Senior Project

2009-05-03T15:16:00.000-07:00

After a long, painful struggle, we have come to the end of our senior project... Since this is the last post, I wanted to go over some of the features we have implemented, as well as functionality that we just didn't quite get to. Sadly, most of the second semester has been implementing the battle system (which isn't done). However, there is a very nice, general framework in place for extending this. Pretty much all you can do in our game is walk between three different scenes. Nothing happens in any of these scenes, and you can optionally enter a battle. As of now, the enemies do nothing, so it's quite an unfair battle. Conversely, you can never hurt the enemies, so maybe it is fair. Nonetheless, you can attack, move, and use one skill. Also, the enemies don't do anything, as of now. They just stand there and let you beat them ceaselessly. Obviously, I would've liked to have implement the entire battle system, but it was a little too ambitious, like most of our objectives. But overall, I am pretty happy about the amount of progress we have made, and plan to continue to work on the project.

2009-04-01T13:34:00.000-07:00

I am glad to say that the inside of the Atonian Tower is complete. So now we are up to 3 1/2 scenes... I say 1/2 because there are still a few things I need to do to the other scenes to completely finish them up(i.e. add a few planes for bounding purposes, add a few planes with pictures of connecting scenes to make it look nice). Derek, in all his programming glory, started, finished, and rubbed in the fact that he got the overlays for the attack text working(sort of? He will need to explain what is up) Sadly, this means all that I really need to do is finish this god-forsaken GUI. I will get the GUI done, this I promise, but the quality may be lacking. It is quite ridiculous how much trouble such an easy task has given me. Until next time.
-Andy

2009-03-17T17:34:00.000-07:00

On my side of the spectrum, I have both bad and good news. Bad news being that although I have made a little progress on the GUI, I have become completely unable to motivate myself to work on it any longer. I plan to have it finished for the final product(though I am quite sure it won't happen for one reason or another). The good news is that I returned to the one part of the project that I have made any actual progress in: scene creation. I have worked on the inside of the Atonian Tower for one day, and have it almost completed. I feel that I should have never stopped modeling(I know... I'm a programmer, but modeling is so easy) and just had a lot of scenes to wander through. But alas, I did not. Also, unlike Derek's previous blog mentioned, we are actually going to implement overlays to display the battle text(i.e. amount of damage done). I plan to get that started here soon and should have no real problems implementing that since it was one of the first things we covered in Ogre. Hopefully the next time I find myself blogging I will be able to tell glorious tales of getting the Atonian Tower done, and with luck the overlay. Until then.
-Andy

For the first time in millennia... Progress

2009-03-16T12:05:00.000-07:00

After removing ODE from the project, I have broken down many of the hindrances that have plagued our project from the beginning, now that the effect radii are working properly (movement radius, attack radius, skill radius, et cetera) I have created several new classes: a base class called Action, from which Spells and Items both will be derived--this allows for a great deal of polymorphism, and reduction of redundancy, in that, instead of determining whether the character has used an item or a skill, just say that the character used an action, then have the subclass itself implement the details. This eliminates the need for a bunch of if-statements, and makes the code a lot cleaner, in general. I have the targeting working, now I just have to implement generic targeting based on the radius of the selected skill (which, with this scheme, 'attack' just ends up being a specificity of a skill--albeit one that has a small radius, a single target, and no mp cost). As of right now, I am working on some of the particle effects, and researching how to get text to appear (without using an overlay). So yeah, things seem to be on track... until the next error. :)

Massive Overhaul

2009-02-23T11:34:00.001-08:00

Okay, we've finally reached that great, inevitable pinnacle in our project where we've realized that everything up to this point has been pointless, and existence itself is simply a bleak sphere of nothingness dropped into an infinite void. All right, so maybe it's not that extreme: but we have realized that our inclusion of ODE as a means of testing collision detection is overkill and has led to many more problems. I have taken out the collision detection that utilizes ODE, and instead, reverted to using Ogre's bounding boxes, which should've been done all along. However, if I had foreseen the problems resulting from ODE, this would've been the clear solution all along, but, no one's that smart. Things are back on track now, expect a decent update next time around.

2009-01-26T13:58:00.000-08:00

New update that will absolutely, positively not surprise any of you. We are no longer doing the overlay GUI. I have just finished compiling, building, and linking CEGUI(Crazy Eddie's GUI) into our project. Ogre already supports CEGUI so there is nothing we have to do on the side, and it already has support for scrolling lists and other tools that we need. Truthfully, we should have used this library to begin with, but stupidity and curiousity took over and we decided to use something new.
Naturally, I don't have much to talk about because the past few days have been filled with frustration from the Overlays, deciding on whether or not to switch to CEGUI, and then lots of frustration with getting CEGUI to work. Despite my past experience with this library, I, naturally, had troubles building and compiling. Next time I promise to have pictures of a semi-awesome GUI, or atleast the makings of one. Hopefully I will have that working by Wednesday so that you devout fans will have something to gawk in awe at.

The system of the battling

2009-01-21T12:17:00.000-08:00

Ok, I know it's been awhile, and I'm sure everyone is just achingly, horribly consumed in bittersweet anticipation of what wondrous improvements, bug-fixes, and upgrades we will provide this time. As you know, I have been working on the battle system since the end of the previous semester. So far, I have implemented the following:

The game loop is functional, and supports moving/attacking.
The main character can move within a sphere defined by his movement radius.
The main character can choose a target (buggy, but somewhat functional)
Enemy/Character attributes are loaded from a file when a battle commences

Now, I'll go over (briefly) what I hope to accomplish, and how the battle system is to ultimately work. I have always thought of it as a hybrid between Final Fantasy Tactics and Shadow Hearts. Combat is turn-based, with skills requiring a certain charge before executing (for instance, an attack with a charge of 1/4 will charge in 4 turns, thenwill execute. There is a bonus for chaining these types of attacks. If another character--on the subsequent turn-- uses a skill that has a charge of 1/3, and they both hit their targets, more damage will be dealt.) Every attack/skill/item will have a radius/area that describes the type of attack. Some are spherical attacks that can hit multiple targets, others will be rays, and some still, may hit enemies and allies alike. Strategically planning attacks will be the key to winning battles, as we hope to make the battles challenging, but fun. I am currently working on a damage formula, as well as the display for when a character is actually attacked, rather, the displaying of the integer that floats above the character for a bit after sustaining damage. Hopefully, this makes the battle system, at least conceptually, a little more understandable. Hopefully, I'll have a screenshot or two for next time.

2009-01-12T12:06:00.000-08:00

First off, let me apologize for the lack of updates, but today is the dawn of a new semester, which means that we are going to post updates on a daily basis... provided that we actually make updates to the code on a daily basis.
Secondly, allow me to indulge you faithful fans with some new information that you can only get right here--We have decided to start from scratch on the GUI, and after much scrutiny and debating we have chosen to recreate it using overlays. They are quick, rather easy, and with enough care they can even become semi-aesthetically pleasing. With the Navi libraries giving us so much difficulty, and with the further hindrance of our laziness, we haven't much to catch up to, so hopefully it won't set us back too far.
Thirdly, we have decided to each update daily. I will update our esteemed fans on my doings, and he his. We feel that writing about what the other is doing is depriving this wonderful blog of the true details behind the picture, seeing that we, ourselves, might not understand what the other is doing.
Until tomorrow... adieu
-Andy

Late Update, A Little Progress

2008-12-17T09:01:00.000-08:00

Within the past month or so, we have made a little progress. Once again, there is nothing that we did that we can really show because most of it is behind the scenes and we have no new art. However, as you can see to the left, a little bit of progress on the GUI. Let it be known that the ninja profiles are purely temporary. We are using them simply because they match the temporary ninja character we are using.
Right now the library that we are using to create the GUI is starting to upset me, so the next update, which will hopefully be at regular intervals throughout break, don't be surprised if I mention that we are using a new library, probably CEGUI.
It should also be noted that we have the scene transitioning completed. Not only does it work, but it works quickly. It jumps directly to the next scene, and although you can see a jump from one scene to the other, there is basically no load time on the all but the biggest maps.
The collision detection is still giving us some trouble and as for now we have decided to let it go. We have a theory on what the error is, and even know how to correct this error but since we are so incredibly sick of this error, we have decided to give it up for now and come back to it later.
One last major update that we have made is that we have started on the battle system. We have the random battles completely working so as you walk throughout the map you have a chance of an enemy popping up. This enemy doesn't do anything for now, he simply stands there and allows you to walk around. When an enemy pops up on the screen, you have entered a battle scene, and that means that the controls change to make battle possible. Right now a sphere surrounds your character that shows the distance that he can walk in any direction, and you have the ability to move around in the sphere. Your character is stopped when he comes to the end of the sphere, so battle movement is completely done.
Until Next time

-Andy

Progress.

2008-10-17T00:41:00.000-07:00

Thought I'd take a moment and elucidate on some of our actual progress and milestones (besides setting a new personal record for most errors obtained in one compilation), bringing to light a tragic aspect of programming--most of the time, you have nothing to show for most of the programming you do. Anyways, in rough chronological order, our current functionality along with percentage completed:

Story 99.9%(.1% reserved for if we actually ever get around to telling it ;)
EarthSculptor/Blender Scene Creation 100%
Pointer Hash Map 100%
Models 1% (ha)
Per-pixel Lighting, Bump Mapping/Shaders 100%
Scene Class/.scene files: Exporting/Loading/Unloading 100%
Dynamic Scene and Entity Management/Classes 100%
First Scene 99% (Might add a thing or two)
ODE Physics 100%
Area (A graph of scene objects) Resource Loading 80%
Second Scene 99%
Trimesh Loading from file 100%
Character/Enemy Class/Character Attributes 10%
Collision Detection 75% (VERY Liberal estimate :)
Third Scene 80%
Scene Transitioning/Exiting 75%
Input Manager/Game State Manager 75%
GUI 5% (Hey, a window pops up, at least!)

Finally, a tentative list of features we haven't even touched, but hope to have implemented within the coming weeks/months/years/millennia (in no particular order):

Battle System (!!!)/Enemy AI(!!!!!)
NPC classes/Basic Flocking AI
Skill Tree
Bestiary
Indoor Rendering/Transitions
GUI/Menus/Status Screen
Saving/Loading (eek)
Event System/Manager
Cut Scene Manager (Maybe?)
Particle Systems/Skill, Magic Effects
Sound Manager/Music
World Map
Experience/Leveling System
Overhead Map/Viewport (Maybe?)

Well, I guess that pretty much sums it up--stay tuned!!! Or else.

-Derek

Progress?

2008-10-02T11:10:00.000-07:00

As of now, we have completed nearly half of the first area--slow progress, to say the least. However, we have spent a lot of time fine-tuning our collision-detection, researching resource loading, working on shaders and materials, and abstracting scenes into a higher-level area class, which will be responsible for the dynamic loading and unloading of scenes, and the respective textures, materials, and 3D meshes that they will use. All the while, we are gaining experience with how to efficiently create scenes, answering questions like, which objects can the character collide with, how tall should the ground be so that the character cannot step on it, et cetera. The better we are able to deal with these problems now, the easier it will be to rapidly develop scenes later on. We expect the first area to be done within the next week or two. After that, we'll begin working on the battle system, and character/enemy models and animations. I guess that's it for now.

-Derek

2008-09-17T11:16:00.000-07:00

I am glad to say that with the exception of a very few small meshes that will do nothing more than make it aesthetically pleasing we are finished with the first scene. As you can see from the screenshot we have successfully captured the bleak and desolate atmosphere that we were trying to generate throughout the game. Also, you may realize that we have decided to go away from cel-shading and stick with per-pixel lighting and bump mapping
We tested, tweaked, and finally decided on the colors of everything as well as the implemented bump mapping on all objects. We also finished several smaller things that bring the scene together: implemented fog, finished the background images, finished the skybox, etc.
One of the two significant things that we have implemented is the TriMesh, which as mentioned before, allows the character to walk on the terrain and take into account the height. Basically it allows the character to walk up hills while being able to calculate if a cliff is too steep to walk off, or walk up, with very little extra work.
The other major addition is the ocean that we added. The picture does little justice to the actual water, but it still shows that it is beautifully done. It is simply a plane with a lot of magic working behind scenes. We made a cube map of the sky box, and the shader then uses those images to create a type of reflection, while adding waves and noise to make it more realistic.
With these few upgrades and the slow but steady progress we are making we have become very excited with the results and potential of this demo. Since most of the models for the rest of the scenes are already done, because we plan on sharing models between scenes, we predict the next few scenes to go rather quickly. Hopefully by the next post we will have a few more scenes done, maybe even the Atonian Cathedral.
-Andy

Slow Progress

2008-09-10T08:56:00.000-07:00

Hopefully we will begin to post on a regular basis once we actually get the project on a steady track of progress, but until then we will continue to apologize for our lack of updates--Sorry for the lack of updates!
Despite a few tedious troubles that we have had to overcome at the onset of our project we are proud to declare that we have made a small amount of progress.
First we have found, fought and triumphed over a nasty bug in our .scene parser which now allows us to upload scenes that we have constructed in Blender. The first scene is complete and now that we have the ability to load the scenes I plan to work on the other maps of the first area.
Derek is going to attempt to get the Tri-Mesh working on our ground meshes so that our characters and, if necessary, other objects will be able to interact with the varying height of our terrain. It shouldn't take much time since we have already implemented it in other projects, but truth be told everything is a guessing game at this point.
Next time I hope to be able to put up some screen shots of our progress to instigate a little bit of interest in Bevontule.
-Andy

Bevontule

2008-09-03T10:34:00.000-07:00

Hello, and welcome to the blog for our upcoming 3D RPG Bevontule. The idea of the game is based on a short novel I wrote several years ago--upon realizing that it didn't make a great book, we decided that it could make a good RPG. The game will feature full, 3D graphics (possibly cel-shaded) rendered in Ogre, along with an innovative combo-based, tactics-style combat system. At present, we plan on releasing just a small sample of a demo (which will contain several playable characters and locations), with the possibility of extending it to a fully-functional game. Stay tuned for updates!
-Derek