Week 7 - Wrap Up

Well, this is our final week of work and aside from one or two little points, we are ready to cook our game's alpha version.  Everyone has all of their parts done in a working form except for the music.  As of when we parted ways today we were still pending the last couple of pieces to have something to put in but things were complete.

That said, going into our winter break I have some bugs I have to work on.  These include:  orienting the crossbow properly to the characters hand, changing the projectiles over for the crossbow, eliminating the randomized nature of the melee weapons attack, and start getting the triggers in game to implement some elements we didn't get time for in the 8 weeks such as spots where cinematic's would go.

That's all for this week outside of cooking and the only thing we have left is to present at our developers talk on Monday.  I'll probably be posting up a couple extra's during winter break but otherwise the real work will start up again at the start of January.

Week 6 - Cram Time

We've had a busy week so far.  Our group goal is to try and have a playable alpha by next week Monday, and considering that we had only the main character and the first level fully complete by last week, this was a bit of a chore.  But it's only Thursday and there is much that has been accomplished.

On my front, I have managed to get the light detection finally working, although in a much different approach then initially planned.  The original concept from the first weeks was to have it set up so the more light that is on you, the easier you are to detect.  I went through trying to find an angle between the light and the pawn, sample code online from the Epic forums, using the outer and inner cone angle/hit boxes and none of that had worked.  The final solution was done in Kismet and it was done using trigger volumes.  One was placed under every spotlight and rigged up to an Actor Factory.  This will spawn two bots every time a light is crossed.  This creates a much more dynamic game where stealth will make the game notably easier and be almost mandatory at parts.

That aside, I also have almost finalized the ladder volumes.  All of them except 3 work perfectly and there is no longer any issue with the character falling into the foreground/background.  I have also started and almost completed work on the weapon code this week.  The ranged weapon was simple and is almost fully implemented.  I still need to change sound effects over as well as reference the model when it is completed, but otherwise the crossbow is complete.  The melee weapon is a bit easier, but having a slight problem.  There appears to be a random variable somewhere in the shot that results in the invisible bullet going anywhere from right in front of the character to 30 feet out.  This is a bit of a spread and almost too much so I'll be conducting research on the issue and hopefully have it fixed come next week.

My parts aside, everyone else has made amazing progress.  Our first level is fully textured, lighted up with the light detection and has the HUD implemented.  The title screen is almost done and is undergoing a last couple of tweaks before it's in.  Our crossbow and sword model are both nearly completed and music has had its rights requested for use.  AI is practically done and the animation tree was nearly completed last I saw.  And to top it off, the main character is fully completed and being added in, the enemies have a few kinks being worked out and the last couple of items are being made and added in.

All things told, it looks like we'll have a passable alpha as of next week, and with next week added in we will need to remove all sounds that we don't want that are inherent to unreal.  If we take care of that, work out some of our last few bugs and finish up a few more things, we will have accomplished all of our goals for the first 8 weeks of progress and will leave off on a great note.

Week 5 - Ladder Volumes

Well, last week was a serious pain.  The light detection seemed like it was almost complete but due to a minor last minute error things didn't work.  In essence, the code wasn't handshaking properly with everything else meaning my code wasn't getting called.  This lead to a whole load of other issues and resulted in this past Monday/Tuesday working things through with my professor to get things done.  The good news is that a valid method has shown itself so by next week, things should be done with the light detection no matter what.  For the time being, I'm focusing on the ladder volumes and kismet triggers.

With the map I have, the ladder volumes have been fairly easy.  We don't have very much to do in the first map to enable to character to climb everything, so I actually have the ladders almost done.  Just a last minute bit of tweaking.  The only issue I'm having with them is that the player can fall of the sides sometimes on some of the smaller gaps.  This just needs a blocking volume that either myself or Vince will be putting in.  As for the triggers, they all need to go into a part of the level that we haven't designed yet.  It will involve a cutscene/enemy spawn trigger and a trigger to let the character unlock doors.  Both should be fairly straight forward and can't be fully done at this time.

All told, this week hasn't been so bad and it's been a huge relief knowing that the light detection is being taken care of.  By the end of this week, I should have half of my work for this done in about half the time meaning I'm still on track for getting everything accomplished.  As for the rest of the group, the HUD/menu's seem to be done/almost there.  The first level is only missing the end section and the challenge parts for items we haven't yet done/are finishing.  It also needs textures put in.  AI is almost done last I heard and hopefully the weapons are nearing completion.  Best of all though is our main character is almost done and ready to be used.

Week 4 - Light Detection

Last week I was unable to finish the light detection and so that has continued into this week.  Also, as soon as I can finish it, I'll also be starting work on ladder volumes in our first level.  That said, I've made progress towards completing the code.  With the help of my teacher I've changed how I'm going about it.  Instead of running a line of sight algorithm, I'm putting a hit box on the light and running a check to see when the character crosses the beam.  That said, the amount of levels of light detection will likely be dropped back to only a yes or no.  I've been having difficulties getting a spotlight to register anything and every example I've seen used a point light.  That means there is no easy way to have multiple hit boxes running within the light.  Not to mention I would most likely need a secondary light object to run a second hit box.

As per the past big, we also have no issues within the group.  Everyone was continuing their sections from last week so we should have a lot done come this week.  That's it for this week and come next week level 1 will be fully climbable.

Week 3 - Game start

This is week 3 and we have officially started production on the game.  My task for this week is to work on the lighting aspect of the stealth system, or make sure that the character is easier to detect if he's standing in brighter light.  I've been struggling with this and it's been somewhat frustrating.  But I think I have a very nice way of doing things, that takes care of multiple facets.  I had started with trying to replicate what other people had done, namely they all had their code trying to trace between the pawn and the light to see if they were being hit.  I wasn't able to get this to work at all.  What I decided on makes is to utilize spotlights and their inner/outer cone properties.  Now, all of this is theoretical at this point as I haven't been able to code it in or find the variables to work with it in unrealscript, but the idea is there.  Using this, I can make the outer cone at 50 degrees and the inner cone at 10.  The character will then have 4 levels of light detection, 1 being not noticeable and 4 being anything can see you.  I will make it so the code performs a check between the light, the character, and the ground below it to get an angle.  This is compared to the angle of the cone.  If he is greater then the outer cone, he's outside the light and at 1.  If he's in the outer half of the cone, or 30-50, then the character is at 2 detection. In the 10-30 range he's at 3 detection and standing in the heart of the spotlight, anything under 10, will result in being a level 4.

I haven't started coding yet because I'm still trying to learn the last bit to make sure I know how to do everything, but this method should garner the desired result.  From this base, It's a matter of working it into the AI that if the player's detection variable is set to 1-4, then the AI will detect him sooner or later.  It also means stealth will almost always reduce your detection variable by at least 1 level.

That said, we still have no issues in group.  Things are going smoothly from what I've heard in everyone else s portion.  Everything from last week was done and turned in when it was needed.  This week, I can see my section bleeding over to next week, but aside from that things are looking up.

That's all for this week.  Next week should be the finishing of the stealth elements and possibly the implementation of our first weapon or two.

Week 2 - The Game Design Document

It's been a week later and at this time we've started getting into the game design document.  We actually have a lot of work done today, more then I would have initially thought.  All but one person has their section in and it all looks pretty solid.  We have a good grasp of what's going on and ideas have started to take form in ways that will be doable.  Pretty much the only thing we don't have at least a basic idea for is the characters/story section fully written out, and we have at least a rough idea of what most of these are already.  All in all, things are looking good as of now.

Since last week, a couple of changes were also made.  The biggest thing is we've decided to make this into a 3D sidescroller.  This has given us some new issues, but simplified a lot of things.  At first I wasn't 100% sure on it and where things could be going, but after thinking it through and talking about ideas, it seems like a great idea while still keeping the original feel.  We have also started research on what needs to be done.  Just from the bit I've done so far, things don't seem quite so bad.  Some ideas of how teleportation could be worked have come to mind and these don't seem so bad.  Also there are some tutorials that explain how to do some of the things were looking at, enough that we should be able to make a sizable dent in the project.

Those things aside, there hasn't been any issues in group.  Since last post, everything got done fine and the new deliverables all seem to be getting turned in as we need them.  That's all for this week and I'll be sure to have another update next week sometime.

Week 1 - The High Concept

This is the first post in a series of 16 that will be documenting progress through the midterm project at DeVry University.  This is week one and as such we are only just getting started.  We have an idea set down and the basic gameplay thought out but we are still working on getting things set up.  Due to be turned in is the High Concept which is mostly a rough overview of the entire game.  Everyone is excited about the project though and we have been getting ahead of ourselves with ideas.

The idea we have is a steampunk universe where you play a mercenary character specializing in stealing and killing jobs.  Discovering a nefarious plot, he has to save the city lest his life style be ruined by events.  The game will be a short, 3 level series which unravels the story and lets you use a series of unique and creative items to get through each level.

Now, this is only week one so we haven't had any issues yet.  Group has started up nicely.  Everyone traded information and we settled on an idea very quickly.  In the future though I see a couple issues.  One is cutscenes.  We'd like to incorporate a few and it may be difficult for our modelers to get our models mouths to move.  We also have a light system used in stealth that could prove problematic.  The only other thing I see us having potential issue with is some mechanics that might include things like teleporting.

That's all for this week.  I'll be on again next week with updates on the Game Design Document and further ideas we are looking at.