Arons Computer Graphics Blog

Tuesday and Wednesday, 21st-22nd

2014-10-22T00:00:00+00:00

Last chance for coffescript. I made sure almost everything had a descriptive comment in the code. Did the final polishing and refactorization of code variables. This was mostly about making the project source code as readable as possible and making the turned in stuff as good as possible. Another big part was of course writing the report.

Sunday 19th

2014-10-19T00:00:00+00:00

I added some icons from kickstart for settings in the scene, like lighting, camera etc. (This menu was not needed in the end though since I didn’t implement them, and just set rotation of the object as default.)
Finally I converted the javascript to coffescript and setup compiling and minifying. Removed all the unused stuff from the directories and restructured a little bit.

Friday 17th

2014-10-17T00:00:00+00:00

4c. I added ambient and direct lightning as well as set the vertices of the model to white and background to black. 4d. I also added “orbit controls” for moving around 5. I used the kickstarter for simple ui stuff like the bar and icons. And added a <input> field and eventlistener for file submission. 6. This step was rather about glueing the file in the input field and object loader in three js together. this took some time. the standard three.js loader expects a url so I had to dig into the code quite a bit and run the obj-file parser from outside 7. The loading of the model was done in step 6 8. The only obvious thing to do for loading of different models was to reset the scene when a new model was loaded so i added some code that removed the previos object before that. 9. Default rendering was done in step 4b so it was already done. 10. I came up with the idea of moving the initial camera position based on how big the model was. 11. Orbit controls was added in 4d, but I tested a couple of other controls examples and the only feasible alternative was the standard FPS “w,s,a,d” with camera centered mouse rotation. If there is time I might add that as a toggle switch.

Thursday 16th

2014-10-16T00:00:00+00:00

Set up project roadmap notes
Downloaded revision 68 of three.js, sublime text, chrome, set up repository on github. I also quickly decided to work remotely from the start though the kth web hosting.
I set up a spinning cube type hello world example from three.js website
I choose teapot and teddy bear from MIT and choose to wait with step 5 for now, because id rather see some successful result of adding an arbitrary model even if it was hard coded. I felt that made the process more modular in that I actually knew that this part was working. 4b. I managed to load this model with the standard object loader. Since the documentation page is “todo” this step required some in-depth research and was pretty hard. I found the files I needed to work with their loaders as it was probably intended though.

Roadmap

2014-10-15T00:00:00+00:00

Think about the report and log all important trials and tribulations during the process
Download and set up the development environment
Get a “hello world” three.js WebGL example showing
Find test model
Create HTML interface (“Load Model” button, title, optionally short explanation, link to code)
Print OBJ file in console (accessing model in JavaScript works)
Load model in three.js
Test with different models (general model loading works)
Render model with default settings
Research rendering options and then implement them
Research 3D view controls, camera settings and then implement them
Optionally polish interface, perhaps with bootstrap
Refactor and minify code
Upload and test on server
Write report and compile the logs

blog take off

2014-05-21T00:00:00+00:00