After 9 grueling months, I’ve finally finished the environment for the centerpiece of my portfolio.
The concept behind this piece was to create a 3D, interactive environment for users to explore my work. I created a shopping mall with shops that are named after my portfolio pieces. Ultimately, gameplay elements will allow the user to view my portfolio work directly through the shops.
Some cool stats on the mall:
Browser based player, small file size for quick download. Current size of
entire player is 10.1 MB. The project should load in under a minute on a
standard cable/DSL connection. A more interesting load screen is on my
list of things to improve.
All shelving uses a combined total of two 256 textures. Wood/metal
texture and a product sheet. The products are aligned on a texture in a
grid pattern. Object UVs are offset in multiples of grid size to give
randomized, unique looks to each shelf. No two shelves look alike.
536 lights in scene baked into lightmaps. Character is the only object
that receives in-game lighting.
Everything in the mall was created by myself in my spare time.
Environment, character, models, textures, lighting, rigging, animation,
scripts, physics. Creation time: 9 months
Future plans are to have rats running through the mall, avoiding the
player. For each rat that is caught, the player gains a token that can be
spent at shops to unlock portfolio pieces to view. Gameplay was secondary
to finishing the environment, but I feel that adding gameplay would increase
the time a user spends in the game and make it more fun to look at.
Before it comes up in conversation, I know the lack of a jump button is
annoying, but so is trying to design a character controller for one.
I’ve got the jump animation done, but it doesn’t sync up with the controller.
The jump button immediately puts the player airborn, but the animation has a
half second of anticipation before jump. Combined, they make for a very
strange looking jump cycle.
That is really good
It reminds me of the mall in “Dead Rising”. Now it just needs the obligatory elevator music in the background. I wanted to jump in that fountain so bad
My MacBook chokes a bit on the frame rate and the bricks used in the mall are WAY too gigantic. Otherwise nice idea, I love the sprint animation and major props for including the Puerto Rico board game on the shelves!
I do wonder about including rats in your portfolio.
I’m curious how it will really work in the end, as well. Doesn’t seem (from your original site) that there will be that much to fill the mall. :?
I’d suggest you modify eye-level and/or store content to make things look a bit less sterile. Most malls are designed to fill your field of view with merchandise – these stores all look empty.
It might help to go to a mall and take a bunch of reference photos.
S-mart ought to have more weapons on sale (it’s where Ash from the Evil Dead movies works).
Nice work so far. I don’t think you’re making the most of your environment. Your lights and colour are bland, predominantly grey and too consistent. I think exaggeration would make it more obvious - is this space friendly or threatening? I’m unsure whether it’s deliberately or accidentally spooky.
If it is intended to be spooky, textures should re-enforce threat. There are plenty of opportunities to create threat with very dark shadow areas, but the evenly spread ambient light, perfect wall lights and time of day is not making the most of that.
If it’s intended to be friendly, textures should re-enforce innocence and reduce threat - with colour, colour colour! Ramp up texture saturation and brightness by 300% to reduce the spook. Also, the space is empty and lonely. What happened? Populating the area with “living” stuff will reduce threat.
Usability
You’ve made some good opportunities to attract users to interactive areas with your model but at the moment I see everything as kinda same-ish. Interactive nodes should stand out more.
Technical
Moving the character to a tall mesh and rotating gets a microscopic view of the player mesh flaws. I’d suggest a larger minimum camera distance;
Is this for web / multimedia work or game work? If web, there’s lots more optimization needed;
The floor tile bump map has some aliasing problems. You could work around this with filtering but I don’t think the floor texture really needs a bumpmap, why not fake it with diffuse?
NIce and crisp … i just loved the fact that you utilized the texture resolution quite accurately … non of them looked pixelated…quite cool…
doing all the art work , renders , scripting … u r a one man army…
cant wait to see more…
Just wanted to say that this is a nice piece overall. Lots of excellent texture work and great animation for the character.
I think the disconnect is mostly in the purpose for the piece. If it really is a portfolio, I’m waiting to see how you pull that off effectively. The whole environment feels a little large. I’m not sure putting rats in there is going to be enough. Maybe some clever game mechanic is needed. One that gets the player to really check out all your stuff in the mall. I don’t know.
I’m sure you’re up to the challenge based on what you’ve done so far.
I thought it was pretty amazing, even more so considering it’s a one-man effort. I can’t begin to imagine the work required to produce an animation that smooth and all the models and textures.
BenH: Audio is on the horizon. I’ve got some elevator music done, but no sfx yet so I didn’t find it fitting to incorporate it just yet.
soad667: Zombies? What do you think this is, Dead Rising? I doubt a nerd would know how to handle himself in such a situation.
aaronsullivan: Which bricks? There are quite a few stores with brick patterns. As for the rats, they are just the dominating idea out of many that I’m shuffling around. I am definitely open to better suggestions for gameplay. About population of portfolio pieces, this is merely a method for which to unlock and open images of the pieces. I’m not going to be actually populating the shops with objects themed after my portfolio. As you pointed out, there’s not nearly enough stuff for that.
podperson: I’m aware of the sterile nature of the mall. Unfortunately, this was designed to be run through the web, so massive quantities of store contents was out of the question due to the amount of inflation it would cause in file size. What you see now is a compromise between load times and content.
Proto: wow, lots of good crit. Let me step through these.
I’m not quite sure I see the ‘spook’ factor you are referring to. I’ve gotten lots of comments from many different audiences and you’re the first to refer to the environment as spooky. Might it have something to do with the lack of other NPCs throughout the area?
There are no interactive nodes just yet. That might be why they don’t stand out
The current camera min position is the distance between the character collider and the camera collider. I suppose I could increase the radius of the camera collision to increase that distance, but that would also make the camera collide with objects from a greater distance. Perhaps a rework of the whole process is in order.
It was previously for both multimedia and web. The CD version is now complete, so the sole focus is web. What sort of optimization have I missed out on? I’ve already made drastic reduction in texture res, cutting the project’s filesize nearly in half.
There is no bump map on any objects in the scene. The aliasing you are seeing is from the normal map used when baking the light map. The floor is such a large object that a decent res light map would be extremely expensive. Perhaps I should kick out the normal from the floor and re-render the light map.
MikeJ: hah. You caught me. The products fall in four categories from things I’ve scanned around my house. I basically randomized the textures to give the shops each a unique feel. The only things sold in this mall are music, video games, board games, and cereal. Now that you mention it, video games should sell cereal. Saves me a trip to get munchies on the way home to play a new game
Thanks GusM, orbitsteve, imparare, charles, and everyone who took the time to view my work.
Mattimus – I don’t think you need MORE content, just better use of what you’ve got. Make individual boxes bigger, shelves should fill your view. Stores try to look like they’re full, not empty.
By ‘spook’ I mean the space is troubling for me… Either I shouldn’t be in here, or everyone has left in a hurry - even the shopkeepers. This is a familiar horror film scenario. I extrapolate from the number of films using this type of scenario that unexpected empty spaces are threatening / spooky for many people. Perhaps I’m just the first to articulate it in this way? I would expect at least some of your audience to recognize the reference to the survival / horror genre.
The file size is still massive for a web app, even though I understand your primary audience (i.e. potential employers?) will have faster connections. I believe 10Mb would be too large for most of your employers’ clients web pages, or for a general interest web site. I think 10Mb is better suited to a download.
Re: camera distance. I’d keep your player collision tight and enforce a minimum distance using script instead.
@above: Try firefox, same happens for me when using explorer ;o)
@Mattimus: Congrats on a great demo. I really think you did a good job and I appreciate the time you must have put into this. There are bits that could be improved but overall the feel is really good and it really “feels like a game”.
You should probably have a look at the smoothing groups on the character as it looks like there are many hard edges in the face region.
Other than that, congrats on a great piece, to me this is one of the best examples of what you can achieve with Unity
