For me, Twoneironauts began preproduction on May 1st when I officially joined the team. Though it may have been going on prior for other team members, I think that marked the first push for art and design direction. Since then, things have changed drastically, and many concept pieces and design documents have been written. It's safe to say that art and design have been moving forward on all cylinders since that time, without signs of slowing. After about a month of solid direction and concepts, it was obvious we would crash and burn - and yes, it happened during Week 03. For this blog post I'm going to discuss some scheduling tricks that I've learned since many of us on the team are running into scheduling issues. This is by no means complete, and will require continual upkeep and time updates so that I can plan things accordingly, but it's a good place to start. Time Tracking and Human Error
The biggest issue we faced is just the one I mentioned at the start of the post - we were moving too quickly, at a too difficult to maintain pace, and more responsibilities (outside of Twoneironauts) are being dumped on the art/design team. A lot of this was due to the course work for ART399 and ART401. While these classes and deliverables may be maintainable with a clear schedule, every member of art/design side of the project has other things they have to worry about that makes it a challenge to schedule. For me, I have an internship that began during Week 03. It's been a difficult adjustment; I get home around 6pm, eat and get ready to work by 7pm, and then work (poorly) until midnight. This is not working for me. It's downright depressing to work a full-day's worth, and then get home and realize you still have hours of work to do on a different project. So we needed a restructure, we needed to reorganize many of our activities to be due during the weekend. But it's important to also have the deliverables planned before the week begins in case someone actually does have time during the week. My response to this personally is to plan out the summer entirely, have a rough draft of where we expect to be throughout different summer milestones, and work to meet those goals. The idea is to minimize the work requested to what can be accomplished within a single weekend, and hope we can spread out another day's worth of work through the week. This is insanely difficult, and sometimes just entirely impossible due to some of the requirements of the classes. So there have been a few unfortunately required "bad weeks". Just to spell it out clearly, let's give these definitions;
As you can tell, creating a character is complicated and time-consuming. A well-detailed character for Twoneironauts takes anywhere from 27~44 hours worth of constant work to produce. And this does NOT contain the initial preproduction style development times, where we experiment with style, shape, and silhouettes. We're talking after all that work has already been done. Even this is a very deceptive amount of time to account for character creation. I would argue that we should take these numbers andmultiply them all by 3 to account for more realistic time constraints, and human estimation error. Meaning the final times would be 81~132 hours of work per character. This translates to development time of about 2~3 weeks per character to take them from start-to-finish. We can double-up on characters a bit, and develop some of them along with others, and this number ends up being pretty accurate. (I believe we will spend 2 months just finalizing the player-controlled characters, we're half-way through that process now, and this looks to be an accurate estimate for 4 characters.) Many people might look at these time estimations and think they are inaccurate or "to big" of estimates (especially the expected times). I would argue that the normal hours are someone working at 100% efficiency, where-as I'd estimate that a more realistic human would be capable of about 60~70% efficiency on a good day, down to 30~40% efficiency on a bad day. (Some days people need to check out entirely, so a mixture of this allows us to estimate an average of 45~55% efficiency.) Wouldn't this mean we only multiply the number by 2? Well, yes - but that's not taking my own lack of expertise and human time perception errors into account. It's proven that humans are just really bad at time estimation, several studies have been done to prove so. In fact, someone who is more interested in a task has a higher chance of shortening the time estimations it takes to complete that task. This is the study of Time Perception - and one of the rules is that tasks that someone is familiar with are likely to be greatly underestimated. (Meaning without actually tracking time, an estimation may be orders of magnitude off, or not taking into consideration down-time on a task.) So we increase by 3 times due to time perception error and efficiency likelihood. Design Limitations Another issue I've run into is that I'm practically managing the schedule for the art team, and in many ways am acting as the Art Director. This has had it's own slew of challenges because I am not very well-versed in managing an art team (at all). But this is part of why I wanted to be a designer! So that I can learn these things, so that I can get better at this stuff, so that I can demystify some of this magical art stuff. I would say I've learned a lot, but it's been at the sacrifice of my time to provide great design direction on the gameplay. There are a ton of things that I know need done even now - that I just simply can't find time to do. Part of this feels bad, I get the impression that I'm slacking or not doing my job, but I am working very hard. I'd say about half of my time in Twoneironauts is spent on meetings, the other half is spent scheduling and writing documents for deliverables. If I were able to apply myself full-time, I'd say that this work would only account for half of my time entirely, and the rest would be open for design direction sessions (which I originally began the project by doing on Week 00~02, but it's definitely dropped off the map as of late.) This is a concern of mine because I don't have time to test or prototype the things that I am unsure of, and can only make pen-and-paper estimations (which any designer knows will only get you so far). The way I will want to mitigate this is by having the development team prototype some of the design decisions. I haven't met with them to request this yet because we aren't ready for such tasks on either the design or the dev time (still getting some boilerplate code working and everyone merged together, to my knowledge). But it would be greatly helpful if I could draft a list of unsure design directions, and proposals for prototype situations that would prove which direction would be the best to go down. The major issues that need solved right now on the design side are still very art-heavy. Environment needs about a weeks-worth of design work to figure out how we will get it into the game, and animations are in the process of undergoing their first in-game experimentation. We are in the talks with another artist for joining the team (much needed, for the scope of the game, I'd suggest 4 artists ideally, but 3 would be doable. I have never considered 2 to be enough artists to work on this project.) So while animations are on their way to being capable and working in-engine, I think environments and the process for getting the environments in-game are more mysterious and unsolved. I plan on tackling this next week, and we are going to begin providing some sample "deliverable assets" in two weeks from now, maybe three. (Many teammates have expressed getting these sooner, but I do not think it's realistic for me to promise anything sooner. At that point me promising something in 1~2 weeks would just be me lying, and that isn't fair to anyone.) Really I need another design-hand, but there isn't enough in place to allow this designer to be used effectively. I think at this point it would only hurt development to hire another design, but in the future I could definitely see a use for another design hand on deck. That's All This Week I think this is an important post, that everyone on the team should read if they want to know all that goes into this time estimation and how I manage scheduling tasks. More goes into it than I think people realize. In fact, there has been a ton of design work even that is just sitting in a little notebook that I really need to make digital so the team can see it! However, it's been a constant battle for time recently, and I think that design unfortunately needs to take a backseat for a week while we figure out how we're physically going to create the art and get it into the game. (That process needs to be roughed-out at latest by next week.) Until next post - Cheers!
0 Comments
Leave a Reply. |
Archives
April 2015
Tags
All
Production BlogPosts by our team members. |