Point of Stealth (Marcus' Entry)

Summary:

Name subject to change

page 1 of 212next »

Point of Stealth (Marcus' Entry)

Marcus42 30 May 2016 03:54

The Game:

Point of Stealth will be a functionally 2D (but technically 3D), relatively casual, puzzle, stealth game in which the player must outfit operatives with gadgets before guiding them through a heist in a simple point-and-click control scheme. Each level will be a building containing a vault/safe with various security patrol routes, security cameras, and obstacles that must be gotten by. The player will be able to win by "solving" the puzzle, AKA figuring out a plan of action, in one of several possible ways and successfully carrying out the operation, requiring creative use of the available gadgets or taking advantage of having more than operative to, for instance, cause a distraction that lures the guards away from the safe.

I have a lot planned that I could and would like to do for this game, but that's basically the minimum of what I want to accomplish during this competition. Hopefully I'll get most of what I've mentioned already done relatively early in the competition so that I can work in some more of my ideas! We'll have to see how Unity and I get along, though, before I make any predictions, and I've never tried making a puzzle game before, so … Also, I'm not sure how crazy I am about the title I came up with - opinions?

Re: Point of Stealth (Marcus' Entry)

Marcus42 30 May 2016 04:06

Sprint 1

Goals:

- Find some 3D models for both the player's operatives and the basic security guards.
- Animate both of the aforementioned models to have basic stand and walk animations.
- Model a simple building to serve as the test level.
- Start the project in Unity, create a repository for the code, and make a basic, functional AnimatorController for the operatives.

Stretch Goals:

- Create the test scene in Unity and set up an Operative prefab that won't fall out of the screen, at the very least.
- Add a script to the Operative prefab that allows for horizontal movement to the mouse's x-coordinate upon a click of the mouse.

Comments:

- Obviously, this sprint is largely artwork related. I'm not really sure yet what my work speed is going to be like this time around, so my goals may be either way too little or way too much. We'll see.

Re: Point of Stealth (Marcus' Entry)

PiscesMike 30 May 2016 05:12

Sounds like you're off to a great start!

---

"May the mercy of His Divine Shadow fall upon you." - Stanley H. Tweedle, Security Guard class IV, The League of 20,000 planets

Re: Point of Stealth (Marcus' Entry)

Marcus42 31 May 2016 01:05

Today has been very productive, but the goals I outlined in the post above are for this week - I didn't do all that in Sprint 0 (which I actually just spent planning my game), but I am going to do that stuff in Sprint 1 (starting today). Sorry for the confusion.

Re: Point of Stealth (Marcus' Entry)

Brett208 31 May 2016 04:03

Hey Marcus,

Point of Stealth is a good, short, exciting name. I don't think it really conveys that you are working with a team of agents or trying to break into vaults though. Although it does convey a sense of stealth actions and calculated moves, so I think it probably hits close enough to the mark.

Spent 5 minutes brainstorming some ideas for your game title based on your description. Titles/Studio Names are always the hardest for me, so I thought maybe a little outside help might stimulate some of your own brainstorming. If not, feel free to ignore them. :)

Henchmen Heist
High Security Heist
Vault Busters
Vault Hacking
Vault Crack
Security Breaker
Henchmen
The Final Heist

Re: Point of Stealth (Marcus' Entry)

Marcus42 31 May 2016 16:02

Wow - thanks for spending the time thinking those up! I'm glad to know that Point of Stealth is already an okay name (and it's kind of a pun, too, because all of the controls will pretty much be pointing and clicking with the mouse), but I will definitely take your suggestions under consideration.

Re: Point of Stealth (Marcus' Entry)

Marcus42 06 Jun 2016 17:53

Sprint 1 Recap

Tasks Accomplished:

- Found models that should be all that I will ever need for this project (as far as human characters go, that is).
- Rigged and colored two of the models to function as the basic operative and security guard (subject to change - the operative looks more goofy and less spy-like than I'd intended, but maybe that's just me …)
- Created walk, stand, sneak, and idle crouched animations, all of which can be applied to every model easily (thanks Unity!).
- Began the Unity project, set up a test scene, organized the assets folder, and set up prefabs for both the operative and security guard.
- Wrote two enormously simple scripts so that I can control when the two characters in the current test scene move or stand still, as well as to allow the camera to scroll to the side.

Tasks Not Accomplished:

- I didn't quite finish the test level. I've finished modelling it (which turned out okay, but at least good enough to serve my purposes right now), but I still need to finish texturing it (which so far … looks sort of bad. Hopefully it'll be better once it's finished and in the game).
- I also didn't get to my stretch goal of scripting the characters to move horizontally to wherever the mouse clicks, but I did get an easier version of this done, so I'm okay with that.

What Went Well, and What Didn't

- So far, I've been very pleased with Unity. I haven't done much development inside it yet, but I'm thrilled with how it handles animations, 3D models, and all that. For example, rather than needing to do the same animation separately for several models and then create texture atlases for each one, all I need to do is apply a compatible animation to whichever model I want to, and it works pretty well. Plus, the AnimatorController is really easy to use to set up how the animations are played, transition, and blend into each other, stuff that I've typically done with many, many lines of painstakingly written code. All of this stuff, and more, hugely simplifies my animation workflow, which has been a big relief. So yeah, that's exciting.
- What hasn't gone the way I've hoped has to do with the change in perspective. Because this project isn't 2D and from the top down, all of my artwork is harder to make look better than it really is (and I say this because I really am a bad artist, and it's amazing that things have typically turned out as good as they have in these competitions). What looks like a really nice walk animation from the top in 2D is sort of stiff when you're looking at it from a more realistic perspective, which is not a huge problem, but it does mean that I either need to try a bit harder or just be okay with mediocre animations. What I ended up doing for the animations I've already made is that I filmed myself doing what I wanted to make from as close to the right perspective as possible, then set the animation to simply cover my limbs and make any adjustments I need to afterwards. Doing that worked pretty well, but there will likely be some animations I won't be able to act out in a way that provides a useful reference, so we'll see how that goes in the long run.

Screenshot/Video/Demo:

- In the video, you can see the animations I've made so far (including the nice transitions), the interchangeability of the animations with the models, and one of the AnimatorControllers I've set up in action - as I change the parameter for the controller on the left, you can see which animation is playing, where it's transitioning to, and stuff like that. It's pretty cool. I'd probably recommend you watch fullscreen, though, and be aware that I quickly grabbed and threw a song by Kevin MacLeod into the background. Also, know that there are several problems with the animations I've got that I still need to fix but am already aware of, so … Yeah, this is most certainly all a work in progress. ;)

Re: Point of Stealth (Marcus' Entry)

mdjdejesus 06 Jun 2016 19:18

I like these character animations (specially the agent.) It reminds me of Counter Spy, which seems to have the sort of 2.5D perspective you seem to be going for.
—Javier

---

—Javier DeJesus

Re: Point of Stealth (Marcus' Entry)

Brett208 06 Jun 2016 20:48

I'm jealous of how comfortable you are at programming 3D assets and 3D character movement. Well done, especially for 1 week of work!

Re: Point of Stealth (Marcus' Entry)

Marcus42 08 Jun 2016 17:16

Thanks, guys!

Javier, I am definitely going for a 2.5D kind of perspective. 3D looks cooler and is probably easier to do in Unity than 2D is (at least, that's been my impression), but I really don't know much about coding a 3D game and need to stick with 2D mechanics. Hopefully it all goes okay!

And Brett, while I'm glad that I look like I know what I'm doing with Blender and 3D animations and stuff, honestly all that I've been doing is slowly feeling my way through the few things I want to be able to do during the 4+ years that I've been using it. By this point I can do everything I need to in order to make 3D art assets ready for a game, but I really don't understand much else of what Blender offers. Plus, Unity has been handling almost everything in this project so far - really, I've barely coded anything, which I'm excited to get to this week, because I certainly prefer programming to artwork!

Re: Point of Stealth (Marcus' Entry)

Marcus42 06 Jun 2016 17:56

Sprint 2

Goals:

- Finish the test level and set it up in Unity.
- Be able to select between operatives and direct them with clicks of the mouse.
- Set the security guards to patrol certain routes and be able to spot any operatives in sight.
- Create a jog/run animation.

Stretch Goals:

- Allow operatives to traverse the level vertically via staircases and ladders.
- Have the security guards pursue any operative that is seen.

Comments:

- This week I'm wanting to do a lot more coding and gameplay related work, which I'm excited for but also uncertain about. I'm already close friends with the Unity Documentation manual, but I haven't figured out yet how I'm going to do a lot of this stuff. We'll see!

Re: Point of Stealth (Marcus' Entry)

Marcus42 14 Jun 2016 02:19

Sprint 2 Recap

Tasks Accomplished:

- Basic operative control. They can be selected one at a time and directed around the level with the left mouse button, and holding down the left shift key allows for multiple to be selected at the same time.
- Better camera movement. It is now able to move in all four directions via the arrow keys and zoom in and out with the scroll wheel, which actually adds a lot for being such a little feature.
- Extremely simple security guard patrolling (just walking back and forth between two points).
- The test level model is finished, added to the Unity scene, lit, and actually looks pretty good!

Tasks Not Accomplished:

- Heh, no jog/run animation.
- Didn't achieve either stretch goal.

What Went Well, and What Didn't

- Overall, I think this sprint went okay. I made a fair amount of progress, but not as much as I wanted to, which I think is largely due to a bit of lacking motivation early in the week. I'm not totally sure why, but on Monday, Tuesday, and Wednesday last week, I wasn't putting very much time in each day - I think part of the reason may have been that I needed to make another animation for turning around before I could get to coding, which I didn't particularly want to do.

- Once that animation was made, though, it has still proven to be an issue. I'm not sure why, but it's not turning the characters in my game around all 180 degrees. In both Blender and the Unity animation preview, the animation clearly turns all the way around, but once I apply it to a model in the game I'm only seeing about 120 degrees. I've ended up spending a fair amount of time trying to figure this out to no avail, and I have yet to get any help from my Unity Answers question, so this has been sort of frustrating. I needed to do something (because the characters should only be walking along the x-axis and so I need them to be straight), so I ended up writing a script that sets their rotation at the end of the turn animation to make the correction. It works fine, it just doesn't look so great.

- I did eventually get to the coding, however, and that turned out very well! The mechanics I implemented are rather simple (just clicking around to get basic movement along the x-axis is really all that it is), but the techniques for working in Unity are pretty different and I needed to figure out how to do a lot of stuff, including raycasting. I've never done anything with rays before, much less in 3D, so that took a bit of time to learn about and get working (to satisfying results, though!). I also kept coming back to that particular script to improve it, particularly when I gave the camera the ability to move along the z-axis for the zoom function and needed to change how the mouse's position is translated into a world point for setting movement coordinates more accurately.

- So far, I've been very pleased with the documentation for Unity. In addition to the manual that I was using during Sprint 1, there's a Scripting API that's also proven to be rather helpful several times.

- Unfortunately, trying to make a video today definitely didn't work out well at all. The software I'm using was decidedly not working properly, so I changed my mind and took a few screenshots to show you all instead, but then it took an absurdly long time to even just upload the pictures to Google+ (although I'm not really sure why), so there's been some problems there, too. All's well that ends well, though, because I do have some pictures to show you guys!

Screenshot/Video/Demo:

-

-
I was also able to make a cool, semi-transparent material in the level that you can see clearly here.

Re: Point of Stealth (Marcus' Entry)

mdjdejesus 14 Jun 2016 16:14

Hi Marcus,

See-though walls look good.

I agree with you on the Unity documentation; I've been very pleased by how easy it is to find what you need.

Hope you get a response soon for the character animation issue. I can't really help; but, manually rotating around Y-axis seems like it should just be a temporary solution. My game is much simpler than yours, and I too had to implement some temporary solutions….I hate those =)

---

—Javier DeJesus

Last edited on 14 Jun 2016 16:16 by mdjdejesus Show more

Re: Point of Stealth (Marcus' Entry)

mdjdejesus 14 Jun 2016 16:49

I've been using the Unify wiki to supplement the Unity Docs.
Found it extremely useful (specially the scripts section.)

Here the link: wiki.unity3d.com.

---

—Javier DeJesus

Re: Point of Stealth (Marcus' Entry)

Brett208 15 Jun 2016 19:24

Marcus,

It looks really good so far. I like how the security guard looks a little overweight. I'm interested to see how the 2.5D lanes of motions turn out!

-Brett

Re: Point of Stealth (Marcus' Entry)

Marcus42 14 Jun 2016 02:39

Sprint 3

Goals:

- Complete basic functionality of the correct movement system. In traditional 2.5D fashion, I'll probably have several "lanes" of movement for the characters to switch between to allow for passing by each other, but I haven't completely decided how I want this to work (much less how to code it … Oh boy).

Stretch Goals:

- Allow security guards to be able to see and pursue (by which I mean trail harmlessly behind) any operatives in adjacent lanes.
- Begin setting up a simple inventory system. (To be used later for the gadgets.)

Comments:

- This week is going to be hard, I think. I'm nervous about trying to set up the movement system, because it's more restrictive than what I usually do (free movement) and needs to be able to work well. None of the characters in the game are directly controlled by the player, so I need consistently working locomotion for the game to, err, not be horrible. Anyway, I'm hopeful that it will turn out well, but I don't think it's going to be easy!

Re: Point of Stealth (Marcus' Entry)

Marcus42 20 Jun 2016 16:13

Sprint 3 Recap

Tasks Accomplished:

- Set up a three-lane movement system. This works pretty simply, where periodically each character sends out rays along the three possible lanes (different z-axis depths) to see which are available and moves to the best one. The middle lane is the favorite, with the lane closer to the camera preferred after that and the one farthest away avoided most. This is still pretty rough, but works well for avoiding obstacles in small distances.
- Set up a waypoint system for pathfinding that is easily created and editable in the Unity Editor.
- Coded pathfinding by brute-force searching the waypoints (a la Dijkstra's Algorithm).

Tasks Not Accomplished:

- Neither of the stretch goals got done, but I have no problem with that.

What Went Well, and What Didn't

- Pathfinding has been going rather well so far. I'm not using a very elegant approach, but for as simple of a 'grid' as I have it's all that I need (and shouldn't lag at all, because I'm doing one step every frame rather than every step in one frame). I've only tried pathfinding twice before (Final Saber and elacc), but for some reason it didn't occur to me until this time around to learn about how trained programmers do it instead of trying how I think it should be done, so I'm no longer clumsily feeling my way through the level with rectangles and such. I was also able to set up a system so that I could manually place objects in the Unity Editor with a Waypoint script attached and take all of the information I need for pathfinding just from that.

- Unfortunately, I'm not very happy with Unity's debugging. I'm still getting coding errors right off the bat with Visual Studio's debugger, but anything less obvious will only print an error message and the problematic script to the console window, so I've had a rather difficult time hunting down numerous ArgumentOutofRangeExceptions or an object reference not being set to an instance of an object the last couple days. There were also a lot more of these than I ever expected to get after writing a big block of new code because of some of the different ways Unity works that are still throwing me off a bit.

- I also had to figure out value vs. reference types better. I knew of them, but have never run into a situation where I need to code something differently before now and it took me a few minutes before realizing what the problem was.

- Overall, I think this sprint went pretty well. I learned a lot and made some important progress, even if I didn't really finish any feature implementations (lane management is still rough and setting a path, much less following it, is still very much a work in progress).

Screenshot/Video/Demo:

- Most of what I worked on this week you can't really see (and switching between lanes looks horrible), but I did take a screenshot of the waypoints I've set up in the Unity Editor. Each of the white spheres is a child object of the level that I can move around and edit the waypoint properties of (like which of the other waypoints it's 'connected' to), but when the game is actually running they don't render and simply provide the information I need for pathfinding.

Re: Point of Stealth (Marcus' Entry)

mdjdejesus 20 Jun 2016 16:49

Have not had to implement any sort of path finding in a game as of yet.
I think I'm appreciating how tricky it might get if/when I do.

So, not getting the way points; are there way points on each lane? or is switching between lanes not dependent on way points?
For example, the bottom and top of the stairs are on different lanes.

---

—Javier DeJesus

Re: Point of Stealth (Marcus' Entry)

Marcus42 21 Jun 2016 18:27

Yeah, sorry, I probably should have explained this better - the lanes are almost entirely independent of waypoints. I haven't decided yet how to manage things like the stairs, which you mentioned, but switching between lanes is primarily just for avoiding other characters moving in the opposite direction or standing still, and the waypoints are for long-distance planning, so there shouldn't need to be much interaction between the two systems at all.

For example, I need the waypoints so that I can know to walk to the right and head up some stairs before going left to the goal (rather than walking left immediately and being unable to get up to the actual destination from there), and while the character is actually following that path he can switch lanes on the go if he needs to get around an obstacle.

And actually, I don't think pathfinding in 2D is all that hard (although definitely still challenging) if you're willing to make a grid of valid and invalid locations and base your approach on a tried and true algorithm. These three articles were enormously helpful for me to just realize ways that I could do pathfinding rather than making a system for genuinely looking around at the environment and somehow planning a complicated path from scratch.

Hope that answers your question!

Re: Point of Stealth (Marcus' Entry)

mdjdejesus 21 Jun 2016 21:08

Thank you for the info and the links. This will definitely come in handy in the future.

---

—Javier DeJesus

Re: Point of Stealth (Marcus' Entry)

Marcus42 20 Jun 2016 16:31

Sprint 4

Goals:

- Finish pathfinding and tie it in with the lane management system (pathfinding doesn't actually move the character yet and lane management is done unintelligently, so I need both to work together).
- Allow the operatives to traverse the level vertically via staircases.

Stretch Goals:

- Make ladders able to be used as well.
- Get pathfinding working with the security guards (which should only require catching up their controller script with the new stuff I've done).

Comments:

- It's hard to believe that if this was a normal competition, we'd be done after this next sprint. Crazy! Anyway, this week I'm hoping to be able to tie up the loose ends I've left from Sprint 3 so that I can start working on other things to round my game out better. I'm also just excited to have a pathfinding system finished and actually working! That's the hope, anyway. :D

Re: Point of Stealth (Marcus' Entry)

Marcus42 27 Jun 2016 16:27

Sprint 4 Recap

Tasks Accomplished:

- Made the looping animation for climbing up staircases.
- Fixed another issue or two with pathfinding.
- Added bounds to the camera's zoom.

Tasks Not Accomplished:

- Aside from fixing a couple of actual problems with pathfinding, I didn't improve it or tie it in with the lane system.
- Didn't finish the set of animations I need for making characters able to move up and down staircases, nor did I do the actual coding to support it.
- Neither of the stretch goals got done.

What Went Well, and What Didn't

- As I'm sure you can tell, nothing really got done. At the start of the week I decided to do all of the animation work necessary before moving on to coding so that I wouldn't need to stop part way, but doing the animation just did not go well at all. As I think I've mentioned here before, during this competition I've been filming myself doing whatever it is that I need to animate and then copy my movements, but I don't have any stairs in the house and I got kind of stumped on how to rig something simple to mimic it. I tried one day to animate it from scratch, but that didn't work at all, and then when I did find a way to rig something up and film myself, the video didn't have all that I needed to make a loop and I had to do some rearranging and extrapolating to get a good result.

- After that, I realized the quantity of animations I'll need to have a complete set, or else I'll have to figure out something clever to reduce the number. I've already got the loop for the operative climbing up, but I still need one for going down and one for getting on at the top and bottom from both the left and right (because the characters need to walk outside the lanes to the very closest or farthest point of the level to get on the staircase and then turn 90 degrees and take the first step before the loop can start). Overall, that's six animations (not even counting the ones I'll eventually need for the security guards to do the same things), and doing just one was a huge headache.

- So anyway, that basically robbed me of whatever motivation I had this week, and nothing significant got done as a result. Disappointing, for sure, but there's bound to be at least one week like this during a 100 day competition. Now, on to the next sprint!

Re: Point of Stealth (Marcus' Entry)

Marcus42 27 Jun 2016 16:40

Sprint 5

Goals:

- Figure out how to minimize the number of animations I'll need for using the staircases and then make all of them.
- Implement the ability to move up and down via staircases.

Stretch Goals:

- Tie in pathfinding with the movement and lane management systems.

Comments:

- This week I'd really like to get back on track, but doing so requires me to potentially spend a lot of time animating and I'm kind of worried about being able to stay motivated and push through. Man, I wish I knew someone who thought learning Blender and doing a bunch of animations for me for absolutely no money whatsoever sounded like a lot of fun, haha! And with this particular project, having a 'partner in crime' would be rather fitting! (Just to be clear, this is not an actual request for an animator, as nice as it would be. I'm just bemoaning the lack of an ideal situation, haha!)

page 1 of 212next »