62–65. https://search.informit.org/doi/10.3316/informit.442611729683014


[23] Daphne Ford and Simon Thurley. 2003. Hampton Court: a social and architectural history. New Haven: Published for The Paul Mellon Centre For Studies In British Art, Yale University Press.


[24] Historic Royal Palaces Blog. 2018. Inspirational Women Through History. Retrieved 27 June 2022 from http://blog.hrp.org.uk/international-womens-day.

[25] Comp-3 Interactive. Active sine 2019. Videos. Retrieved 23 May 2022 from http://www.youtube.com/c/COMP3Interactive/featured.

[26]Sha Zhao et al. 2021. Player Behavior Modeling for Enhancing Role-Playing Game Engagement. in IEEE Transactions on Computational Social Systems. vol. 8. no. 2. pp. 464-474. doi: 10.1109/TCSS.2021.3052261.

[27] Benefits of JSON. 2017. https://docs.progress.com/bundle/ openedge-abl-use-json-117/page/Benefits-of-JSON.html. Retrieved 26 August 2022.

[28] Simon Thurley. 1993. The Royal Palaces of Tudor England: Architecture and Court Life 1460-1547. New Haven: Yale University Press


bottom of page