Friday, 13 May 2011

Bibliography

Robinson, Josh. (2010) Using Maya’s Unfold tool so that it actually works. Vimeo. Available from:

(Accessed 06/04/2011)

Thursday, 28 April 2011

Post Production Edits

The post production side of Modelling was quite large for me personally due to some huge oversights on my part. The first oversight was that I had gotten my UV’s wrong in several places and I did not notice:

Of course at the point of noticing this, I had completed the texturing on Mudbox so it was impractical to change the UV’s and re sculpt/ texture my door so instead I changed the UV in Maya. Then using the UV snapshot tool, I overlaid the image in Photoshop on top of my textures – realigning them:
The texture on the right is the updated version complete with improved UV overlay.

Of course this wasn’t the only issue that I faced at post production. When extracting texture maps some issues had caused various irregular anomalies in my model:

These problems could be traced back to texture map. By using clone tool I managed to clean up any problem areas though:



Wednesday, 20 April 2011

Sculpting and Texturing

The next step in the workflow was to create a normal and texture map for my low poly door in Maya. I did both of these steps using Mudbox. Here is the finished Sculpted Arch and door:



After I had sculpted the arch I needed to texture it before exporting the paint layer into an image format for Maya. This was the final texture:

The textured, sculpted door in Mudbox

Wednesday, 6 April 2011

UV Mapping

UV mapping seemed possibly the most terrifying stage of the modelling process, but once the concept is known it is much simpler to understand. UV mapping is essential to texturing of an object because it dictates how much detail and information is given to a particular area of the model (much like the way a costume might be cut out of a large piece of material – the amount of material used for each area limits just how big or detailed that area is but at the same time too much cannot be taken from just that one piece of material so it has to be mapped out economically).
There are many ways to UV though. Chris Goodswen taught the ‘Planar Mapping’ method which due to finishing the modelling element late I didn’t get a chance to learn (and others I have spoken to have learnt the ‘Automatic Mapping’ method), so I looked instead at tutorials on the ‘Unwrapping technique’ by an experienced member of the animation community – Josh Robinson.
Using unwrap to UV is again simple if you know the right process (click on the below to zoom in

Of course there are likely much simpler methods to UV but this is the method I taught myself when UV mapping my door. This is the final UV:



(Note to explain the checker pattern: Window > Rendering editors > Hypershade then select the Lambert material before applying a checker pattern in the options under colour. This isnt entirely necessary to UVing but serves as a good indicator to the amound of detail each area has)

Monday, 4 April 2011

Post Modelling ( A look at the two door attempts and the final model after finishing the modelling stage).

Having to focus mainly on modelling, I hadn’t posted in a good while. Modelling has both been both a quite amazing and insanely frustrating task thus far while much has been learned. In the few weeks modelling I have created 3 doors – two with crucial errors. I will just briefly look at each door and where I went wrong (in addition the subsequent improvements made to combat those errors).
The first door I created had several errors despite looking very closely like the template:
-There were too many subdivisions for a “low polygon door”
- The subdivisions were at times very uneven
- Many objects went through each other (thus creating internal walls)
Plus the many that would have been picked up in the Maya/ Mudbox transition – basically a catalogue of issues meant that a complete rebuild was in order.


Given the issues with internal walls and un-even subdivisions, I focused on having absolutely no internal walls and much more even subdivisions by extruding out the entire door from a primitive polygon. This went seemingly much better but again internal/ hidden faces occurred, I had several ‘tri’s’ and N-gons (5 + sided shapes) and again I was forced to remake my door.


Given the issues I had though in both instances, I chose to make my ‘back up door’ whilst utilising what I had learnt:
-Extrude can be a modellers best friend
- The subdivisions must be even while remaining somewhat ‘Low polygon’
- Append to poly is much more useful in “hole filling”

The end result was as follows:


All in all I was pretty impressed with the modelled door. The door is essentially two linked objects (the arch and the actual door). I kept the subdivisions perfectly equal at the cost of the detail for the door while the arch’s shape meant I had to be a little more ambitious. Still with the next stage of UV mapping and texturing looming, I need to make sure that it works in Mudbox before I can begin.





Saturday, 2 April 2011

Key tools and Terms for Maya Modelling

The crucial points of manipulation for a modeller.

Invaluable tools for modelling in Maya



Split Polygon due to vertex placement can be a little inaccurate so often Insert Edgeloop is much better. Still should it not work when used often it can cause Hidden Faces.

Fill tool can be tempramental and not work (also causing strange inner faces) so I find Append much more useful.

 


When modelling there are definate issues that can cause problems during the Scupt or post production so these are a few of the modelling 'Do Not's'



Thursday, 17 March 2011

Orthographic Sketches

Orthographic Door Sketches:
For the Maya Modelling process, having a good basis for size and proportions is invaluable so an orthographic sketch is needed. These are the sketches made, based upon the two doors researched:
Practice door based upon St Mary Croft's door (due to uploading reasons all three views are not up but have been made)
My chosen door - with all three views

The next stage in the Modelling process is to begin modelling so after next Monday's lecture I plan to make notes and begin production on Tuesday.

Tuesday, 15 March 2011

Door Research for Modelling

This unit requires me to construct a digital model that is ‘production ready’ and also look at the methods – or production pipeline that is used to execute such tasks. The asset that is to be built is some kind of doorway so to give me some kind of grounding and basis for my model I went out and photographed and filmed two possible doors:

St Peter Mancroft:

It stood out because of the simple basic arched doorway, but yet there was room for a great amount of detail to be carved into it in mudbox.





'Saint Mary's Croft' Door

Just found this on the way to the above door. The actual inner door is simple but the outer archway is detailed and interesting - a 'second choice' should I have issues with the Mancroft door.

 
The next step in the production pipeline outlined is to orthographically draw the door chosen. I will do this later in the week (probably on both to practice my orthographic drawing (plus it would serve as a sensible back up should any erratic problems happen)