next »

[Tyrus Peace]

Hi, I'm new here.




Cloudbase Prime is a first-person game where the world around you is composed of hexagon pillars that you can move up and down with your sweet glowy hand. It's called an architect. It can be used to access new areas or launch yourself and enemies skyward.


You're piloting a robot on a gas giant. You fall into a lost mining platform hidden in the clouds, full of wild robots. Things get violent on occasion.


But there are robots you don't, shoot, too! That happens.



I've been working on this game off and on, part time, since I made it as a 7DFPS entry in 2012, but I've been working on it full time as of September 2015 and things are moving much quicker now! The next alpha will likely be the early access release.

Cloudbase Prime's out now on Steam! It exited Early Access on July 26, 2017.

[Tyrus Peace]

Things are fairly far along now, but I figured I could cover some of the more interesting dev things I've done since the original 7DFPS build. I'll go into more detail on other stuff, too, if there's interest and I actually have useful stuff to say.

7DFPS (2012):


now:


Walker mechs: (started Aug 2013)

Big enemies in games are cool. When the game allows the player to suddenly move the ground underneath them, things can get a little weird. Because of that I designed walker enemies to dynamically handle complicated terrain in the same way roaches and (I think, from fuzzy childhood memories of documentaries) the Mars Pathfinder used: just have each leg scoot toward the destination on its own and don't worry about the rest! The main walker AI just decides where it wants to go, and if it can pick up another leg without falling over. Each leg figures out the rest on its own: where the nearest hex it can get to is, what path it should take to get there, etc.

I tried a few different solutions, but a simple bezier curve for the legs ended up being the most flexible way to dynamically animate a noodly leg with a foot that can move to arbitrary locations. I  also use that bezier curve to place (less detailed) colliders and shadow casting meshes!

I figured out most of this in 2013, but I reuse this walker leg code for all kinds of things (tinier walkers, landing legs for flying enemies, multi-form boss transformations), and I seem to do something to break it in new and exciting ways in every other update. Disabling colliders on far away hex terrain for performance and adding angled hex terrain caused some head scratching, for example.

The flipside to that is when the AI does cool unexpected stuff, like being a good climber!

Optimizing performance for hex terrain/tiles: (started Sept 2015)

Having a game with this many colliders can be pretty performance intensive, despite the simple look. Once I made a heightmap->terrain generator, it became easy to create levels huge enough to cause performance problems.

The player can move a lot of terrain in a short amount of time, so I leave any hexes close to the player as separate colliders/rendered objects. I merge meshes when rendering far away hex terrain (usually anything that's too far from the player to be moved up/down). I cull colliders for any hex terrain that far away, as that's cheaper than building and calculating collisions for a new mesh collider. This is actually where I hook in a lot of my LOD logic: most things that could inhabit a hex also listen for it being renderer-merged and LOD/hide themselves in response. My target machine is a 2008 MacBook and it's still running fine there .

Angled terrain: (started Oct 2015)
I actually worked on the much more complicated support for dynamically rotating hex terrain last October so that this giant robot would be climbable while animating. Honestly, most of the work just went into removing a lot of game-jam-era assumptions I'd made about "up" always being in the y direction. Doing the actual vector math to figure out the terrain movement and new launch vector based on the current transform's rotation was all I really had to do, and it was all stuff I should've done in the first place. Noice.

And I got it pretty well tested when working on this dragon boss:


That was definitely most of the heavy lifting for placing hexes at more arbitrary angles, but there was still more to do once I decided I wanted to reuse them elsewhere. The chief expense has been fixing some bugs that weren't producable the first time around and tweaking enemies to behave a bit more intelligently around it.

Now that I've been using angled terrain in a good few other places, I'm pleased with the new gameplay I've been able to fiddle with because of it .

[TonyManfredonia]

Oh boy. This looks wonderfully abstract, in a "Portal"-ish way. Very artistic and unique in design!

It'd also work really well for VR, given the leaps and distant falling. Super neat! Do you have any plans for VR? Simply curious

I can't wait to see more!

[jctwood]

Love the enemy design so far!

[bdsowers]

Love it!

Some things I'd like to see:
* Enemy hit responses. Right now the enemies feel very static while you're hitting them, like they're brushing it off. Maybe a little wiggle or something to indicate that your hit did something important, or they have a larger flash. You have a little particle effect, but it's not much to run with.
* A more dynamic "fire" animation. it seems a bit stiff as-is. Maybe look at Orendi from Battleborn - she throws projectiles with her hands.

[Tyrus Peace]

amanfr01:
Thanks so much! There's actually a joke in-game where the suit AI realizes you're using a mouse/keyboard to control it and it has to... reconsider the intro tutorial. I've considered making the inside of that suit visible in VR having a mouse/keyboard/desk behind the suit's window that you look down at while playing VR. It'd be a pretty expensive joke... It's not the only suit I'm planning, though, so it could be worth it ! This is what the inside of the suit currently looks like in the new menus.

JctWood:
Thanks! I tried to make them simple but "functional", at least according to the weird internal logic of the game so far.

bdsowers:
Agreed! Damage flashes are pretty high up on my todo list. Some of the enemies already flinch after taking a certain amount of damage and/or losing a leg, but I'll be cautious about adding flinch more often than that because of how fast/low damage the player's weapon tends to be. Immediate feedback after hitting an enemy is definitely necessary, and flashin' is the main way I plan to do that.

The firing animation's slower than it oughta be; I've sped up the firing rate and added auto-fire since I did that. Hopefully that, and maybe adding multiple different firing animations, will help make the relatively boring "shiny ball of pain flyin' outta my palm" firing style more interesting and obvious.

Thanks :D.

[Tyrus Peace]

I've released a new alpha! It's been two months since the last major alpha, so there's a kinda big pile of changes. It's got four new levels: one boss, one boosty water level, and two story oriented levels that have lots of angled hex terrain experiments. I've also finally started adding hidden upgrades for the player's robot suit to various places in the game. So far they all take the form of weapon modifiers or new skills/powers.

New levels:


These rings basically turn the player into a missile once they swim/fall through. It seemed like a fun way to have boost rings without pinning the player to a specific direction/path. It does make things feel a little slow when not boosting; such is the cost of super speed. Hopefully that does drive home how big the new underwater area is for players, though!


Also, those boost rings are good for getting this fixbot .


This boss.


I've added a fair amount of dialog, in the form of a little fixbot campfire song and a giant 9 headed "Atlas core". I hope it does a good job of explaining why wild robots wandering around everywhere. I don't want to assume that *anyone* can consistently understand all of these babbling bots, so I'll probably make subtitles mandatory. I should make them look a bit less horrendous before that, though .


The player can just walk away from any story sequences if they're bored, but I'm also considering a "next" button to move on to the next line of the conversation. It's pretty non-standard to bring up any sort of dialog UI in a game where talking's just in the "background" as you run around, though. I'll probably wait for players to complain first.

FYI: listening to the same fourteen lines spoken by a hive of robot voices at over and over again for a day is a really good way to break your brain's language bits. Took a bit to get my good talkins back .

SWEET NEW MENUS:

YEEEEAAAAAH! Hopefully this does a good job of exposing the new combat systems, gives players a better idea of what their robot suit looks like, and makes it less of a pain to navigate the level picker.

New combat systems:

I've added a big chunk of new gameplay by introducing the idea of stealing fuel from enemies by launching them. This fuels a great bit pile of new powers that I've added for the player to find throughout the game. For the most part they're fun and I'm happy with them!

• Holo tiles let you stop unwanted falls and block enemy projectiles.
• Shockwave lets you launch a whole bunch of enemies at once to get that sweet, sweet fuel.
• Architect grenades blow stuff up and leave nice dents in terrain .
• Also there's a death beam.

I'm a little worried about the weapon upgrades, though. It's an intentional wrench I'm throwing in to combat balance to try and shake new things out. Some of it's working, some of it's not. But some of it's fun and busts balance! Dang it.


Charge shot's alright! Not the most novel thing but it feels nice and can do a lot of damage.


Homing shot's really accurate for a game that's mostly designed around clumsily missing enemies before launching them a bunch. It feels so cool, though! I'll probably nerf the default number of lockons allowed while making the weapon upgradeable via other pickups later on.

Combust ammo needs to be manually loaded, and causes an explosion in any enemy that gets hit three times in a row or killed by combust shots. It looks cool, but may just boil down to "do more damage for 30 seconds", which isn't as fun or balanced as I was hoping for. Perhaps I'll just make it a more passive boost, rather than a temporary bonus on "reload" like it is now. This is one of those things where I may be seeing through to the core mechanics a bit more than players will, though. I'll fish around for feedback before bailing on the idea, but I'm glad I have an exit strategy.

I guess I should post this to playtesting as well! I'm definitely diggin' around for feedback for this new alpha . My next major release will likely be the early access candidate!

[Tyrus Peace]

Food Chain Mode! (Feed bunnies to your friends to defeat evil.)



I should be working on adding more story scenes to the game, but as my next release is going on Steam Early Access, I really wanted to add more gameplay for people, too. So this mode happened! It's long been an idea in my game's world that the wild robot food chain exists as a way to accumulate fuel at the top. I figured I should have a little bit of that show up in actual gameplay.

The basic idea for this game mode:

• There is an evil robot factory spewing out enemies that attack you and your bunny friends.
• A hungry hungry robot will eat bunnies to fuel a powerful beam attack against that factory.
• Get the bunnies to that robot. Don't let it die.

This album of GIFs shows the basic gameplay/flow for this mode.

What do you think? Sorry I didn't embed directly. This thread is getting... meaty, and these gifs are far smaller as imgur video embeds.

[Tyrus Peace]

Cloudbase Prime's coming soon to Steam Early Access! I'm excited! And yeah, a bit scared. Mostly, I hope the game will get some attention and feedback. It's tough to create in a vacuum.




Cloudbase Prime's Early Access trailer, made by Respawn Series


About working on the trailer

I was fortunate enough to find out these guys were interested in making trailers for indie games, and had a really good experience! This trailer's quite a bit better at showing the game's full spectrum than the gameplay footage I've been posting.

I'm fortunate that my videos aren't the only game footage on YouTube, but Tubers aren't providing a very good example of the overall game, either! Most "Let's Play"ers only play the first couple levels, say it's great, and never play it again. I imagine a lot of indie devs are familiar with this: tiny indie game streamers, especially those playing free releases like early alphas, feed their audience variety and don't really stick with any game for a long time.

It's exhausting/repetitive to go back and make a "comprehensive" trailer for every new release, and I don't recommend that. It obviously had to be done this time around, though!

I talked with Respawn Series a lot to be sure the trailer had all the elements I wanted, and it was a really nice iterative process. They had a lot of great ideas on their own after playing with the game, but I wouldn't recommend getting a trailer done in general unless you've got an idea of what in your game you want to show off, and how to succinctly show and tell that your game is worthwhile. How can you be sure you'll be happy with the result if you can't verbalize what you want?

Also, you'll probably be a real pain in the neck to work with if you don't know what you want, which'll make the process more painful for whoever's cutting your trailer and you.

And yeah, I did the voice acting . Matt Javanshir did some great work making a trailer-specific music track, too!

----

About my first Early Access alpha

The game's fairly far along, and some folks have told me I shouldn't be calling it an alpha any more, but I'm still calling it an alpha. I plan to call it a beta once I finish the main storyline and story-levels.

This is a fairly big story update, but the story bits in Cloudbase Prime don't take a particularly long time! And in terms of new level content there'll only be a few more levels. I hope folks aren't going to compare this to the last (free) alpha and say it's not worth the "extra" cost! The intent of the free alpha was to entice/thank testers with a free alpha build of the game, not to devalue the game's paid release.

Frankly, the game received so little attention during the free alpha phase that I doubt anyone will notice. I wouldn't be releasing to Early Access in the first place if I had enough player feedback for this to be a problem !

It is hard to work on a big story update, though! I like sharing my progress, but there's not much I can do that's not a huge spoiler at this point . This guy looks cool though.

[io3 creations]

Overall, I like the visual style and the gameplay.  There are quite a few things that work well for a 3d action platformer but some aspects felt off.

I actually have a somewhat similar game mechanic idea that's similar and have questions about how to handle the mechanics in terms of fun and challenge.   I wonder if some things were intended for the demo but'll be different in the actual game.  It felt like that the range of the "platform moving hand" was like almost having nearly infinite ammo and most areas were very easy and didn't provide much challenge.  Starting with shorter (but upgradable) range could be used to create more tension and increase gradually with additional challenges.  Also, none of the other weapons that I saw in images were seemed to be needed to finish the levels.  Well, I got stuck with level 5-2 (might be intentional?) around the beginning of the level where can rotate a bouncy platform.  It seemed like you could jump to an area with enemies but those enemies would kill your character quickly.  Maybe that's where those weapon upgrades could be useful?

The other aspect is level design and challenge balancing.  I personally would preferred if the game started out with smaller areas and would increase level size later.
You've probably played such platformers and another devlog: Suzy Cube https://forums.tigsource.com/index.php?topic=49900.0 seems to do go for that approach, that my require a different level design approach and sufficient feedback.  Feedback can be interesting and I'm also thinking that one benefit of Eary Access would be that *hopefully* there'll be feedback from a larger playerbase regarding bugs and gameplay.  Of course, you have to keep in mind your vision for the game and only focus on the feedback that is relevant as some may request changes, features that would make the game something that it isn't. 

As far as bugs go, I only noticed one on level 5-2 with the mentioned rotatable bouncy platform.  Sometimes, my character would just fall through the top part and land on the "base" of it. 

I'd say Steam Early Access covers all phases of game development that isn't actual release, so it's not necessary to add Alpha to the end.  Chances are you'll be making multiple minor updates to the game (bug fixes, adjustments) so it might be better to go with version numbers anyways and update that number for newer versions.

As for gameplay length, you probably know that you can request a refund on Steam if you play the game for less than 2 hours.  That might be one aspect to consider and also what is the total gameplay time relative to the cost e.g. some/many people found Inside's ~3hrs gameplay for $20 expensive.

[Tyrus Peace]

io3, thanks for the feedback! I'm not sure how you got a demo, but it sounds like you got the old 0.8 version, which has some bugs with the hitboxes for those rotateable terrain. It also has a bug where a giant poop icon with an arrow pointing down is missing above the bucket directly across from that rotateable hex terrain.

I think that, without that bug, it's a lot more obvious what to do on 5-2 .

None of the found powerups are necessary to complete levels; they just make some things easier to do and give the player more options. Story-given powerups will probably be more required, but I'm not prone to designing challenges that can only be passed one way.

I call the game alpha because it's not feature complete. It's an attempt to be honest with folks about the game's content, although alpha/beta are misused often enough that I get that's not helpful for everyone.

[io3 creations]

Yeah, it's "Cloudbase Prime alpha 0.8 demo" that I actually downloaded and played about a month ago (probably from a link here).  As I was thinking about a few aspects of it, I forgot to reply ... until now.

Yeah, I also wonder about how especially non-developers perceive alpha/beta notation.  But if you release the game as Early Access, then you can specify what has been already implemented and what hasn't.  In terms of game development, that should be more obvious (and transparent) for everyone without the alpha/beta labels.  At least, that's my current assumption.

[Tyrus Peace]

YAAAAY! It's very neat to see this on something I made.

So! The promo bit? I've been contacting youtubers and twitch streamers for weeks, and wrote a few journalists/outlets that wrote about the original 7DFPS release. Honestly, jam coverage doesn't seem very eager to translate into further coverage down the road. It's very much a "hey this game idea! look at it! okay, moving on." Really, it's hard to find any writers that go more in-depth than that for indie games, in general.

I think I'm getting a good response, though! It'll be an exciting week.

Also, I wrote about how fatherhood and indie dev are connected for me.

I actually added a level unlocking system in this alpha based on the number of fixbots you beat and whether you defeated the alst zone's boss. A couple days ago realized that the jet dragon boss fight was still a huge pain in the ass, so I also decided to make that a lot easier. The dragon will now catch you if you fall.


"The dragon will catch you if you fall" should be a song. Maybe I will sing it one day.

[Tyrus Peace]

Primians: Filling up a robot city

I actually made these guys a while ago. When making the levels for zone 5 I really wanted it to seem like a little robot city, so I modeled these little fellas who'll wander around, stop, and look at ya. Naturally, they start running if things go wrong. There are actually four types of lil' fella. I'll only post two, to keep this short.





I also made a font for their language, which is designed to be read from top to bottom, and is designed to work mirrored. For some reason it seemed obvious to me that robots would make sure their written language had these properties. The font is, technically, poorly structured, but it produces clean textures for my signs, so I'm happy.



There are a lot of windows in the city, so I figured I'd program some window peekers. They hide when you look back.

[Tyrus Peace]

I can give this guy eyebrows with my crosshairs. hehehe.

[Tyrus Peace]

Megabunny WIP. This is a quick demo animation I did; for the most part these'll play separately in game.

It'll actually be a mech walker, but the legs will be separately rendered, as they'll grab hex tiles and move around that way. Like the game's other mech walkers.

[Tyrus Peace]

Megabunny has arrived:




Gameplay-wise, it'll be a a friend you build up during the game's foodchain mode by collecting bunny balls. You won't be able to spawn more than one at once. Megabunny's pretty deadly until tinier enemies bite its legs off. Then it just looks sad.

[a13dev]

Megabunny is great!

When we were researching 3D platformers I came across Cloudbase Prime about a year ago, great to see you've made it to Early Access. You've got a lot of cute content that you should be spreading across the social media-verse, you could get loads of people involved with your EA campaign!

[Tyrus Peace]

Thanks a13dev, but that's not quite how it's working out! People tell me the game looks like it should be popular a lot. It is not! Please feel free to share it with folks if you think it's worth their time.

I'm glad you like the game's art .

[Tyrus Peace]

I've gotten into pretty spoilery work lately, which is frustrating! I can't share as much as I used to. But here's part of the next boss I'm working on :D.