Game Design Challenge (PC)

Hard challenges are what makes you best. I'm sure I'm gonna learn a lot!

 

Yep It feels crazy to quit a hobby, but I've got little time for other things. The bright side is that dice.com is fantastic for looking for work. 6-8 calls & emails a day, yo. Most of them from people who don't know what 'programmer' means and call me about entry level customer service positions, but eh.

There are enough calls now for me to be a little snippy, actually. "You say you read my resume? What experience or skill on there under 'programmer' indicates I'm a good candidate for sales or customer service?" I've made 2 recruiters say sorry so far But to be honest, if the job search goes on too long, I might just have to take some crap like that. If the work load is light I can focus on my hobbies!

I encourage you to succeed and show the world how silly it is to teach programming in college. 4 months of googling over summer crippled my college experience because I didn't learn anything new in school. I learned new things myself, but the professors spent the semesters covering things I learned in a week.

 

Well best of luck, I know Dice are always looking for skilled DX / GL engine developers.. Not sure if that's your bag?

 

@Tomnnn
Exactly! Exactly! You're the first person to say that! Here when I was new , 40+ or even 60+ programmers here told me that I'll learn basics of programming in 5 years , then in another 10 years to grasp concepts and then another 10 years to learn something. Gosh! Those were lies! I have learned C# so much in 8 months than I would ever learn in my college.
That's why I hate those people who say , "DUDE , I started programming at age 8 , that's why I'm a good programmer , you'll have to wait years to be as good as me" . Well I know how bull crap experienced programmer these people were. lol Starts GW Basic "Basics" at age 8 , calls themselves programmers. Basic language has absolutely nothing to do with programming required in Unity.

What they learned from age 8 to age 15 is literally equal to ONE youtube C# tutorial. Now I won't be fooled by those bastards , they literally discouraged me so much that once I even thought of getting out of this field. But now I know , programming is extremely easy , just requires your attention and interest. That's why I started this way before college.
Thank you!

 

@Haseeb_BSAA

Most fields in game dev require you to be a jack of all trades. Especially as an indie, you'd better get used to Art / Shaders / DX / GL API and coding..

Even when I was doing engine dev for large corporates, you needed a good eye for rendering.. Pretty much pointless coding shaders and sorting out lighting integrations when you can't tell if it's working properly or looks bad.

So understanding the concepts is half the work, the rest is a google away ..

Being a coding Guru never seemed to get anyone anywhere, especially in larger companies because they'll replace your ass with Junior's who can do the logic / will work more hours and get paid less for it. But if you're like a weed stuck in several places, it changes your stance.

 

Yeah , that's exactly what I'm doing. There is absolutely no way I can just learn programming. At first I found it so hard to learn 3D Modeling , Animating , Programming and of course learning Unity at the same time but I kept on watching tutorials for every program. I made a lot of mistakes , I asked a lot of questions. GOD! It has been so tough for me but I'm so happy I finally grasped it! I remember I used to watch blender rigging and modeling tutorials , it was because I was alone working on my project. All my friends left this field , some went into medicals while some went into other fields. But I was so passionate about it. I knew one thing "I HAVE TO DO IT"... And so slowly slowly I kept on learning. I started with downloading pre built models and rigging them. Then I started making my own. Now I make my own models , I animate them on my own. It's so fun! Definitely it's tough thing but I don't have a team or something and literally I want to work alone and learn as much as I can.
I'm glad I made it this far and I hope to learn more in the upcoming future Thanks for suggestion sir!

 

BTW , I started game development in 2013 for the first time in a software called "3D Rad" , it didn't require any kind of programming , only visual scripting like Unreal Engine 4, it had a few built in models too. It had Nvidia physics engine , so it gave me a very good environment to learn the basic concepts. I think the person who is completely new to Game Development should give it a try. I developed a racing game , very simply , it just had a loop of track , I named it "Sky Racing Extreme" , my friends loved it even though it was bad, they appreciated it because I was a Biology student and I developed it It gave me courage and passion to get into it. Also I had a very great gaming history , I always loved games. So day by day I got more interested in it and eventually it became my ambition!

 

I like to program. The details don't matter. IT302 showed me I could pick up new languages and technologies in as little as 3 hours. I made a little game in javascript, changed the click events to touch events to get around 300ms button delays on mobile, then used cordova to build the app into a mobile game. Start to finish 3 hours, which was the class period.

@Haseeb_BSAA I don't remember when I started programming. But I do it several times a week consistently since my 2nd year of high school. Besides minor things, languages are mostly the same to me. The logic works out the same except for value vs reference things in C++ and Java and that annoying thing python does with evaluating booleans.

If you're interested in unity and C# check out quill18creates on youtube. Best tutorials I've seen on the subject and I recommend it to everyone. Good for you for not being discouraged, programming is as easy as thinking. So it makes sense why some people struggle with it

 

I started building a boat to go on my water. I think I'm just ignoring the sci fi theme, unless I do some cop out like "there was this big fancy scientific world, then they all blew up, and now you are back to building rafts".

I'm currently facing significant performance issues. Apparently 400 spring joints between 200 ridgidbodys is just too much for the engine to handle. And that's just a small fraction of the scale I was intending to build, initially it was 10,000 rigid bodies. So I may end up using a custom physics solution for this.

I suppose to meet the intent of this challenge I'm also going to have to learn to model. I have the feeling @ShadowK will toss me out if I try and use Unity primitives.

But one thing at a time. First fun mechanics. Then a acceptable performance. Then models and textures.

 

@Tomnnn Thank you for suggestion , I'll definitely check that out!

 

You could have space ships occasionally pass overhead lol. Maybe give your player 1 human arm and 1 robot arm, which functions as your crafting table.

 

Week 3: "Arrgh! My Back!"
I really need to catch a break. I pulled a muscle in my back. As a result I lost three days that could've been gainfully spent building the biome generation. So...that's a thing.

However, I did take the time to flesh out my concepts for The Scout (both male and female versions), as well as the Scout Rifle.

I was also able to start using suggestions from @ShadowK on implementing the procedural biome generation. Specifically, I'm starting with making a volume of 'cells' (cubes), which get replaced with room data; each room has dimensions both in world space and 'cells' consumed. This is how I will be able to set rooms and connectors.

Week 4 Objectives:
While I'm messing with the biome generator in the background, I plan to make - at the very least - the meshes for the Scout and their gun in its base form. The various alt weapons will take the form of mods that are attached, but power on and off visibly. I also am going to make a 'template' connector and room with which to more accurately begin getting the Biome Generator up and running.

I'll attach some renders through the course of the week.

P.S. - I seriously hope the rest of the challenge doesn't continue my theme of "Week #: [Insert exclamation of misfortune here]" I would love my next entry to be "Week 4: ...Wait, nothing bad happened!?"

Edit: Here I go...some initial in-engine renders of the basic Scout Rifle.

Pictured Here: First-Person view of a door versus the Scout Rifle. You should totally put your money on the door winning.


Pictured Here: The Scout Rifle in-scene. It's a bit stretched due to the fact that the character isn't in the game, so realistic proportions aren't a thing yet.

I'm going to be periodically updating the geometry to make it more interesting. So far, I'm liking the rather simple setup; it comes off to me as a futuristic assault rifle, which is what I'm going for. The screen where iron sights would be may or may not stay; I might change that into a holographic sight, or even excise it altogether and make sights something that can be crafted in-game.

 

So far I have the rough skeleton of a combat system in place. It's been pretty slow goings since I haven't been sure how I want to do any of it. I'm going to try getting parrying in, and that's going to force me to make up my mind on a lot of this. That and movement in combat are what I'm a little stumped on because I don't have good answers for how and if they should be done. At the very least, I'm trying to make it clean enough so that I could rip out whatever I have without too much self loathing.

 

Week 4: "Yay! Nothing went terribly wrong this week!"
Finally.

That said, the game is in a rough patch (but, this is game development, that's a given.) I was able to get a basic version of the Scout Rifle in the game, along with a basic projectile that's not a red cube; I went back to my old projects and borrowed my old 'blaster bolt' setup, so that was easy. I briefly experimented with 3D modelling the Scout from a basic cube, but I think I'm going to just use something like MakeHuman to start my models.

As far as the Room/Biome Generator, stuff got interesting this week. I started with a premise - world generation is a purely processing and I/O process (and, that sentence is mildly redundant.) It would follow that, not being an implementation of gameplay, that this is the perfect setup for making Unit Tests (just so that my life is easier.)

However, Unity intervenes.

You see, UnityEngine.dll has parts written in C++...but that's the thing - only parts. Some of the framework is native C#. This leads to a Unit Testing minefield. You can write tests that involve code that uses Unity Engine code. However, if you use certain bits of code (I've so far identified MonoBehaviour and Quaternion, but I'm certain others exist), you will start to get a very strange Security exception (which, upon Googling, is how I learned about all of that.)

Long story short, I've spent the last three days screwing with Blender while trying to figure out how to get past this. Granted, it's not too late to change course entirely. The reason I want to procedurally generate levels, is to add replay value to this Unity Short. However, it may not be viable - not because procedural generation can't be done (it can), but because it may not be feasible within the remaining four months. So, I'm weighing my options.

Week 5 Objectives
Work on creating the male and female Scout, including the mesh, texture, and bones (no animations...yet.) Instead of a capsule with a gun, a default-posed Scout with a gun floating in midair needs to be running around the test scene. Additionally, it won't hurt to get a little bit of UI going, since that's something I can code reasonably easily. I'll still look at procedural generation, of course, but at the end of this week, if I'm not sure it can be done, I'll scrap it and start working on static geometry. I can still procedurally place objectives.

 

Going to sticky this for a bit, to make it easier to find. Eventually, when it feels right, we can drop the sticky.
Gigi

 

Thanks GIGI, my update is pretty slim.. Not been too well over the last couple of weeks and busy renovating the house.. So time's been a luxury I didn't have..

Anyway, I'll extend the project by a fortnight and crack on this week.!

 

Ok so I'm back doing things, over the weekend I've been messing around with my two favourite subjects lighting and post..!

I used an asset pack for testing to make sure it wasn't me who SUCKS!..

Ok, with that out of the way..! Here is the original image from the asset pack:



And here is the one I messed around with:

(Custom post, Cubemap (IBL) and variations of lighting / colour changes)..



I was going for semi-realism, not sure if I got there (BTW you can use that pic as a colour grading ref if it helps (at all)?

One final one with some slightly modded shadows:



Well anyway I tried, this is in Unity BTW and if people don't like that then I officially give up on trying to make Unity look good.!!!

 

Hi guys, not sure if there still is time to join, but finding this thread was too much of a coincidence ... on the other hand, since I am working full time on this and I've been working on bits and pieces for this game over the last month or so and I feel like I might be around where other people are at. In fact I was aiming at making a mobile game in 2 months, but I haven't the faintest idea if I could finish it since it is the first serious game I will have ever made.

The theme is a space combat game with the flight model based around Starlancer.

Backstory
Having achieved spaceflight throughout the solar system, mankind is looking for a way to cross galaxies and explore the stars. Suddenly, a routine mining expedition in the Kuiper Belt finds an alien artifact in the form of a stargate. Within weeks of activating it, a hostile alien force comes through, destroying ships and stations near the stargate. The stargate is heavily guarded and no further attacks take place, and over the course of the next few months, many probes are sent through the gate. They find a seemingly endless network of gates, and soon a discovery is made of a habitable planet near one of the gates. At the same time, attempts by the aliens to breach the defense systems intensify. Concerned that mankind is isolated and concentrated in one star system, plans are made to send a colony ship to the planet.

Gamestory
You are a spacefighter pilot who's job it is to defend the colony ship as you traverse several gates to arrive at the new planet.

Mission overview
Starts with training mission. As you enter each gate, several objectives have to be completed before moving onto the next one. These missions are mostly based around combat, but may also involve mining for fuel for the colony ship (all first-person) and whatever else I might think of.

Gameplay mechanics
Flight: Similar to Starlancer. Atmospheric-style flight with roll-yaw linked.
Weapons: Lasers and missiles
AI: Simple repeating hit and run tactics (see how it goes).

Story/dialogue mechanics
For dialogue, with a bleep, a message appears on the screen with the responsible face in the upper left corner. Not even going to try to get voice acting.

Mantra
Tools, tools, tools, and workflow...and did I say tools?

What I have so far
A space combat framework (which I will be releasing soon as a starter pack). This includes:
- Spaceflight controls
- Camera
- Weapons: lasers(using raycasts) and target-tracking missiles
- Basic AI (hit and run type tactics) as well as collision avoidance using raycasts
- HUD using the new uGUI:
> On/offscreen target tracking as well as where to aim the weapons for leading high-speed targets)
> 3D radar with dots representing targets in vicinity
- Radar targeting functions (next, nearest, closest, freest, etc)

A few skyboxes

What I need to get done
- Story mechanics (working on now) - if anyone has any pointers I would be very grateful, my experience with text dialogue in games is very limited
- Cutscenes: at this point I'm going to try to keep this simple, such as swinging shots of enemy forces as they arrive, shots of the stargate as you arrive, death animations, objective completed animations etc. (see how it goes) - I don't have much experience with this
- A workflow for setting up new missions

Tools
Unity, Blender, Substance Designer/Painter (all hail Allegorithmic, tryingtomakemetaltextureingimp see you later! )

Workflow
-Setting up new ships: my framework (see above) has modular scripts (engines, avionics, weapons, shields) which I can whack onto a new model when it gets imported.

- Art (modelling): High/low poly models in blender --> bake normal map and get materials ready in Designer --> normal details and material mask in gimp --> whack it all together in Designer and export model, substance/textures to Unity.

- Art(skyboxes): Create texture in gimp, import in Blender and add to subdivided cube (clone over the seams), add planets etc to blender scene, bake environment map, export and slice/assign with script in Unity

- Art(character portraits): Sculpt in Blender, texture in Painter and take a snapshot

-Sound: stuff from freesound.org edited in Audacity, probably buy some music


Game Philosophy
I've heard so much about open-world and sandbox games that I almost feel like going back and playing COD modern warfare . I feel that so many indie games lack momentum and fail to drive the player from one objective to the next. To this end, I want this game to have heavily directed story and missions and a strong sense of momentum, and to be relatively short.

The gameplay will be somewhat simple but I hope to communicate a decent story. What I want to have is regular communication with a scientist studying the stargates and the aliens, probably in between missions (and probably just one still shot of the two of you standing around or something). The mechanic should be similar to Mass Effect/KOTOR involving a few short branches to each conversation that always wrap back to the main thread. I've always enjoyed writing stories so I don't think the dialogue would be hard once the framework has been set up.

The art will be as low-detailed as possible without reducing quality. Halo games have the style I'm aiming for: very clean, low-detail textures with most effort put on the materials and shaders. This to take advantage of Unitys new PBR and the beautiful procedural textures in Designer.

My Influences:
Starlancer, X series, Homeworld series, Stargate show, Alien movies etc

Last but not least, thanks to @ShadowK for starting this, it's just what I needed to give me a bit of that community feeling

 

A bit of my artwork - these ships were made a while back and not specifically designed for the game, and probably won't be used. I decided not even to make a single ship until the space combat programming was finished, since I normally get hung up on artwork before I get any programming done.

Heres a good guy (was made for mobile so a bit low poly):



A bad guy:



Some recent skybox practice:




And a bit of recent random sculpting:


None of these models were made with a single style in mind, so they probably don't look like they mesh together...anyway the pipeline comes first!

 

@Billy4184

Welcome aboard, looking forward to seeing what you come up with.!

-------------------------------------------------------------

So to pick up on my previous post with a fresh mind, I was trying to prove a point. I didn't change shaders, neither did I touch the artwork..

That stark contrast is nothing more than lighting / shadow techniques and post effects, when such things as colour grading post can change the whole spectrum of colours and composition, you can see how important it is..

In short, some who say engines don't make a difference and it's nothing but artwork that makes a game look good is completely and utterly wrong. Like the who game itself it's an accumulation of everything together working in balance..

--------------------------------------------------------

Ok another update, I'm currently finishing off detail and un-wrapping a scene ready to go into Substance Designer.. At the moment it's 32 Cells (So 32 rooms pretty much)..

Also made a holographic material shader for the centre console pieces.

 

I've programmed (self-taught) for a few years but only started game dev mid last year and started learning Blender and Unity side by side, I've been lucky enough to be able to spend most of my time on learning them rather than having to work full time, if you ditch a few luxuries . The great thing about Unity is the fact that there are so many tutorials on the site, every time I need to know how to do something I go to a video on that general topic and sure enough it is usually spelled out for me very clearly. That and the incredible mass of information on the forums. I've only asked one or two dumb questions because they've all been pretty much asked for me

I also signed up to blender cookie and modelling wasn't hard as I've always drawn stuff, but my weak point always was materials. I could never get a decent metal texture for example, always looked like shiny paper, and Designer patched up a big hole in my knowledge there. So with Unity, Blender and Designer (and a bit of Gimp) I feel like I can finally bolt together a decent game on my own. Now to knuckle down a bit and churn out that art!

 

A screenshot of the test scene for the space combat framework. I've turned off a lot of the hud because the placeholder art looks horrific and I've been reorganizing the ship setup so a lot of stuff is not on-line right now ... web player demo coming soon!

 

Life sucked this week, so I didn't get much work done on the UI that at this point I need to actually make sure combat is working the way I think it is. Week before last, I got combat to a point where I can leave it til I've got actual models that need to be animated. I ended up roughly setting up some stats, which was me going "it wasn't supposed to be this way!" over and over again as this slowly devolves into an RPG with forty different stats. Fun times...

On a side note, I love how SerializeField has the second sentence which says "You will almost never need this."... Yeah... I don't think so.

As for what to do next, I'll finish up the timer UI, which is mostly a bunch of gauges (non-interactive sliders). After that I might either hunt down anyone associated with the blender foundation, or work on editor tools for dungeon graph making.

 

I've decided to change the game story and focus more on gameplay than story. This is mainly because I don't have too much experience playing games that tell stories with still shots and text dialogue and I don't think I want to spend a lot of time learning about that right now.

Also, since I like aliens, I won't be casting them as the bad guys I just remembered a story I had for another game that involved a future in which a person/corporation set up an army of robotic/cyborg war machines somewhere in the solar system and then returned to invade Earth. I quite like not-so-distant futures that don't bring in too much extraterrestrial influence, and also I would love to have a game that took advantage of what we know about the solar system, rather than being Earth --> other galaxy.

So you will be a fighter pilot who is sent as part of a military force to take the fight to the enemy throughout the solar system. The game will be very mission-focused and fight-focused rather than story-based, although I hope to tell at least a good backstory, if not one that continues throughout the game.

Anyway, I started work on a ship, probably a heavy fighter/torpedo bomber:

 

Actually I quite like the top picture artistically, although I'll admit that the bottom ones are much more realistic. The top one just has a nice warm look and is not too stark, it's less noisy with the detail. Reminds me more of the sweet, clean graphics in Halo.

 

I prefer the tweaks.. But I get your point, in my own level the textures will be more along the lines of an arch viz material setup with cleaner whites and metallic materials. So it'll suite that type of post processing much better (as a sci-fi game) it's all about that glossy SSR and SHINEZ..!

It may look nice in a screenshot, but all that DOF and bitz n' pieces obfuscated the fact it fell to pieces when you moved around. The noise from the aliasing was horrible to say the least, so it was a good test.. It'll be better when it's adapted to my own stuff.

Also it was to prove a point, post effects and lighting make a massive difference.!

 

I often look at this pic, one of my favourites, when I think I might be adding too many greebles to my ships .. to carry it off with so little detail and virtually no grunge/wear and tear is fantastic..

 

Yeah I prefer a cleaner look, the layout of the underground is starting to resemble Resident Evil's Hive.. Completely by accident, but it's cool ..

 

Final Week: "F#@! it, this was a mistake."
While a bit strong of a conclusion, and emotion, I have to admit - this was a hell of an experience.

The reason Worldseeker is over, as early as the game is in its lifetime, is because I made a few mistakes early on, starting with the concept. While the concept can work, a couple of weeks of reflection and looking over my resources tells me something @ShadowK feared - as I am now, I can't pull this game concept off, at least not to a satisfactory level.

Why Asvarduil should fold 'em
First things first - this idea could have worked. A procedurally generated FPS-Metroidvania would be the best thing since waffles. Where I initially fouled up, was getting the right ideas from the wrong places...specifically, Metroid Prime 2, Halo 4, Awesome Games Done Quick 2015, and others.

Each of those influences has an agenda. The problem is those agendas don't intersect with what a procedurally-generated FPS-Metroidvania by very narrow definition can accomplish.

Metroid Prime 2 is a story-driven FPS that happens to use Metroidvania mechanics to advance a storyline - not what I'm trying to do. Halo 4 is a story-driven FPS that happens to use modern FPS mechanics to advance a storyline - also not what I'm trying to do. AGDQ 2015 shows MP2 done in the quickest time, and while it was something I was paying attention to at the time, it was complete noise - any game can be completed in record time, given the willingness to find said solution to the game.

Worldseeker would have been/might one day still be an adventure game, that happens to use FPS and/or Metroidvania mechanics to allow a player to explore a procedurally generated world. Note how different that is. Also note, how far beyond my current abilities that is.

The problem, is not that I found an idea that "doesn't work" - the problem is I found an idea I'm not advanced enough to yet make work. In fact, this might just be the best idea I've ever had in regards to game design. This might be the game that I struggle to figure out how to realize across my entire career. Where some people start off with their Wow/SAO clone, this might just be that thing for me...what @Gigiwoo would call "the shiny, glamorous" road in the "Road Less Traveled" metaphor used in the Gigiwoo Two-Week Challenge™.

I spent two weeks thinking about Worldseeker, starting with random-level generation, then consulting with my mechanical vision, then looking at what I had and where the path of least resistance could make it go. What I've built is the prototype for an FPS, not an adventure game. Even if my graphics were the equal of anyone's - they're not, but that's the least of my problems with Worldseeker, graphics can be solved by throwing money at reputable artists - this means the worst thing I think I've ever realized about a game I've started: I prototyped the wrong damn thing.

Let that sink in. Now you know why Worldseeker can't go anywhere, in this contest.

The good news, is I've learned from it. First step - leave this contest, I've as good as thrown it. Second step - admit when things have "not gone to plan." (I am deliberately understating.) Third step - learn from it. Don't do an Asvarduil and prototype the wrong thing. In theory I could restart from square one, right now, but I'm six weeks out of a four-and-a-half-month competition behind. The fact that we're playing a game where the primary mechanic is building a game figures into this too (that's step four, by the by.) This isn't to condemn taking part in a contest - on the contrary, I encourage it, you learn lessons like this, but hopefully not as harshly, not that it matters - I did it to myself.

Fifth step? Apply this knowledge. I'm going to finish Sara the Shieldmage - I did prototype the game well, it was coming along great when I figured out the whole Nemesis thing. What's better, I made some improvements to some of the setup I can back-port from this version of Worldseeker to Sara the Shieldmage so it wasn't a waste of time. Work resumes on that game today, starting with worldbuilding.

Sixth step? Write this - all of it. Worldseeker is a game I will one day write; it is a vision that showed me a worst-case scenario, and how most efficiently to screw up a great idea. Just because this incarnation stood no chance of greatness, doesn't mean it can't ever be. I have limitations to overcome, but more importantly, I have to discover the "true" concept of Worldseeker. A "Procedurally-generated FPS-Metroidvania" is entirely too dry to describe the vision, and I did it a disservice by describing it that way, and using that description to start the game. I did much better in this reply about giving up, "A game about exploring a procedurally-generated planet that happens to do that using FPS and Metroidvania mechanics."

@hippocoder once said, "A game without heart is a game no one can love." It's funny how I quoted that, and not six months after the fact, banged my self all over it. Still, while scraping bits of Asvarduil off this topic, some good can come of this: learn from this (monumental, but quick) failure.

I cheer on the other contestants, encourage them to make ShadowK lose his bet (I'm going to be googling, "ShadowK loses bet" on YouTube in a few more months just to see what happens as a result), and hope this helps someone. Next time, things will be different.

Or...

"Failure is but a door, the Game Design forum is but a window: I will be back."*

*: And now you know why I don't write movies.

 

Meh, I don't expect much by the end of this anyway. At the rate I'm going, I'll probably only have a prototype by the end of it. I can already tell I've got too much I'll need to experiment on. I've got tools I need to make, an art style that might work for something human~ish, and animation which I need to figure out normally, plus make it work for my bizarre idea.

Hell, here's what will probably be the main character.

Yes, that's just a hat (it is supposed to be top-down), and I would probably leave it at that for now if not for the shadow. I have no idea how I would be able to animate a human with that sort of low poly aesthetic. I'm thinking I'll have to Rayman it with parts, but we'll see.

 

Firstly, it wouldn't be a good idea to stop.. Sure, you've hit a roadbump and you'll have a 100 more across the road. It's realising it, recouping and then simplifying.. What you're essentially tried to do is a improve on a AAA title and I'm glad you did, I knew it wouldn't work but sometimes it's about finding it out for yourself. It's not a mistake because you gained something.

What would be a mistake is giving up the opportunity to see how far you can take a concept like this, sure drop the procedural generation. But the basis is sound enough.

It could of worked if you had two + years to work on it , I've done procedurally generated dungeons. It comes with a lot more issues than just generating the corridors and getting everything attached, it is truly a time consuming project. Hence the reason Metroid Prime and Halo are what they are ..

I have a stack of prototypes yay high, hence it's a prototype. It becomes a finished product when it's out the door .. What begins as a bog standard FPS can morph into whatever you see fit, if the point was to create a Metroid style game it really needs to be a third person component.

I'm not sure where you wanted it to go, but if you wanted to morph into balls and start the adventure that probably would be the best place to begin.

P.S I think I said earlier it's been extended 3 - 4 weeks as I've been in hospital and still not feeling great.

In conclusion, you've hit some of the base sins of AAA on a shoestring. Which is not a dig, I was hoping someone was going to get to this point ..

• If it sounds complex it probably is, complex equates to lots of time. Get the basis of the game working before any super coding ideas get in your head. (AI / Animation / Block out's / basic sound / basic UI / basic interaction, triggers, camera, movement, cut scene's).
• Time management, being able to pull something like this off requires a lot of experience. Not to say it was a pre-requisite for the challenge, it was a learning effort. To discuss with your peers what would work / what wouldn't and keep the wheel chugging.
• There was a reason I chose SCI-FI, hard surface modelling isn't quite as complex as organic. If modelling is an issue to anyone, they need to reach out to asses how to do it properly. Simple bevelled geometry with vector based colour overlay / noise etc. can look amazing. It's all about effective simplicity...
• Try, Try, Try again.. @Gigiwoo bangs on about this constantly and I whole heartedly agree. Do you honestly believe you can take a challenge on like this without at least a couple of epic failures?
• Intelligent design and tools, get the best out of what you have. Skimping won't get you anywhere, efficient effective workflows are a must.

Now the balls in your court, we can either break it down into it's base components and discuss what's salvageable / where to go next. Or it's quitsville, but you will miss out on a big opportunity to potentially release a game that WILL earn you money and a learning experience that will last for your entire development career ..

Have a drink, a think and let me know .

 

I don't think about failure because that would imply I've not already won. I've won because making games is what I want to do.

What do you want to do? Whatever it is, do it for you, not to compare or compete. I think everyone can relate to that.

 

Take a page from binding of issac, tower of guns etc, design little self contained "rooms" as prefabs. Then you just build the world using a random assortment of the chunks/rooms. The guy who made axiom verge took 5 years, so I dont think it will be a 6 week project.

So maybe some doors couldnt be accessed (if you didnt have the right powerup(s) drill, or the drone, or the double jump etc), maybe the end of the floor boss drops a powerup. Although that might be a FPS roguelike metroidvania (which Im surprised no ones done yet).

 

And you could procedurally generate the rooms later too, once the larger frame of the dungeon has been built. I haven't been able to play Bloodborne so I haven't been able to tell if the chalice dungeons are using predefined rooms or not, possibly even a weird hybrid approach (with a shell of a room that could hot swap components).

Either way, trying to do metroidvania style progress is not an easy formula to generate. A simplified idea would just use gates and keys, where the keys always generate in a location before all of that type of door. In a linear graph with all branches leading to dead ends, this is pretty easy. Once you have paths loop back, you probably need to have some type of pathfinding (BFS?) involved.

 

If Asvarduil is willing, I'll help him make the script. Nothing fancy, create a grid, randomize (room prefabs) as @Aiursrage2k says and then add the connecting blocks.. Shouldn't take long to do. Probably end up looking like this though:

------------------
!!------!!------!!
!!------!!------!!
!!------!!------!!
!!------!!------!!
------------------

With lines the other way connecting of course. Even if it's procedurally generated, it doesn't necessarily have to be anything fancy ..

 

Personally I think this is more like a legendary failure - purple text on this semi-random drop(ping of the ball) isn't good enough to describe this kind of fail. In fact...well, let's ask Captain Picard's command staff:



...Right...well, let's move along.

You do make a point, though, about not giving up, if only because I'm stubborn. That's not to say I should continue the competition; where I'm at, to say I barely stand a chance of winning is like saying Enterprise was an intolerable affront to Star Trek, but I'm only Lt. Commander Obvious; that one requires command-level clearance. The point of the sarcasm, is that I've screwed up so much, my stubbornness says, "dude, you may actually want to sit this one out."

Still, it's a good lesson. I'll take the drink, the think, and consider if, in the next week, any great ideas come that would let me get back into this contest, let alone stand a chance of winning. Boning up from Day 1 is a tough thing to fight back from. Worldseeker is the most genre-neutral - and, generic - title ever, I don't have to make it a FPS/Metroidvania at all. Hell, if I changed Sara the Shieldmage, the most awesome thing since Duct Tap, from a sidescrolling platformy-type game (whatever it was back in the day) to a semi-competent JRPG, surely I can do something as good, if not better, with this one. In fact, I'll do something I'm bad at - listening. Given what you've seen of me (not a lot that's impressive, but...hey. I'm supposed to be listening,) what direction would you recommend me taking Worldseeker in? Bear in mind - graphics aren't my strong suit - music and code is.

EDIT: I'm throwing the whole "procedurally generated" thing out. I don't have to have it. It's just not necessary.

 

I think that a game with any significant amount of procedural generation is probably realistically far beyond the expectations of this challenge, unless you have a lot of experience and code base already. Granted, once you have it all set up, you can produce a level with the press of a button, but to build the brain in such a way that you're sure that what it spits out will make sense takes a lot of work and a LOT of tweaking. For example, it is easy enough to make terrains with cell and perlin noise, and maybe some erosion to get fancy, but will it be walkable, strategically and visually interesting? I haven't made dungeons but I imagine it would be difficult to procedurally generate the x factor into dungeon design. Good quality procedural generation is really the bleeding edge of game design and I would give myself five months or more just to make the the toolkit, let alone the game.

@Asvarduil if I were you I would push ahead and make your game one way or the other, even if you don't have procedural generation now, you will have the art and gameplay mechanics set up and ready to go, and you can add all the other stuff you want whenever you get it done. I have no idea what I will be able to make either by the end of the challenge, but even if I have to add another few months or more onto the end to get it finished, so be it. The point is to find the limits of your capabilities and push them back a bit

Edit: your answer popped up as I was writing

 

By the way, @ShadowK, do all assets have to be original? For those not interested or experienced at making art maybe an asset pack(s) would be a good idea, after all, thats what they're there for...and pulling off a good quality game with a generic asset pack is probably an art in itself ... besides, making assets is what takes 90% of development time

 

Ive seen a few greenlight games that didnt have any art really. I guess just get some simple geometry and maybe some pleasing style people dont seem to care. Then again I started looking at the greenlight list.

https://steamcommunity.com/sharedfiles/filedetails/?id=463312881
https://steamcommunity.com/sharedfiles/filedetails/?id=450389956

 

Well the main point of this challenge is to get you to stand on your own two feet, resources aren't necessarily easily available for more complex matters. So were here to help...

Firstly dig into your weak spots, get the game world built and get the basics in. If you want to avoid a 3D modelling tool for the most part, there is pro-builder although for some things you'll need a modelling tool. Pro-builder will speed things up.. I'll show you some quick workflow methodology for building scenes as well.

If you're looking to do a Metroid Prime sort of game (I'd recommend it) looks fun, then to stop it being a straight up shoot-em up you'll need an interactive environment and UI, so that's definitely where I'd start. Main char, interactive environment, UI and a base char (first person for now).

Later on, what we could do instead of procedural generation is randomisation. It's far simpler, what we do is have a grid block attached, where you will instantiate variations of the tunnels and corridors. Whilst the room placement remains static, the enemy's AI / Path finding will remain the same. If that's the case, you need to make sure all the art is of the same X,Y,Z and the connecting blocks are the same so they can be swapped out.

A while ago I prototyped something like Metroid Prime, below is some pictures This was nothing more than cubes and materials. Hence a proper blockout... If this is something towards what you're aiming for, let me know and we can take it from there.

I have probably prototyped most types of games under the sun LOL!.. Yes folks, that's why I never get things released:

 

Not all of it no, obviously it's better to re-factor and add your own touches. Make everything coherent, but for a lot of basic models that suck up time it's not worth it.

 

Incredible. Made me remember that I saw this over at the Allegorithmic forums, all flat geometry and normal maps, and on the asset store too:



Made me think about adding 3rd person (space station?) gunplay to my game. Too bad spaceships need an outline

 

Exactly, looks great to me.. Another perfect example of simple geometry manipulated to look awesome, add some crates and random paraphernalia to that sort of style and it'll look AAA worthy. Bar the texturing and un-wrapping I'm not sure why you couldn't do something like that in Blender or whatever in a couple of hours.

Of course with mine I like to go OTT and put detail everywhere, that's why it's taking me longer than usual.. Putting holographic consoles in etc. LOL.!

Sure yeah outside's of spaceships with tubes lumps and bumps can take a while..

 

So I've made a lot of progress with my space combat mechanics over the last week. The current version is what I'm putting it on the Asset store as a starter kit so I'm not sure how relevant it is to post stuff here about it, but here's the the link to the WIP thread with a gameplay demo, I'd appreciate feedback if you have the time (sorry for the spam but it really will be the backbone of my game).

Anyway since my game is a bog standard space shooter at the moment I was thinking about gameplay mechanics to set it apart and something occurred to me last night. First a quick story recap/update:

Backdrop:
Sometime in the future, advanced, fully autonomous AI war machines are implemented and something goes horribly wrong, causing a human vs AI battle that results in massive damage to Earth and its population. Humans win because of the built-in weaknesses of the AI system but such systems are banned from then on by international treaty. 'Enhanced humans' (i.e., cyborgs, you!) are a relic of this war, an attempt to match the human mind with the combat capabilities of the robots to confront them on equal footing.

Now:
Another cyborg veteran of this war has moved out into a desolate region of the solar system to build an AI army to return and invade Earth. You are a combat pilot (soldier as well?) who is part of a desperate attempt to neutralize this threat which could well spell the end of humanity.

Novel gameplay mechanic - Mind Control:
As you engage in combat with enemy AI machines, you will have the ability to hack and gain control of their systems using your cyborg enhancements, giving you the ability to set their targets or cause them to self-destruct. However, this ability will drain your resistance to the same tactic and enemy AI can then take advantage. The result of enemy hacking your mind will be foggy vision (probably implemented with motion blur) as well as the tendency of your ship to do things you dont want it to!

What do you reckon?

 

@Billy4814

Sounds good to me, actually in some places sounds a little too similar to the plot line I had. Except without the Game of Thrones twists every 10 minutes..

Also the robo's In mine were actually made just like humans were, it's a complete mystery and one of the speech snippets were..

Protagonist: Who created you, machine?
Machine: Who created you, human?



P.S also sounds a bit Mass Effect (the Geth)..

 

Yeah that could be a cool idea.

 

Week #I don't remember what week we are on:
Can't say I worked on the game as much as I should have... I played way too many new games, and life got in the way too much but at least I am still attempting this!

Stuff I implemented:
- Scrolling combat text ( customized to take in any type of colour gradient, crit effect, any numbers or text, and converts numbers in a more readable format ex. 10000 -> 10,000 )
- Buffs / Debuffs displayed on monsters and stats
- Range / Melee Skills
- any skill has an "action" that allows the user to do any amount of attacks, heals, buffs, etc
- any skill can be single target / Aoe / self / everyone targetted, etc
- any skill can have several actions that could be anything ( attack, and heal after, and buff, and attack after that etc while any being aoe, single target or random target)
- dynamic and can change at runtime and will read it properly
- Nicer looking terrain and UI
- Some type of FPS limiter ( just to make sure everything works under lag lol? )
- All formula's implemented like total damage after defense and multipliers are caculated, crit rate/damage, accuracy and resistance calculations, etc.
- Working model in, to move away from boxes! ( unity asset store model, I don't take any credit on this one )

Everything works nicely with no fps problems on my android phone, and some gameplay is shown in this video

I need help with a problem:
I am very happy with my current implementation but if you watch the video above, I made the SAME "monster" with the SAME skill on each monster and decided to attack like normal against any type of monster and realized that my timers are off when I attack monsters further away compared to a monster in front of me. I made the video only to show this problem lol.
Now looking at my code I use coroutines and I don't take into consideration the distance of my target, I just simply say "move forward" and "move back" with a coroutine timer and I say "deal damage" at a certain time with my coroutine as a "simulated" attack or do 3 attacks after a certain amount of time.
Now I would like to keep it like this because its simple and allows me to simulate different attacks easily, but how do I deal with different enemy distances properly? Like only playing an attack when monster reaches target or maybe an invisible square reaches the target to test its distance for a range attack? I do not want to use Unity animator to solve this problem.

Thanks for advice and watching the video if you watched it.

 

You do realise this is a PC game challenge right? It shouldn't be able to run on Android and I can guarantee your poly budget can be higher.

Anyway, do your damage system separate.

 

Yup, np! What do you mean by damage system separate?

 

Well the cool thing about a turn based system is that it can be relatively simple, what I'd do is move the vector position until you intersect a raycast from the opposing player (or you could do it collision based, your choice). Then apply damage, the timer thing seems a little un-necessary..

You can follow up with sequential characters triggering after, it could be something really simple like a switch case with some bool's. When the character has finished his sequence the next part will trigger.