Unit Introduction
In this unit we are being asked to produce a model pillar with no more than 8000 triangles for the low poly bake, in a effort to help us improve our skills moving forward from Unit 9, going by the criteria set out for the splash damage brief.
Requirements for the unit are:
- Produce a high poly mesh, low
- Low poly model mesh, no more than 8000 triangles
- Good knowledge and understanding of how to set out UV's, also texture pages.
- Diffuse, specular, normal and ambient occlusion texture maps
- Second set of UV's for light mapping purposes
- Screen shot showing the model and mesh topology for the low poly model
- All textures used in the scene, (Diffuse, normal and specular maps)
- The models that make up the scene (Preferably in Maya)
- Render of the model, (Preferred if presentation of model within UDK fully textured.)
As previously, the every stage of the development and process will be blogged.
Time Management
For the brief, we have been asked to set out a timetable to help us manage our time effectively through this unit, in helping us being more organised and achieving a better quality model and work, thus hoping to achieve a higher grade. Effective and efficient time management is essential for success.
With 15 hours a week spent in class time, 15 hours a week should be spent outside of class continuing work, as well as another 8-10 hours spent doing research and homework. In total 38-40 hours a week in trying to complete this Unit to a high/ decent standard. On the whole, requirement for a unit is 150 hours.
It is essential to make sure time has been allocated correctly, to cover things that might not go according to plan, or that working on some elements may take longer than others, hence the reason for allocating the right amount of time when needed to work when possible.
Started Monday 23/11/15.
It is essential to make sure time has been allocated correctly, to cover things that might not go according to plan, or that working on some elements may take longer than others, hence the reason for allocating the right amount of time when needed to work when possible.
Started Monday 23/11/15.
High & Low Poly Modelling
High and low polygon models are used in varying ways from each other. From research, low poly models are predominately used in in computer games, where as high poly models are used in films, advertising, and still images.
The reason high poly models are used in the likes of films, advertising and still images is because a high poly model produces a lot of detail. Thus making high poly models ideal, as they are very well detailed, showing off good quality by appearing more realistic with their smoother edges, which is great for audiences watching a film or an advert etc. The company selling that said product is showing how great their product is. Sadly with high poly models, the draw back is that being this detailed, the file that produces the model is large, and only a certain amount of movement can be created with the model, before it becomes unrealistic defeating the object of the model being that well detailed and appearing realistic. At present, sadly high models could not be used in computer games as the files would be too large an would cause the models actions to be too slow. Although the model would look better and more realistic, its slow and unrealistic movement would undermine the realistic appearances and would not go down well with gamers.
Low poly models are primarily used in computer games, due to them as stated above, being ideal or a happy medium, in that a decent level of graphics and detail can be created, whilst giving enough movement and animation for the player to enjoy the game and the model to move at a fast pace without any lagging. Downside to the low poly as stated, as that the model is not as well detailed, or as realistic looking as if it were a high poly.
The evidence can be seen from the screen shot above of my own character I created for Unit 2, with the high poly character on the right and the low poly character on the left. Although the differences are obvious, the low poly character still has a fairly decent level of detail to satisfy gamers needs, whilst producing a small enough file to be able to function to a satisfactory level in a computer game.
High and low polygon models are used in varying ways from each other. From research, low poly models are predominately used in in computer games, where as high poly models are used in films, advertising, and still images.
The reason high poly models are used in the likes of films, advertising and still images is because a high poly model produces a lot of detail. Thus making high poly models ideal, as they are very well detailed, showing off good quality by appearing more realistic with their smoother edges, which is great for audiences watching a film or an advert etc. The company selling that said product is showing how great their product is. Sadly with high poly models, the draw back is that being this detailed, the file that produces the model is large, and only a certain amount of movement can be created with the model, before it becomes unrealistic defeating the object of the model being that well detailed and appearing realistic. At present, sadly high models could not be used in computer games as the files would be too large an would cause the models actions to be too slow. Although the model would look better and more realistic, its slow and unrealistic movement would undermine the realistic appearances and would not go down well with gamers.
Low poly models are primarily used in computer games, due to them as stated above, being ideal or a happy medium, in that a decent level of graphics and detail can be created, whilst giving enough movement and animation for the player to enjoy the game and the model to move at a fast pace without any lagging. Downside to the low poly as stated, as that the model is not as well detailed, or as realistic looking as if it were a high poly.
The evidence can be seen from the screen shot above of my own character I created for Unit 2, with the high poly character on the right and the low poly character on the left. Although the differences are obvious, the low poly character still has a fairly decent level of detail to satisfy gamers needs, whilst producing a small enough file to be able to function to a satisfactory level in a computer game.
Normal Map Bake
Normal map baking is when you take a high poly model and bake the detailed information from the high poly on to the low poly model, which is pretty much a empty basic shell. For the bake to work, the models need to be at the centre of the world, using the centre pivot icon or dragging the the models pressing the X button into the centre.
Construction History
Construction history is what it says on the tin. Gives the user the information about the history of the models construction.
Holes
During the baking process, there can not be any holes in the model likes the ones shown below in the picture. This can cause problems during the baking process.
Grids
Grids can be used as part of the baking process to press patterns against the surface of the model.
Wires
Another possibility for the baking is using wires. Giving the model a different prospective in its design.
Kit Bashing
Looking at more part to play with in terms of kit bashing as accessories to use to add on to the model in the baking process. This can bring in new elements to the design process by bring extra features.
Projection
Projection meant that when kit bashing you can not bake at an 90 degree angle. Creating shadows behind the extrusion when creating your own kit bashing pieces, so that when you come to bake, it looks good and how it should. To be honest this has caused me some problems initially.
Detail Scale
Silhouette has to be the same from model to model. Meaning the models have to be the same size. Otherwise again this can cause problems during the bake. Probably one of the few things I did not have an issue with this time.
Exaggerating shapes
This involves exaggerating the parts on the low poly to make it look similar to its higher poly counterpart.
Curves
Brakes
With these elements, it has been a process of learning what to do and what not to do in terms of the modelling, Uv-ing baking process's. What can go wrong and how to correct any issues that may occur. How to get models to work in a sense that no major problems occur during these process's and that they can be corrected. Also recognising possible problems that occur during these modelling process's.
Panel Joints
Panel joints, in giving more options again for kit bashing.
Get Bent
The get bent tool makes it possible to bend surfaces appropriately to how you want them to be. Pipe work is the first and obvious thing that springs to mind, but there are many more possibilities. From research it looks like it would be possible to bend more tougher looking surfaces like the for example panel joints above. Creating what you require it do for your project.
Bevel tool
The bevel tool to smoothed out edges by adding edge loops near the edges of the model. Usually one loop of each side of the edge but getting loops as close as possible, making the model look really smooth. By not bevelling before the bake process, this can also lead to problems after the bake, and not having the desired effect on the model.
Boon leans
The boon lean tool makes it possible to to merge two objects together by merging the geometry together, rather than per say extruding a shape to then create another shape from the faces or edges.
Design process - Mind Maps
After unit 9, I decided I wanted to try and do it a bit more with my pillar. So I began putting ideas together for a new pillar. Initially just building mind maps with ideas, then moving on to Pinterest and then sketches.
Pinterest Mood Board
This time I decided to categorise my mood boards with each board having its own theme, trying again to get as many ideas as possible together and then narrowing down what I would use on my pillar etc.
Sci-fi Pillar
With this mood board being science fiction, looking at different types of technology, sci-fi films etc that might be of useful inspiration.
Historical Mood board
This mood board I was looking at possible historical and religious influences.
Industrial mood board
https://uk.pinterest.com/stewartmarr/pillar-3-industrial/
This mood board, looking industrial influences and what sort of thing I might want to add to my pillar.
This mood board, looking industrial influences and what sort of thing I might want to add to my pillar.
Kitbash mood board
I made a kit bash board just to gather more ideas and possibilities together on what the pillar could look like.
Surfaces mood board
With the surface mood board, I began looking at possible textures at what I could cover the pillar with. Clean, reflective, battered, rusty and so on.
Nature Mood board
With the nature mood board I was trying to gather ideas on the battered worn look and of course the overgrown look.
Ideally, I was wanting to merge somehow all the ideas into one. Even if it was just small elements from each mood board. Technology meets religion, covered in a good surface, whilst over grown maybe battered by nature.
Sketches
Below are some sketches I have done on various ideas from inspiration I had gathered together in and from my Pinterest Mood Boards. I wanted to use some sort of Religious art mixed in with sci-fi technology or even heavy industry, making the pillar look like it had some sort of alternative purpose.
Pillar
I started off with a simple shape, and just extruded the top of the box to give to give edges, then copied the pillar, so I had ready a low poly and a high poly to work on.
After my research with mood boards on possible designs, I took a sun image from Google and put into photoshop, removing the outer layers so that a basic silhouette remained to use on the pillar. Then I placed it on the what would be high poly pillar.
Being inspired by the kit bashing ideas, I began develop how I could use my own kit bash ideas with what I have learnt in Maya so far and what I wanted to achieve from the mood boards. Panels and bolts were something straight away that I thought might look good.
The next phase was to begin Uv-ing of the pillars. This required going into the Uv texture menu and lining up all the parts of the pillar to make sure the model would bake properly. By giving the it a chequered pattern (for example), it helps with lining all the parts up, making sure all the small boxes in the pattern are fitting inside and are boxes so everything can work as it should.
With the bake, highlighted problems with my original pillar ideas. Panels and bolts were not baking properly on the low poly. Also, the sun image I had wanted to use on the pillar was also causing problems with the bake. So basically design was all round failing epically. The panels, bolts and vents, also the sun image were causing too many shadows and causing the the after effect on the low poly bake to look really smudged. Not really what I had in mind. With this I tried to simplify things.
From here I tried a more simple approach, by using less panels and not having the panels I had, be smaller, with the vents an bolts extruded to a more suitable angle facing outwards from the rear. Sadly the sun image went. I really would have liked to have kept more to the original design with more panels and the sun image but it caused too many problems after the bake process. So it became quite disappointing, but I've tried to stay to a similar uniform to what I had to tried to build earlier.
Light map. For using when the fbx file is imported into udk. This needs to be spaced out to prevent shadows on the model.
Diffuse map (left) This is an altered normal map, created/ altered in Photoshop.
Normal (below) This is produced automatically after the bake.
After the bake, I went into Photoshop and began develop the diffuse map by painting panels etc, trying to give the low poly model some details. My skill in Photoshop are as limited as Maya, but I was able to add some colour to the panels and bolts for effect, even if it was limited. By the time I got to this point, imagination had gone out the window, hence the plain grey colours. It could have been a opportunity to add more detail to the pillar at this point. Obviously to say, I missed that train.
Although the work is not to the greatest standard, the time in Photoshop as giving me some ideas for the future. At this point, I can start to see how all the systems link together and go hand in hand.
Ndo
Although I did not use Ndo in my pillar production, Ndo can be used to create accessories in the same terms like kit bashing in maya, except doing it through Photoshop and creating for example bolts or panels on the normal map, and then putting the normal map back on the model in Maya, giving the impression that the model has bolts or panels for example when really it is just a texture. Set out what you would like to create i.e. bolt, create a round shape, place where you want the accessory to go, and build. Then click on the ndo button and press play. Ideally, it should all go according to plan. In hindsight, I could have done this to resolve the problems I had in Maya with the kit bashing and the baking process, and would have saved time. Instead I did get caught up with Maya and making a bit of a mess in there. Knowing what I know now, I would take a different approach to it all.
DDO
DDo is the next step from Ndo, in that you can import your normal map and you can add further textures to the model. Make it look worn or battered. Its pretty impressive what the suite has to offer, it looks pretty endless with what it can do. However, this time I have not used ddo, due to complications with the normal map, which I can not work out why the normal map had gone a bit out of line. Otherwise, it would have given me a great finishing touch on the pillar, mixing what I originally had planned out from my mood boards and sketches, in mixing technology, religion, different surfaces and giving it that edgy worn look. So it is a bit disappointing that I could not achieve this.
Once everything done, this is the finished project, with the high poly on the left and the low poly on the right with its textures placed from it. To an extent I am satisfied. Mostly because after everything that failed and all the problems something worked. Something working better than nothing working at this stage I am happy to take. As I have pointed out, not what I originally wanted starting out, which is very frustrating! However, getting to know more what can be done with Photoshop, ndo, and ddo, it gives you more toys, tools to play with in the future. Another positive to take away from all the problems I have had with the construction phase, is knowing not to panic (too much) as there is an alternative should things go wrong else where with the construction part. It has however been interesting to see how Maya and Photoshop can work together in conjunction. All these possibilities do give you food for thought, again its just disappointing that I could get the original idea to work the way I would have wanted.
UDK import
Finally, the last thing left was to import the pillar into udk. As you can see from the screen shot, it worked. Just a case of importing the fbx file into udk. One of the few things to work so far in this project. That was quite a relief.
Conclusion
Above is the end product with the two pillars. Using the normal map from photoshop, then creating a diffuse map, by adding colour, then transferring the diffuse map from photoshop to maya, and adding it to the low poly model giving it some texture. The end result is not exactly what I would have wanted it to look like or what I had in mind that it would look like, but as nothing really has gone to plan in terms of construction and baking process and having to constantly back track, it has given me time to to think about all the process's involved, when not getting frustrated at every stage with the construction part that is. I would have liked to have kept more in line with my original design, keeping the sun image, and having more panels, also developing my imagination more. Admittedly the colours I have chosen are not interesting or great. The whole thing in truth is too boring and simple. So feeling frustrated. I would have liked to have given it a more worn an battered look, like of that we had been sampling in Ndo & Ddo. Maybe something for the future. I had grand ideas, but due to my own frustration with the whole process, I have let my temper get the better of me. There is a lot to look back and think I could have done x differently, and instead of looking at the alternatives, I got distracted by what was going wrong, and wasted time by doing so. In short, disappointing and much to be desired. Sometimes you have to move backwards and look at mistakes before we can go forwards.
No comments:
Post a Comment