Tuesday 2 March 2010

Problems...Solutions...Outcomes...What did i learn?



This section of my blog i am going to focus upon the problems we faced as a group and how we went about solving and if we went about solving them properly, and in turn whether i learned anything by my experiences.

One of the initial problems we came accross was the bad weather and snow in the first 2 weeks of term which hindered our group progress. The tutors argued that we could have still been getting on with the project at home, but it wasnt the same without direct contact with eachother. We were still able to talk over MSN and via phone, hoever it isnt the same when we are all in college helping eachother out whenever we needed it. I was quite confused to find out that we werent going to be given any sort of extension for the project, as in theory we only really had 7 weeks of time to produce the finished outcome if you take into account the bad weather which is a rediculous demand to make whilst having another term long project on the side, and having to worry about a PPD and contecxtual essay until 2 days before they were handed in. I know that this wasnt the case for just our group, and it was for everybody, but i do feel that if we had that extra weekend of solid work, we would have been able to get the project done to the standard we would have liked and it is pretty stupid to sacrifice the final quality outcome of the work for a couple of days where the work is sitting there and not even being marked. But that is just my opinion. Taking a different perspective though, maybe this will enlighten us to time management in the future and having to quickly and effciently re arrainge our schedules if something as random as bad weather comes up when we are working in industry.

One of the most frustrating problems we came accross as a group, and more specifically one that i found most challenging to overcome was the project folder issue that came up at the beginning of the last week. This problem threatened to jeapordise the whole outcome of our project if we didnt get it fixed fast, as we couldnt progress on to animating lighting or rendering if we didnt first pass this hurdle. We found ourselves having problems transferring scene files from one persons computer to another without having to re assign the textures every time. And even after we thought we had solved this project by re setting the project to the correct destination and then re assigning the textures to the correct destination the textures were not showing up. This took me and Guy 6 hours of frustration to work out, after having talked to Dan Dali, who suggested making anew project, and copying the old scene file data and old texture file data over to the new project, as this would eliminate any Preference confusion in Maya. Thankfully this worked, However it didnt work every time we tried it from then on and we found ourselves regularly having to re assign textures to their correct destinations which was in truth very very annoying.
Another problem we ran into was rendering. We left it till the final day to render one of our scenes which was stupidity at its highest. We were warned many times by Jared and Alex not to do this but we fell into the trap and there arent really any excuses for doing so, we didnt even get a test render done until the penultimate day which again was a stupid idea. The main problem with rendering was too much data in one of our scenes, the textures on the outdoor scene were too large and the poly count on the outdoor scene was far too high too. Clym helped with tidying up the scene on the final day of the project, and allowed us to get a 1 shot render time down from nearly 2 minutes to 27 seconds which was brilliant, and we were very thankful he was there to help us out on that otherwise we would have been stuck. Also we had mysterious problems cropping up with rendering the outdoor scene which started to frustrate me and Steve. The church wasnt showing up in the render view on Maya software, it was showing up as a blank black screen and test renders showed it to only be rendering out a blank black screen too, which was very strange. We found a solution which allowed the scene to show up in the render view, but only whilst using Mental ray which was annoying, as this increased the render time for each frame. In the end we ran out fo time to render the final scene, and we had to include it as a playblast in the overall compositied movie which was dissapointing to say the least after all the effort the team put into it.

One thing i think myself and the rest of my group will agree with is the importance of communication throughout the team. In my opinion this is vital to the smooth running of things right through the begigining stages of the process, right up until the end where all the work is handed in. I feel strongly that it is important to keep in touch at all stages of design to keep continuity, i also feel it is important to air issues we have with nthings within the group, sometimes if we are confronted with a problem, calling a group meeting and talking ti over until it is resolved can be the only way to deal with things, as it was in teh case of the skull debate. I feel i have learned a lot since the start of this project, not only the technical aspects in dealing with Maya, like UV mapping, texturing, modelling, and lighting, but also moral issues like how to deal with other people within a team and how to respond if things go wrong. All in all i have mixed emotions about this project's outcome... i am somewhat dissappointed with the final movie, however there is definitly room and time to go back and improve upon that outside of class time. I am very pleased with some things too, such as now having the confidence to go on and learn about maya more deeply. I honestly feel that this term has taught me about 90% of the information i know since starting this course, that is how much it has allowed me to develop on a technical level.

No comments:

Post a Comment