Friday 15 May 2015

Post-Mortem : Off the Map



This project is another strange one. Throughout this project we as a group have been constantly teased about how our group is probably producing one of the best projects.

I personally was constantly uneasy about this as I could see the manic nature in which we were dealing with tasks, especially towards the end.


I personally right from the start of the project was quick to try and separate myself from the rest of the group. I did this by asking for full responsibility over a whole area. This was going to be the canal area. I also was always going to be responsible for the Puzzle room.


I did this so that I could work solely right from concepting to the finished in engine world that I could put straight into a portfolio. Also working like this puts more pressure on myself to be good at all parts of the process, and also to always work consistently to a high standard. This area was always going to be photorealistic. Basically I wanted complete control over an area within which I would not have to compromise on quality.

This area was going to be first person so that all aspects of it in all dimensions would have to look good. Very early on I suggested that I would be willing to make this area a side scroller, as a compromise so that I could sooner finish and help in other areas.

This at the time was deemed unnecessary at this point. I also throughout the project expressed that I would be willing to drop this entire area, I never wanted my decisions to be detrimental to our level. There would always be plenty of areas, props and general assets for me to get my teeth into, even without the canal area.



 I personally really enjoyed working on this area especially when I was starting to get some really nice screen shots of the grass. I then started to get efficient at the foliage workflow, and was past the beginning, slow learning faze. A lot of the foliage was created relatively fast.


At points during this project I felt like I wasn’t working fast enough and was stalling. But this was due to learning new things about complex materials/ shaders. I now have a decent working knowledge of creating complex materials including Tessellation, complex grass wind, macro variation, Decals, multiplying normal, using multiples of all the maps tiled and not tiled, foliage shaders etc.


I also spent a week learning how to create game ready rock assets. So I learnt How to create the high poly in Z-Brush and how to artistically approach rock creation, which brushes to use, decimation, Dyna Mesh, unwrapping in Z-Brush and applying surface detail in Z-Brush.


I also during this project really pushed high to low poly backing, nearly every part of every asset has been detached and had a high poly made, which has then been backed down onto the low poly to create an AO map and a normal. This has been extremely successful especially in simply smoothing out the edges of relatively low poly assets.


I have also used many other processes during this project although I didn’t learn them in this project I have used them. For example a projection mapping pass is often a part of my asset creating, as is poly painting in Z-Brush and using spotlight. 

As with our projects as a group I feel like there was three core issues

Skill set differences

Our group of 6 had 4 character artist, we had no concept artist, and at the start I don’t think anyone actually wanted to really push engine work. So we had an imbalance.  We gave out the character art roles to these people. But this brings me to the next point.

We had a major skill level difference across the group. Half the group were past the learning faze for many processes so could produce work much faster than others. Where some people felt that they needed to spend only 4 weeks on a character or major asset, for others 13 was more appropriate. This basically took some people out of the project completely, instead of being part of the group and producing assets, they would be responsible for one asset. This would then leave our group 2 or 3 people short for the vast majority of the project. Obviously we as a group could have made compromises here. We could have dropped a character, this would free up one person. Or a stronger member of the group could have been given multiple characters, thus leaving some budding character artists character less. I understand that doing this would have been extremely selfish, although this would have resulted in a better project, this would have stunted the personal development of certain members of our group. If you want to be a character artist you are going to want to develop skills in this area regardless on weather this will actually be detrimental to your grade.

Art style differences

Some members of our group would have felt allot more comfortable working much more stylised. The majority of our group wanted to work realistic so this is what we did. This placed some members outside of their comfort zone, and thus may have stunted there enthusiasm and willingness to really push the project.

Bottle neck with engine

For the vast majority of the project two members of the group were working on two separate engine files for different areas. I feel now that we merged these files much too early. This lead to a bottle neck right at the end of the project when we were importing the finale assets. The singular person now responsible for all the world building and re-importing had a near impossible work load. This lead too many assets being forgotten, having the wrong textures applied or no texture applied and generally poor quality world building “propulasion”. The problems here can be fixed extremely easily and I know that it can be done extremely well given time, but for the DMU deadline the above was the case.


Personally during this project I did spend allot of time developing two complex assets that eventually were not used for the DMU hand in. These were the Bridge and the Boat Shed. This means that these assets were actually a waste of time. This obviously would have been detrimental to our level. The effects of this could have been reduced had we dropped the canal area sooner. Sadly we didn’t, it was dropped just before the last week. This occurred because the canal area potentially wasn’t our worst looking area up to this point it’s simply that this wasn’t one of the core game play areas. We had to drop an area in order to finish on time, so the canal the smallest part of the game had to go.

Basically to summarize this project I am going to use this single screen shot from our level.


This is our menu and start screen, yes this is a good image. But I know that it could be allot better. As this is a singular image the grass here could have been denser to hide the planes. The trees going off into the distance should be denser. I created many many different plants during this project and only a small amount of them are in this image. Their is potential here to actually use the boat shed and or the bridge. Also the Gunneras and the Gunnera flowers I created mid project aren't actually in out finale level anywhere. The models that are being used in the current level are unfinished, but they were never replaced in time for the deadline. My finale Gunneras could have been used here at least as they are only present currently in my personal engine file. Sadly none of the above is present in the above image. Although I say these things this is not a personal gripe at any member of out team, we simply ran out of time, given the circumstances we pt our selves in we did the best we could. But that doesn't make the finale screen shots any less frustrating to look through.  


Hope fully all the issues I have listed can be easily resolved for the Off the Map official hand in. 














Stalagmites : Off the Map




I made this in Z-Brush using Z-Spheres to create the basic forms, I then worked in to this to create the high poly. Here I have included the Stalagmites and the Stalactites in the same mesh.

Super Last Minute Texturing


At this point I was just grabbing any assets that were yet to be textured.

Even though these were extremely rushed textures, I tried very hard not to compromise too much. After high to low poly baking nearly all my assets up to this point and seeing the effect of baking a simple chamfer on to a 90% edge it was important to me not to miss this step.


This is the very last asset I textured, right on the last day. I would have liked to have included more material variation, but I was seriously limited by time at this point. Also I did this on my home computer so I had limited axes to Normal map creation software. I only managed to borrow a flat mates computer for about 5 minutes, during this time I created the normal map and AO map for a tillable Hammered steal texture I created.

This was really quite enjoyable, hammered steal made for a nice material study. Also its always fun to see just how far you can push normal maps. 


Oh and I also did the label and cork for this bottles. After I put so much work into all the other textures, I actually felt guilty not high/low poly baking the cork of this bottle, But I was simply out of time. 










Monday 11 May 2015

Last Push Before DMU Hand In


Although I have done tones of work towards the Canal section of our game. In this late stage with the deadline looming over us we have decided to drop this section for now. This means that the Bridge and Shed won’t actually be used. But these will be used for the off the map dead line.



Nearly all the other assets though are being used throughout the forest level.


 So I have gone straight back into finishing the Puzzle Room. Here I have grabbed the roof which I already made, a large sculpture wall thing, the hall way door and wall that Jake made. Now I can get a feel for the room again. Firstly I re-made the wall sculpture thing, I made it reminiscent of Christchurch ceilings.



 To do what I wanted with Jakes door turned out to be nearly impossible, so I modeled my own.



I taught Hannah how to projection map. As a result she created these complex designs, which I used in my own work. I’m really happy with what she created here these were extremely successful.



Here I have added in the floor. One of the pieces of feedback we received was to make the floor pattern draw focus to the Glass table. So that’s what I have done here. Originally this floor pattern had circular rings converging on the centre but after I started texturing the floor I decided this was going to take too much time. This was due to the floor texture being a single large texture, it isn’t a tillable. I was individually selecting tiles and filling them in in Photoshop, the rings had far too many tiles in them for this to be a feasible design.



 I did a planet




This is the finished Puzzle Room. I am really pleased with this.