5th Developers Diary - The making of the.... graphics!

I don't even want to count how many weeks or months has it been since our last update. However, we're back with another
developer diary of the Until I'm Gone project by Faraway Studios... If you have visited our homepage recently then you 
probably know the reason of the delay of the updates as well as the development itself. If you haven't visited the homepage,
then let me explain: some of the team members and especially the team leader had to take the final high school exams. Because
of that the development slowed down considerably in mid-April 2008 (because we're all such nerds :-)) and then started again
in the end of May after the exams and some rest. As of now everything's back to normal, but nevertheless the project and the
team will undergo some minor changes.

Anyway, with todays theme we'll be starting a sort of a small series as a part of our developer diaries - "The making of...".
In these we'll be talking about various parts of the game creation process (Graphics, Animation, Sound, Music, Programming
and maybe even the screenplay) from time to time (when we run out of proper ideas). However don't be expecting anything in
general, or God forbid something that might actually help you with creating your own game. It'll mostly be just a description
of our creation process and its quite possible that amongst our readers there are some professionals who've been working on 
commercial projects already and who will be shocked and appalled by our methods... But even if you're an average gamer and 
don't give ... about game development, we believe you might find this subject interesting and it could serve as a brief look
behind the scenes. Now as the defense of the quality of the following article is almost complete we can finally start with 
the article itself... It will be divided into two parts (the second one will come out next week) as we've really outdone 
ourselves this time and the diary (should it come out as a whole) would really be EXTREMELY long. After all, if you scroll 
down (as obsolete as it is :-I) you'll be able to see that it still is an impressive amount of text we've sweat out.

The graphics - a very wide subject. As you may know, our adventure game is 2.5D - obviously that's nonsense, nothing like 
that exists in reality. The term itself was invented by some wiseguy and everyone loves to use it (as we do). But why not, as
it illustrates quite well the situation: it's a mix of 3D real-time characters moving in "flat" 2D images (scenes). And 
2D+3D=5D (yes, there really are 5 dimensions :-)) and the mean is 5D/2 which equals to the proclaimed 2.5D. 

The characters (and a few other things) are fully 3D. The "environment", that is the scenes, locations or let's just simply
call it the "images" are in 2D. These are the so called "pre-rendered scenes". That means that these pictures were a 3D 
environment in the beginning, but after their completion (modeling, texturing, lighting and rendering) they were rendered 
into the exact picture (from the exact viewing angle) that we were aiming for. The renders are used primarily to keep  
the hardware requirements as low as possible. Thanks to this the game should run even on very old computers - the unofficial
hardware requirements for now should be something around a 500MHz CPU, 128 MB RAM, 32MB GeForce2MX... but we're still not 
promising it won't change. By using prerendered backgrounds we're able to achieve nearly photorealism with only a fraction of
the hardware requirements that would be necessary to display the scene in a fully 3D environment. Well, that was maybe a bit
clunky definition but hopefully even those, who in the beginning had no idea what I was talking about, are getting it now.


The animations are usually mentioned along with the graphics as well as non-rendered 2D images (in our case various 
wallpapers, the GUI (graphical user interface)) and other things. We're going to omit these for now but we'll hopefully get
back to them later - to the animations at least - in one of the upcoming developer diaries. In this one we'll focus "only" on
the pre-rendered graphics of the scenes (locations).

So how is a scene like that created anyway? In general you could say that in the beginning there wasn't the Word but the 
Idea. Then a draft or a sketch, some words maybe, then a more elaborate sketch, a basic model of the scene, a lot more words,
downloading/taking a few pictures or videos, modeling of the scene, texturing of the scene, lighting, feedback from a senior
team member, adjustments, a final render... Well, and then the post-production and adding some tiny details that you've 
forgotten (every graphic designer loves that part). This is how the creation process of the scenes for an adventure game 
should look. However, the reality is often very different from the theory. In the following paragraphs we'll describe how we
do it. 

The idea - before a scene is created, there should be an idea in the beginning (if there's none, horror ensues). The idea 
should be developed then and thought out well (if it isn't, even MORE horror ensues). If you come up with the idea that the
first scene will take place e.g. in front of the gates of a magnificent castle floating in the skies on a cloud and you're
hoping to "come up with the rest later" (as you already have a general idea, the setting and the atmoshhpere) and you
promptly shift to other parts of the development process trying not to lose precious time, then you're doomed! You'll soon
find out (if you're lucky that is) that you have no idea how big is the castle supposed to be? What should the walls look 
like? Should there be crenellation? What about the towers? What ground-plan should the castle have anyway? What color and 
opacity should the cloud have? What should the main gate look like? Should there be a draw-bridge? After all, there's no damn 
moat around! Will it look good if I put there a moat after all? Can there be moats on a cloud anyway? And do I want a castle 
or a chateau :-)... You'll find out that you've got the basic idea, but the other things that "you'll come up with later" can
be thought out later but it will only cost you more time and effort. That is if you still want the scene to be worth it (if 
not then you can really come up with it later but it will sure as hell reflect on the quality).

The draft or the sketch - a picture can say more then a thousand words... Everyone should really follow this rule (and that's
exactly what we didn't do in the beginning :-) ) If there's no sketcher on the team and the author himself (that is the 
person who should "know it the best") doesn't really have a talent for drawing sketches you can make it up with a ton of text
and some photos of real places, but still: there's some truth to this saying. On the other hand, if you don't specify all the 
details of the scene you leave some freedom to the graphic designer and that's definitely positive. However, you're still 
gonna need some skill and good nerve to keep a unifying graphic style. In our team, we solve it (or used to solve it as most
of the scenes are already done or not far from it) with text and real photos. Firstly it was faster ("Why bother with getting
someone just to draw sketches?" I hissed to myself as a rookie in the beginning of the development) and secondly we gave the 
artists more freedom than they would get if they received a detailed sketch or an artwork (that usually looks much better 
than the final scene itself (as I've seen in many other projects) raising the question if both of them are really 
necessary...?).

The word - from the previous paragraph you can probably understand that we've written quite a lot of text already. I should 
point out that it is advisable to keep the text clear and concise (and probably take this advice myself considering the 
length of these diaries). Also it's a good idea to take a look at text after a while and review whether it really is clear
enough. After all everyone thinks a bit differently and while reading you can't hear the intonation of the voice and see the
mimics of the face of the writer so in the end it can be understood in a different way. 


The new "development" screen shows how "time goes by" in the project and the scenes themselves. Here we can see several 
versions of the same scene in:

- The very beginning of the creation process created by two different  
graphic designers (pictures 1 and 2) that haven't been used in the end.
- The first version of the "final" scene (picture no 3)
- The "almost" final version (picture no 4), that we'd considered final till
- The virtually final scene (picture no 5) came along

(The picture can be seen here:)
http://www.farawaystudios.com/gfx/5thuig_dev.jpg


But even that one underwent some changes and therefore you can see the  
real final scene (for now :-)) in full resolution here:

http://www.farawaystudios.com/gfx/room5.jpg