Feedback Feedback Friday #140 - November 20 - 23, 2020

Hi all,

It's Feedback Friday time!

Proper playtesting and feedback early on in your game will save you tons of time - according to this postmortem, it might even save you a whole year!

If that sounds like something you could use, you've found the right thread! Feedback Friday runs from Friday to Monday every week. Share a build of your work in progress game to get playtesting and feedback, get new perspectives to help you overcome your biggest obstacles and take advantage of the Unity Community to boost your game development!

What To Show

• Minimum Viable Product (MVP) - Core game play > everything else
• How To Scope Small (Unity tutorial)
• Post a link to a playable game, preferably WebGL. If you don't have a playable game, post something substantial, not just text.

How To Ask For Feedback

• Be concise.
• Specify what you want feedback on and what you don't.
• Resist the urge to write an immediate defense. Take the time to understand their points. Remember that your friends here are taking time out of their busy schedules to help you for free.

How To Give Feedback

• Be positive. There's something of value in every game.
• Focus on the design, not the designer.

Feedback Friday #139 is here.

 

Gif first:

Okay, so I've discovered that

1. WebGL still works with itch, I just wasn't looking in the right place

2. ...Unfortunately I don't think I can make a WebGL build, because my code is heavily reliant upon multi-threading (lots of Parallel.For), which WebGL doesn't seem to handle appropriately. I'll see if I can run everything on the main thread, but I fear that will cause a bunch of slowdown. The downloadable version online is updated, though (https://eternalambiguity.itch.io/ca-shooter), and for now:

3. I added a second "enemy" type that shoots directly at the player. My original idea was something like "bullet hell" so it's in line with that, but I don't know if that's the right direction to take things.

4. I added a "wall" for the player to use to defend themselves. It isn't shown in the gif because I didn't stay in one place, but they can be destroyed. Right now they're set up directly on the simulation grid like the enemies.

5. "The Blob" was more popular than I expected! I modified the growth rules so that there's only growth if only 1 neighbor is of the "growing" type (in other words preventing growth when more than one neighbor has changed). In my opinion it looks far weirder/more sinister.

Something else from my original thread:

I imagine something like that would work well here, though I'll have to figure out how to create it. That image was created with diffusion limited aggregation, which works backwards (cells on the edges are added to the center when they reach it). I'll have to figure out a way to create a similar effect when starting from the center (this has bigger implications than a video game, but I'm cool with a game being the impetus).

Any thoughts?

 

@EternalAmbiguity there is nice progress, from what I have seen other time.
But what is the exact feedback, you would like to receive on?

Probably you want to look into fractals. Or something like L-System (Lindenmayer Systems), to allow kind of more organic growth.

However, that may not be as easy, once you allow to break chain, like shooting in the middle for example. Complexity of the algorithm may grow dramatically.

For the inspiration, are you familiar with a Creeper World game?

https://duckduckgo.com/?q=creeper+World+game&iar=images&iax=images&ia=images

Have you considered Project Tiny?

 

Since the growth rules are simple and deterministic, maybe the player could somehow take advantage of knowing the rules. There are too many cells for the player to track individually, though, so I don't know where that thought really leads.

 

Initially I'd hoped people could play the game and then tell me if they think it's interesting, along with any suggestions for gameplay. My goal is to get a very simple game on Steam just to get some experience with the whole process, but I want it to be reasonably interesting to play. I wanted it to be based on a cellular automaton, but I wasn't really sure of where to take that in terms of mechanics.

I'm not familiar with that game (Creeper), nor Project Tiny. I'll check them out.

Something like that might work better for the moving agents. Right now it's just a random walk, with 0.25 for each direction, but that could be adjusted (it might also be adjusted on-the-fly, a la Pacman). And there's no reason the grid has to be completely empty/open - it could have borders/different areas.

 

Creeper World to be more correct. Version 1 is free to play using flash, while version 4 is on steam. Demo available.

https://forum.unity.com/forums/project-tiny.151/

https://github.com/Unity-Technologies/ProjectTinySamples

 

I liked the version where it was a solid blob more.

Kind of out there, but maybe add another layer on it where there are "seeds" that cause the growth and you can cut into the blob to hit the seeds and kill off a chunk of it, all the while it keeps making more seeds.

It may not work as well, but I think it may be worth exploring.

Also, LOVE the walls idea.

But maybe instead of plopping down walls in solid chunks its actually an alternative shot that either A) goes a short distance then makes a single block where it lands, or B) you turn the blob its self into a wall.

Maybe it's like a "freeze ray" that only temporarily halts growth.

Also, this is BEGGING for a explosives.

I could see it also potentially being a double edged sword that takes out a chunk, but sends particulates of the blob outward that in turn start growing.

 

I like the way you've gone about making a shoot 'em up style game. The cellular automation aspect sounds quite complex, but I bet you could make some very interesting enemies with that, as well as power ups. I second Not_Sure in that this does seem to be begging for some explosion effects.

---

https://visionbreak.itch.io/sunsear

I'm also making a shoot 'em up game. Much simpler though, and for Android along with WebGL. This is my second project, but this will be the first one I end up releasing (hopefully). Here's a video of the Android version:

Please mind the sound, as mixing hasn't been done yet. Feel free to lower the volume or mute.

Feedback I'm looking for:

Anything really. Nothing also works. I know this isn't a support forum but if someone has a suggestion for that android bug of sprites being black when a material is used, I'll take that feedback.

 

@cyangamer1 - You mentioned WebGL. Any chance you could share a link to a playable WebGL build?

 

Done. It takes over 45 minutes just to switch between Android and WebGL, hence the delay in getting a build out. I put in much more optimization for the android version, but this should give a good feel.

https://visionbreak.itch.io/sunsear

other post has been updated as well

 

@cyangamer1 unplayable on PC web browser.
How to move?

 

I was able to play it in Chrome on Windows 10. I really like the powerups and abilities, especially that big purple power attack.

The controls definitely feel like they're designed for mobile. Using a mouse is somewhat awkward. For web, it might be nice to have some keyboard shortcuts. For mobile, what would you think about moving the square buttons to the sides of the screen? Right now, I'd have to cover up the screen to activate the shield, etc., since they're in the middle.

Very minor thing: The loading screen looks a bit strange:

 

I had the same problem. That said, the video looks great, gameplay seems flashy and enjoyable - I'd say it definitely has potential. Thanks for sharing your early work on this. I'll take another look when there's an answer about the controls, hopefully it's just something obvious we missed, as @TonyLi had no problem. I just tried again but still can't move around.

 

When the stage starts click down on the lower part of the player ship. There's currently a region around the model that you have to click / touch in order to move (that will obviously be changed). it won't work if you click anywhere on the screen, unfortunately.

Thanks for the feedback! Which sides? Like 1 on the bottom, 1 on the left edge, and 1 on the right edge? wouldn't hurt. to try it.

Yeah, no polish has been done for webgl yet since I want to get Android in a good spot first before I switch.

Thanks! The only thing I can think of is the suggestion I made above (clicking directly on the player ship in the mid/lower region). If that doesn't work then I think something else it a play, but the issue will likely go away once I fix the touch system.

 

I gave a second try go.
I managed to find that region to click and hold, which follow the ship.
I don't know how practical this is on mobile, but on PC it will quickly frustrating.
Now add an invisible mouse cursor and you dead in no time
Also that skills after click, annoyingly (for Desktop user) pops up.
It definitely need different control system including keyboards, if targeting desktop users too.
That will help preventing negative reviews in future.

It is indeed good, to find these things early.
Other than than I like presentation vid.

 

Anywhere that will keep the player's hand from covering the screen as much as possible.

 

Cool - it's working. I did have one input-related issue, where I had dragged my ship low on the screen, and released to select a powerup, then I couldn't click low enough on the screen to grab it again (I eventually did, but pretty much had get the cursor over the bottom row of pixels).

The explosions are nice but there were times I couldn't see there was still an enemy. Maybe that's a gameplay element.. but if not, you could reduce the alpha on the smoke a touch so things seem more within it, as opposed to being covered by it.

 

I may join a wagon, before end of the weekend

So I am currently working on L-System DOTS based, to generate procedural plants.
Still some issues / matters to resolve.

Spoiler

Anyone interested more about details, I discussed it here
https://forum.unity.com/threads/wip...ype-dots-main-menu.773210/page-2#post-6525340

I am aiming implement it into my main project,



replacing current voxel base lifeforms. They do evolve and grow new cells, affecting their functionality etc.
But I would like to see more organic evolution of plants. Of course, that introduces problem of none-plant lifeforms, which can potentially move and share plant cells traits, which I need still to resolve. No issue grammatically, but rather, how to represent cross-mutation in logical and appealing way. Spore may give some ideas concepts for reference. Of course, I don't want to go spore creator route (at least not entirely).

I will also keep plants trees structure very minimalistic, for performance reasons.


My question is:

Do you think implementing such procedural tree feature / mechanics is a good idea in my project?
Any thoughts?

 

@Antypodish - There's something to say for breaking new ground technologically. So that could be good. But I think it also needs to support your project's core hook. Would you please reiterate for us what the core concept of your project is?

 

This is very good question.


Main Concept

So my main concept focus on lifeforms evolution.

As the core base is, where lifeforms bringing new offspring, then these mutate, affecting their own traits and cells.
Having more cells, allow to have more traits, or strengthening existing one.
Some may be good for specific conditions, other don't.

I.e. chloroplast cells as part of the plant traits, allow to gain energy from sunlight. Energy gain is low, but is pretty much free, like solar panels. However day / night cycles have an effect on produced energy level.
While muscles cells allow to move, which cost more energy and along with herbivore traits, allow to hunt lifeforms with plant traits. In long run, much more efficient for gaining food and energy production. There is more into this mechanics, but no need for details at this point. However, these are in game already and are to stay.

Also, there is faction mechanics, where lifeforms from one faction, will be more hostile, against other faction and can compete for a space and domination. Could say, evolution race has started.

While game can run on its own as simulation, I am introducing more mechanics, which allow mutate new offspring, in more desired direction.


Building Cellular Lifeforms

Primarily I had lifeforms represented as cells, which are voxels. So each cells, or building block, adds new functionality. Potentially "building" large species, like trees, or elephants Sure they won't have moving lengs, or trunk, but can resemble of such. Of course, game simulation (AI) may be looking for more optimal cellular evolution and its structure. And I would like keep that mechanics somechow. At least to certain extent.. Even with new proposed L-System, I can check for position of plant components, to find out, if has benefits, for specific environments. I.e. chloroplasts / leafs lower to the ground, will gain less energy, than up high, or in middle of the lifeforms (useless and cost energy to breed such offspring as well.).

Problem with building by hand mechanics is, It seems, or feel too tedious, to modify larger cellular lifeforms.
Not sure if it will be fun, while game is running and things are changing around. I would rather have something, which allow player to guide evolution direction of desired lifeforms or species, than getting focus on to micro level.
I think there will be plenty to do already anyway.

I mean, I loved building in minecraft or From The Depths, and making stuff move and fight as well (as meny do too). That is micro management sure. But main difference here is, I am dealing in EcoPico with potential RTS alike situation. What if I would like to have implemented multiplier in (far) future? (rhetorical question, thinking aloud).

This is why I thought, bringing something like L-System, to give more organic behavior, could be potentially interesting and hopefully fun to play with, if executed well (yeah, define well if no one done that? ).

I am hoping to find some sensible and interesting solution.


Visual Aspect

Saying all that, as in current mechanics and new propose L-System mechanics, I want visual parts to have more meaning, than just visual. Like specific cells appearance, carry also their specific functionality.

I do realize, that going from cellular level, to more macro level, like plants and trees, may break my some initial concepts. But I am keen on exploring options.I just hope, not to diverge too far from main scope. But on other hand, I want make game more appealing to potential audience.

Consider just pure plant tree. Here for example, starting with single branch lifeform, as if first cell. Gaining certain traits over evolved generations, would allow to grow leafs, or rather first leaf (maybe). Other traits would allow to make branch bigger, storing more energy etc. Few generations later, new branch would appear. Our tree now having two branches or so. Then can focus more on leafs evolution. Should be going higher, or wider, etc.

Now you look at lifeform and can judge briefly, what is capable of and what some of none hidden traits may carry.

Player then can focus, on evolving specific strain of lifeforms.


I hope above makes sense.
Any thoughts, on new proposed system? Again I would like have trees more simplified, than per my vid from preview post.

Also, I would love abuse ability of DOTS.
Ehh, I think I made myself a bit carry on
But that helps me think about core mechanics and where I am heading vs where I want to go.

 

I've be interested to play with an MVP at some point. Maybe a god game where you can change the environment or terrain and see how it affects a region's evolution.

 

Yep, dynamic terrain changing was something in my mind for a while. I was even thinking about an asteroid impacting the surface, killing nearby lifeforms and leaving a crater.
This isn't something however, which I am going to implement anytime soon. However, I know how I could do that, with current mechanics.

Elevating and reducing terrain levels, could help building land bridges and barriers. I.e. separating from enemy factions, by making water channel. Or placing mountains. Or simply to gain more convenient ground, for specific species.

All these is doable at this point. Which is really cool thought. Could be as part of God's skill set, if going this route.

But not the priority focus at this point.

 

Use Cache Server to reduce time between switching platforms. I think it works for all platforms. Let me know if it doesnt work for WebGL.



Read more about it here.

 

Looks like this was nice and busy Feedback Weekend
I hope that everybody gained some useful feedback.

Thank you for you participation and please come back in next Feedback Friday.