29 posts tagged “thesis journal”
So with four days away until due date, I'm going to start doing a run down of the elements in this film. Mostly to show what they are, but also a bit of discussion as to how these pieces fits (or doesn't) within the project.
First off the motion graphic pieces that shows up on the TVs and jumbotron in the film. These were definitely the hardest elements to lock down. I originally planned to create up to six versions to be mapped on to the CG screens, but with production time for the CG elements taking up more time than I had hoped, I decided to cut down the number of screens to three.
The most difficult thing about the production of these screens comes from the fact that everything was all based on the same "brand" 'The number one product', I wanted to keep typography consistent thought out all three spots, but that created a quit difficult situation in terms of designing them, since to have each screen holds it's own in the scene, they have to be stylistically very different spot.
The three themes I went with were, a pop culture based spot, a product/infomercial spot and a personality/celebrity spot. The idea behind these spots and actually the whole film are based on a lot of Asian advertising styles, where advertising literally only feature the brand and an A-list celebrity spokesperson and most of the time, the celebrity really have nothing what so ever to do with the product. I've always found this concept to be very interesting where a brand is purely driven by celebrity tie-ins. Especially compared to western advertising where the focus is associating an emotional attachment to a brand and less about having some celebrity vouch for a product.
Of all the elements I created for this film, I really hoped to have more time to develop the design of these spots. Personally I really want the designs to be a bit more thought out and more elegantly put together, but not having these spots built were really holding up the CG rendering, plus given it only shows up for about 1 and a half second or so in the whole piece, this was a compromise I am happy to deal with.
At the same time, I really do like the somewhat over the top boldness of the moves and treatment. In a way, it fits the very Asian corporate design sense. They want it loud, they want to show that they spend a crap load of money on the spots, I was definitely thinking of my art director when I was working in web design in Hong Kong, I remember the very line I got was "Keith, your designs are too American, you have got to add more to it, the clients are not going to be happy to see they paid all that money for a few lines and blocks. Needs to be louder."
The past few days have been a lesson in learning when to call it quits. For all intents and purposes, I've been done with this project for two weeks. The good thing about this is that I have ample time to tweak and get feedback on the piece, of course the ample time also led to an overly obsessive attention to minor details.
For example, should anyone really need to spend 3 days tweaking a 15 frame transition? It is my hope that the attention to details at that level will show, but at the end of the day, people will most likely not notice, or pick up on something very obvious that I've managed to miss. As a friend of mine commented after showing him the piece, he said, "When you stress out over a detail, I can't tell if you are just being neurotic or that you are just really good at this." I didn't say at the time cuz I'd like the think I'm really that good, but yeah I was just being neurotic.
One of the things I was glad I got feedback on was on was for the titles and credits. During production of the motion graphics elements, I was obsessing over a type treatment that I really wanted to try out, but it didn't fit with the overall styles of the motion graphic pieces I was making for the screens. So I decided to save that for the titles and credits as it felt like something that would work well with my piece. So I spend quite a bit of time making that work and convinced myself that it fits the overall style. But stylisticlly it was such a huge deperture from the actual piece that I decided that it actually took away from the actual content of the film. This was definitely one aspect that I was very glad that I got very constructive feedback on, as if no one had pointed it out, I'd have convinced myself that the very elegant, fine and curvy type treatment actually fits with a CG piece that features a lot of really bold type and hard angles.
I think I found a middle ground with the type treatment, I really like the outline to fill transition for me to completely give up on it, ending up going all caps and a bolder serif font (Breughel Black). It's a good thing to have a departure from my love for San serifs.
I've gone back to reading my thesis research paper and reread my goals as to what I am trying to achieve with this project, I was really hoping what I wrote a few months back will help me make this decision. But they both work within the style I'm looking for, I think the warmer look fits better with the overall theme of what I'm trying to say with this piece. But the cooler tone I think looks better... eh, decisions...
With 10 days to go before thesis due date, I find myself at an odd situation. I'm done, and have been for a few days now. This is new to me, very very new to me. I've been known to be the type of person who's always stressing out, working down to the wire, to the 11th hour, sleepless for weeks, making deadlines minutes before it goes on air kind of deal.
But here I'm done, ok done not in that "this is a masterpiece, you can't improve on perfection" sort of done, more in the likes of, I'm content with what I've created and will be fine with handing it in, if I am told that this is actually due tomorrow.
I mean as with any creative work, is it ever really done? You either run out of time or you've given up on trying to improve it. The only reason I've stopped is because I have solid final renders for all my scenes and decided I needed to get started with compositing and check if all the scenes are working.
This is a rather scary time, the next 10 days, for I'll watch the film over and over again, looking for the tiniest flaws and attempt to fix it. And I'll show it to a friend, who will watch it and make a very innocent comment about a color choice, a lighting angle or a pacing issue and have that comment throw me into a frantic emergency mode as I attempt to redo complete scenes because that never occurred to me during production.
Right now I'm busying myself with color grading. I have two color styles I'm toying with, one sticking to the cold stark sci-fi blue, green thing. Another one where I'm pushing a warmer color set over a what should be a stark cold environment. Both looks I've played with in the past year photographically, both styles I love equally.
This is the toss up now. I'm leaning more toward the warmer tones, since it's something I've started playing with recently and also a style that I want to develop further.
While the colder looking blue green style is much more indicative of a stark urban distpoian night scene. As a color and a style, it's easier for the viewer to identify with the mood I'm trying to convey, as they can relate to prior usage of such color combination.
I have both version rendered for a screen color test tomorrow, as much as a creative decision this is going to be, it may very well end up being a technical one where one of the styles works better with the projector we are using during thesis defense.
Also a good thing to have noticed this early is that, I'm already seeing presentation issues right now, my D1 footage looked a lot brighter after going through a mpeg 2 compression to go on DVD. This may just be the difference of how Quicktime plays back on a computer VS how the apple DVD player plays back footage, the issue right now is the Apple DVD player pushes the gamma up way too high, it would be really bad for this to ruin my presentation. The color test later will help answer a lot of these questions.
And so it's on. Basically now other than the end scene where I am still having problem with getting the camera move to work, all my scenes are locked and it's pretty much all rendering from now on. I'm a little annoyed that I had to move my projects to maya 8.5 to render, the mental ray integration in 8 is just pack full of issues. It's not really that big of a deal having to move to 8.5, I just don't like upgrading software in the middle of a production, especially this late into the process.
Overall though it looks like it's working but it does look like I will not make the soft deadline of April 10th, I'll be two scene short, so really it's a day or two worth of work so overall it means I'm actually ahead of schedule in terms of CG work. I'd like to lock all CG work by Thursday April 12th and spend the rest of the time focusing on tweaking final composite and title work. This is one area I really want to spend a bit more time on, sure it's not the focus of the project, but I feel the overall packaging of the piece is important. It's quite exciting right now actually, things are looking good, sure I'd like to be further ahead, but overall progress is steady and solid.
It's getting close to due date, things are moving along, not as well as I hoped but good enough for what it is now. I was hoping to make the soft deadline of April 10th, it's still possible but following the better judgment of personal wellbeing, I decided that I'm not going to kill myself trying to meet the soft deadline, plus I'd like to give a few scenes a bit more time to tweak.
On the note of tweaking, I was going over my storyboards from previs and really there were quite a few scenes that are quite different now in production form. Most of the changes spawn from technical problems, but also there are aesthetics issues that arisen in the translation process from storyboard to CG. A lot of the framing I had just didn't work out as well once in CG form. I tend to frame wide, whether in my photographs or in my design work. This is something I've mostly picked up from photojournalism and landscape photography. The need to provide context to the subject, the environmental elements are a big part of a visual presentation.
Wide in the context of a cinematic narrative, of course doesn't work as well... ok it does, just not in my case. The issue in this piece comes from the fact that the subject is the environment, not only was the wide framing making the scenes very objective and lacking focus. It also created a major problem with rendering time and on a more logistical level, created a massive extra set of buildings to model and texture. The wider shots also means there are large distance the camera has to travel, which created difficulties in making motion feel natural in such a massive scene.
Maybe it's a little odd that I am talking about camera motion that feels natural, because I'm trying to stay with camera moves that's physically possible to achieve with a real camera and a good camera rig. And really there's nothing natural about that. Maybe more appropriate would be to say I'm looking for a more traditional cinematic style. Sure it'd be fun to do one of those massive Lord of the Rings fly overs or a Fight Club style building fly-through, but these moves are starting to feel like something you drop into a movie because you can. (Yes yes, it's also very hard to do right and I can't pull it off... please let me finish executing my well formed excuse masquerading as an artistic decision.)
The change to a tighter framing is a departure from what I usually do stylistically but I feel that in this case, it worked out for the better. The tighter framing created more tension in the scenes and also made it easier to direct the viewer to specific subjects via variation in brightness and converging lines.
About 24 hours ago I was writing about this render. I'm still writing about this same render... granted it's actually working now, but it took way too long to figure out just what was wrong with it. Memory issues...
This scares me a bit, knowing how 5gb of ram on my mac is actually not enough to make this one scene work, I threw it over to my PC only to remember I didn't get that ram upgrade when I got the machine. No love either. So I spend bulk of my day trouble shooting the scene, cutting down, removing elements that I don't need. And I still have things to add to the scene...
I'm not happy about it, but I will most likely have to do this scene in passes, just not something I'm interested in doing right now, this scene already being a massive 12 hour render, I can't afford to lose a computer to rendering for a full day. I don't even want to think about going to school to render, I'll be lucky to find a chair to sit, getting on a fast, working computer will indeed take some major hustle and possible physical violence... 40 something thesis students and 20 usable computers... Way to go NYU.
I guess it's a good thing that I'm getting to the stage where I'm doing production level rendering at this time of the process, it's good to get an idea as to how long things are going to take in the final render. It's odd when you think about the difference in time, when I was doing After Effects work, a six hour render for 30 seconds is insane, I get worried when it takes a minute or two to render a frame.
Going to CG it's something else all together, 6 hours to render 90 frames seems quite normal and even after all this time, it's still something I have to get used to. But progress is good, as of now every scene are lit, all but two have final textures applied and now only one scene I haven't done a full res render yet. I've managed to be ahead of schedule by a bit, not a whole lot, but it's good to know that I'm on track.
On rendering, since my desk sits right next to my bed, I've gotten used to falling asleep to the smoothing sounds of spinning hard drives and revving cooling fans. I used to complain about how loud the fans gets on the G5s, now it's gotten to the point where it has become the reassuring sound that my computer is working on my project. The sound is also why I'm up right now, as the fans stopped spinning and literally I woke right up like an obsessed boss, looking over the computer's shoulder to check if it is indeed working or just pretending to work. Turns out it went for a break between frames and forgot how to start working again...
Above is a frame from a production resolution mental ray test render of my thesis. While it's still ways to go, I'm so happy as to how it came out. There are some edge buzzing issues that I'll have to address in my next render test, but as of now, I like how my shaders are looking, camera move on this shot got a little more tweaking and it's going well. I have three more scenes that's close in lighting stage. Things are good if maybe a little behind schedule, but good none the less.
One of the things I am very happy about is how smooth the shoot went, doing two test shoots with my actor before the real shoot really help work out the problems. The most important thing was working out issues with working with a green screen. It'd seem that as something I've done a lot in the past, doing a green screen shoot should be one of the things I am not concerned about. But with the space, budget and equipmen limitations that came with this project, it was a good thing to have worked out all the issues that could arise during the shoot before hand.
I did end up buying my own green screen to work with, seeing how the super high tech, expensive and hype invested screen we have at school is about as useful for chroma keying as shooting against a moving street scene, the money spend on the screen in the end of the day just looked like a smart investment for my sanity. Plus seriously, I don't think I've ever lit a screen that text book perfect, sorry had to brag a bit, I've always had problems with lighting green screens and getting good separation. This is a major break through, yes, it's been a good week.
