Monday 29 July 2013

29th July 2013

I have been working on various elements of the project over the past few days. Last week I processed some of the impulse that were recorded during the site visit to Spike in the various areas of interest. The preroll silence was trimmed from the recordings and then they were deconvolved in Apple's Impulse Response Utility to remove the sweeping sweeping signal from the recorded tracks. What then remains is the acoustic signature of the space. I was then able to export this and use it in Logic as a reverb plugin and apply this acoustic signature to any sound I wanted.

I have also spent a good bit of time working on my landscape. This is very much a work in progress at the moment. I began by trying to use the CAD contour drawing to create my terrain. I imported the contour drawing into 3D studio Max and converted it to an editable spline which I then converted to a terrain. This however, result in a very 'messy' mesh which was very difficult to extrude from. I also used the terresculptor plugin for 3D studio max which produced a similar result as below.



Continuing from this I am now using the Unreal Landscape editor to import a heightmap of the Island which I created in Photoshop using the contour lines from the AutoCAD drawing that I have of the Island. I assigned a different shade of grey to a different level on the contour map which resulted in the heightmap below.


The black area represents the lowest point (sea) and the lightest shade of grey represents the highest point of the Island (the fort). When the height map is imported in UDK the result is a lot better than trying to create the terrain mesh in 3D studio max. I plan to do some more work on the landscape in UDK tomorrow as I am still struggling with the scale of the Island as there does not seem to be a way to enter real dimensions into UDK to define the size of the landscape.


Thursday 25 July 2013

24th July 2013


I have had a lot of trouble with UDK and importing my meshes from 3D studio max into the UDK platform. I created my mesh and textured it in 3D studio max. I then exported the mesh as a .FBX file. I opened UDK and went to import the mesh into my assets browser, however, every time I went to import the file, UDK crashed and I had to force quite the program. I tried a lot of things to get the file right. Firstly I selected all my vertices and welded them, just to make sure my mesh was contiguous. I then added a normal modifier and unified normals to correct any vertex alignment issues. I reset the xform and then re-exported my file. UDK still crashed. I then removed all my textures and applied a standard material to my mesh, exported the file and imported it into UDK. Finally UDK did not crash and my model was loaded into my assets. I will need to figure out why the texture was causing so many problems as this could prove to be a serious issue as it would severely slow down the whole modeling process.

Once I had imported my mesh into UDK I then was able to place it in a world to see what the model looked like. Here a second issue arose and that was to do with scale. When I added the mesh into the world it was quite small and not to scale relative to the user. I tried many different unit setups in 3D studio max to rectify this. The one that seems to work now is to customize my unit set up to 1uu(unreal unit) = 2.56 cm. Then in my system set up, set 1 unit as 2.56. Then when I import my model to UDK the model appears to scale.

Tuesday 23 July 2013

23rd July 2013

Today I spent most of the day continuing on with the models. I managed to fix the issues I had with the shadows yesterday however I made only a little progress as some of the UVW unwrapping caused some difficulty which really slowed me down. I also did some research in the evening.

Monday 22 July 2013

22nd July 2013

I have made a start at accurately modelling the buildings within the Fort. My plan is to start with the buildings that I intend to model externally only and get these done in a short amount of time, leaving a bit more time for the buildings of interest.

The first model that I have concentrated on is known as the 'B' block and is one of the more recently refurbished buildings within the fort.



I have not yet applied any materials to the model but I have more accurately modelled the facade which already makes the buildings appear to have more depth and interest than the previous sketch models that I had experimented with (as seen in previous post). 


One issue that appears to be arising at the moment is the unusual shadowing that seems to be cast under the over hang of the roof. It should be an evenly distributed shadow however, it appears to be broken and unnatural. This will need to be sorted.



This morning I also applied some texture maps to the sketch model that I had been working on. They all seemed to enhance the appearance of the texture and make the model appear more realistic. However, the displacement map caused the mesh to become distorted when this map was applied.

Tomorrow I plan on working on the acoustic information that was gathered from the visit to Spike and to get some more of the buildings modelled.

Saturday 20 July 2013

19th July 2013

I made a rough sketch model of one of the buildings yesterday. The modelling of the building was very simple as it consisted of the same bay mirrored twelve times. Many of the buildings with in the fort are very repetitive like this. One of the most important aspects of the modelling is going to be applying the materials and textures, along with good lighting.

At the moment the sketch model is very basic. I created a basic wireframe model and then applied an Unwrap UVW modifier to the mesh. 



I created two separate UV maps for the roof and the front facade. I rendered these UV templates as .png files, saved them and reopened them in photoshop. I then was able to apply different materials to the various elements of the facade.



I then applied new bitmap materials to my model and used the .png files that I created. The result was very flat and unrealistic when I rendered my views.



This is due to a few different factors. At the moment I do not have way to create good quality bump maps and normal maps. These can be created in a number of different ways. I will need to test a few different techniques to see what is the fastest and most economical technique available which also produces good results. There are numerous plugins available for use with photoshop and also separate applications which create normal maps. One application that I have tried with a UV texture map that I created for an earlier sketch model is CrazyBump. 


I have not yet applied this normal map to the model so it will be interesting to see how it effects the appearance of the model. This application also creates other types of maps such as occlusion maps etc. to enhance the appearance of the textures on the models. 

The textures that I have used for this UV map were also not of very high quality so this is definitely something that would need to be addressed before creating the final models. Also the lighting for the models will be essential to create good quality and realistic models.

18th July 2013

Took a second trip to Spike Island on Thursday with the aim of gathering more detailed first hand information and data on the buildings within the fort. As well as this another important goal of the trip was to decide which would be the main areas of interest and therefore the areas that would be modelled in most detail and with most accuracy. This resulted in us focusing on three main spaces which were:

1. A Prison cell in the former NW casemated barracks - 'C' class cells
These cells are within a single-storey fifteen bay casemated building. The building is constructed of large coursed ashlar limestone blocks. Casemates (i.e. vaulted chambers) are brick vaulted and lined and all originally extended the full width of the building. Most areas inside this building now have modern internal finishes with cement floors and cementitious render to walls and ceilings. 



2. Former Prison Cell Block - Solitary confinement
These prison cells were much more oppressive and dark than the first block. The building was also constructed of ashlar limestone blocks. The building consists of a two-storey main block and a single-storey block extending north from the north-west corner. The interior is brick lined which had been plastered but has since been removed. There is a timber gallery which is supported on decorative iron brackets along cells to the first floor which is accessed by way of limestone steps with an iron handrail. This is reputedly the building where John Mitchell, the Irish nationalist activist and political journalist was incarcerated.



3. Bastion and Ramparts
The bastions were generally geometrically designed to defend unrestricted views of the moat along flanking walls. Each bastion has a flanking gallery with slender window openings to offer the best vantage points for defending the moat. The gun is in situ with associated stores of vaulted brick contraction with later concrete tunnel construction by the Irish Army providing access from the parade Ground via the casemate stores to the gun position.



I aim to model these three spaces in detail and then to model the rest of the buildings in the fort as external models only, with some of these structures being ruins as can be seen in the 1915 Building below.



We also performed some audio recordings within the three significant spaces in order to be able to replicate the aural properties of the spaces.


Wednesday 17 July 2013

17th July 2013

Today I spent most of the day in the VR lab trying to make a start on some of the models of the buildings in the Fort. Tomorrow I will travel to Spike for a second, and more detailed, visit. Hopefully I will have all of the information on the Island available to me by the end of this week. 

Thursday 11 July 2013

11th July 2013

Today I took my first trip to Spike Island. The trip was brief, as I didn't get a lot of time to freely explore the Island, so it was more of an initial introduction and overview of the Island. I will definitely need a more in depth trip to get a better feel for the spaces within the Island and buildings within the Fort. The weather today painted the Island in a beautiful light so I was not presented with the desolate and oppressive "Spike Island" that I was expecting. Instead, I landed in a place that seemed so far removed from a prison, with the most brilliant panoramic views all around Cork Harbour.

The fort itself is a lot bigger than what I had pictured in my mind, with a lot of open space inside the fort walls. The state of the buildings covers a broad spectrum, with some of the buildings being newly renovated.....




 and some being shells of what they once were...


I did not get to visit any of the prison cells which I am very keen to see, so this point alone warrants another visit soon.... I have spoken with Tom, the man who everyone says knows most about the Island, and provisionally arranged a guided visit by him for next Thursday.

From a modelling point of view, many of the buildings within the fort seem quite repetitive in their structure and detail (facades). I could perhaps model one bay of the building and mirror this down along the length of the structure as in many cases each bay contains the same pattern of window and door openings. I will need to take some detailed photos of the building materials to get a true representation of the building textures on the model. 


As of now, I plan to create a model of the Island and the Fort. I will model most of the buildings within the fort externally and I will also internally model in detail one of the prison cells, focusing on realism and trying to give the user a sense of the space within the cell.

I also spent some time today researching the (technical) concepts behind game design and the different approaches involved in serious game design. I need to sketch out and plan how my "game" will work in a more definite way as I feel it is all too fuzzy at the moment.

Wednesday 10 July 2013

10th July 2013

Most of today was spent working between 3D studio max and UDK. I created a simple static mesh with materials and textures and exported this as a .fbx file. I then imported this into UDK in a package and I was able to drag the mesh from the assets menu into the world space. The mesh came with materials and textures intact. I was able to preview this in the game window and I was also able to add collision properties to the mesh so the user collided with the surface of the mesh and didn't travel through it. 




I also spent some of the day working on various types of texturing techniques in 3d studio max and how to implement these. I spent the morning researching some of the concepts that I had begun to research yesterday with regards to the technical aspects of designing a story and also some of the psychological aspects of game design.


Tuesday 9 July 2013

9th July 2013

Today I spent most of the day in the VR lab working between 3D studio max and UDK. I managed to make a textured mesh as export it from 3Ds max as a .fbx file. However, while I was able to import the mesh into UDK as a package and view the asset in real time with its textures and materials intact, I was unable to get the mesh to display in my world space when I dragged it into the viewer. I will work on this tomorrow as it is very important to get these things right and iron out any wrinkles before I begin more detailed modelling.

I also spent some time this evening researching the effect on emotion in interactive environments and some of the technical aspects of story design. I hope to do some more research into these areas tomorrow. Hopefully I will get to Spike Island for an initial site visit later on this week and as I would like to get a first hand perspective of the Island itself.  

Monday 8 July 2013

8th July 2013

Most of today was spent in the VR lab working my modelling skills and the best way to work between the different platforms. I also spent some of the day chasing leads to acquire some data and information to get me going on the 3D models.

Friday 5 July 2013

5th July 2013

Today I spent half of the day in the VR lab working on the computers. I also did some analysis on the different platforms which have potential for use in the project. 

I spent a few hours researching the concept of "serious games". A serious game is a game designed for a primary purpose other than pure entertainment. When discussing the subject of serious games, the term "game" itself can become ambiguous and you can really begin to question what a game really is? In my mind, I always thought of a game as something which always has a definite and eventual ending. 

However, in the context of serious games for use with cultural heritage, there is the potential for a user to explore a "virtual" environment with no real aim or end goal in mind, other than exploring and discovering the virtual environment itself...

Wednesday 3 July 2013

3rd July 2013

Today I spent the first part of the day researching, specifically in relation to interoperability between platforms and file formats. I also spent some time in the VR lab working between the various platforms. 

By the end of this week I aim to have two main items tied down. Firstly I need to finalise on the platform(s) to be used for the implementation of the project. Secondly I hope to have enough data sourced and gathered in order to start "building" the project by early next week as this will take a lot of time. I also hope to visit Spike Island next week to carry out my own analysis of the Island and gather some first hand data.

Tuesday 2 July 2013

2nd July 2013


Today I spent the first half the day researching and the second half of the day in the VR lab working on computer skills. I have also made contact with some people involved in different areas of research related to Spike Island. They have come back with some interesting data that will hopefully be of use to the project.

Monday 1 July 2013

1st July 2013

Today I spent most of the day researching. I found an interesting project on Clonmacnoise which is an immersive 3d model of the ancient town of Clonmacnoise. It not only uses touch inputs but the app can also be used on site using GPS input which is used to control the user’s position automatically. When using this functionality, the user does not touch the screen to move around in the environment, but as they walk about their position in the 3D world is constantly updated to reflect their position in the real world. The user can then see how the area that they are in looked 1000 years ago.


I would like to develop an accurate 3D model of spike island, that the user is free to interact with and explore the model. The user would land on the Island and be greeted by an animation which would bring them through the fort gate. They could then freely roam about the Island. There would then be 'hotspots' around the immediate location of the fort which would encourage further interaction, possibly through the use of other media elements such as audio and video.