Feedback Friday #148 - January 15-18, 2021

Hello everyone!

We had a bit of a quiet one last week, let's see if we can get things cranking this time around

For this week's bit of encouraging game design news, check out this video tweet showing the development of Jedi: Fallen Order, an incredibly popular and well-rated game (some might say the best Star Wars game for a very long time!). You can see how similarly the the levels play, even though one is just a pile of cubes and the other is the AAA result.

https://twitter.com/80Level/status/1328363114748850176

Spoiler: Blockout vs Result

Great game design starts at a very basic level! If you're looking for free design feedback for your work in progress project, then you've found the right place. Feedback Friday runs from Friday to Monday every week.

What To Show

• Minimally 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 #147 is here.

 

Moving took longer than I expected and I ran into some issues and had to refactor some things (still more to do), so I don't yet have my "real" level fully working. The images below show what it looks like (the first is generated from the grid representation of my background simulation).



Interested in feedback on the design, and suggestions for filling up space.

This is meant to be the "hub" of the space station, and the station is of course a research station but also acts as a "rest stop on the long, long highway between planets." The 12 "indented" areas lead to different modules.

Spoiler

Hanger/Inprocessing
Rest/Sleeping quarters
Entertainment
Food production
Research labs
Cargo Storage
ATM - module with constant poison as the atmosphere
Med Bay
HOT - module with constant high temperature
COL - module with constant low temperature
Security and Power
Bridge

The 12 rooms have four general types: a store, a restaurant, a bathroom, and a medical facility like doctor's/dentist's office (how obvious are those from the silhouettes? Does that matter?).
The half-curve in the middle is for administrative assistance, and radiating outward that's a waiting area, with seats (with small tables in between each).
The room in the middle is for administrative duties, and will have an elevator going down to other areas.
The yellow areas are the interactive zones around doors (my intent is to change doors to automatically open when you're in the zone).

(for anyone seeing this for the first time, the game is basically a top-down shooter available here)

The edge rooms seem like they'll provide interesting gameplay, but the center space seems too open to me. Does that seem like a problem to others? Does anyone have recommendations on what I might fill it with, given the intent of the area?

 

Why not make the edge rooms bigger and connected? This would reduce the empty space in the center, and the connections between rooms might add more interesting possibilities.



(The black dots represent doors.)

 

That...makes perfect sense. I'll still want the areas leading to the modules to be separate, but that's just more walls, doors, and thus interactions.

 

I think some potential for spill-over would be interesting. So if you leave a door open between two rooms, or if an enemy can somehow open/destroy a door, the blob could spill out from its original room.

 

Hmm, I was thinking I wanted 24 total rooms, but that's too many I think. 12 looks better.

 

I agree with TonyLi about extending the rooms inward a bit, but my instinct would be to put something 'spectacular' in the center (since it's the center of the station and has some kind of symbolic value) such as a reactor core or something.

I think I can tell which rooms are the bathroom and restaurant, but I'm not sure it makes a lot of difference unless you plan to add more context.

I had a quick play of your linked build, not sure if you're looking for feedback on that? Maybe I'm doing something wrong but the enemies basically just reproduce until they're overwhelming. Pretty soon the place is packed and I can hardly move around. Am I supposed to be pushing forward rather than clearing out rooms? What's the objective for this level?

Just a side note, the doors are basically invisible when you approach them (because the camera is looking straight down on them). Would be good to have it clear where they are even when you're right next to them.

 

So the "reactor core" type thing is actually going to be below on a separate level, and it's kind of tied to a secret bad ending the player can get (basically blowing up the station instead of clearing out the blob), but I like the idea of having something impressive in the middle. Not sure what it should be.though.

The difference between the different rooms was really just a check that I was doing an okay job of building recognizable spaces, rather than just a bunch of blocks that don't make sense.

Always happy for more feedback on the gameplay. The idea was to go to the furthest back room and pick up the McGuffin, fighting enemies on the way, but that was really just an excuse to experiment with the enemy types. This hub map is really the first attempt at building something that will be in the final game.

And yeah, as I mentioned I'm gonna try having the doors open automatically when in range, though it's still an issue for certain situations where doors don't open...I'll have to think about how to make that clear.

 

I had another play and retrieved the diode - might be a good idea to tell the player what the objective is. I also managed to guess the 'interact' button but maybe add that too.

Since you want the visual elements to be simple, I think it would be really good to focus right now on clear communication to the player through level design, bits of text information, visual and audio feedback, and/or whatever you plan to focus on.

For example with the level design, there is really only one path to the diode room, and it's not direct (basically you need to go clockwise around the map). That's an instinctive thing when designing a level - to make it not a simple thing to traverse - but when the path is still linear, it ends up feeling simply convoluted rather than tactical.

Also, when I started the level, I assumed that I had options for where to go, and instinctively went to the bottom right for some reason. I found myself in a dead end room, from which I had to reverse and go back into the main room I had already passed through. It was like the game saying "no, you're wrong, go back and find the correct answer".

Here's an example of a level that I think would be more tactical, which is what I would like to feel when playing this game.



The advantages are:
- No matter which way the player goes, there's a route to the objective. The only room that doesn't lead anywhere is in a completely different direction, but still offers some reward (maybe some equipment or perks).
- There are different routes for different play styles. The player can go full Arnold through the middle, do a tactical room-clearing route at the top (which offers potential for varied enemies and tactics in each room) or sneak up at the bottom and sprint to the diode room from the bottom right before anyone notices.
- The player doesn't have to return to the start. I'm not sure if this is something you want, but I'd avoid it unless something fundamentally changes the dynamics of the level after the player has reached the objective (e.g. previously locked areas are unlocked, some event makes enemies move to a part of the map etc).
- The map has replayability. Not only can the player try different routes/approaches, but the reward at the top left might give the player something that changes the way the whole level plays, so you'd essentially have six ways of doing the level.

 

I do wonder, is this kind of layout of rooms you going after?
How that fit to the circlic layout, you presented at beginning of the thread?
You are going to change, to be more in circle layout?

BTW, what I would try to do is, do some mixing up here.
Take this case

I would divide station not to 12, but rather 6 distinct areas, plus center. But only one, or few areas to be connected with the center.
Now, each area is connect with another neighbour area. Or at least most. You could have some areas, which is connected only with center.
Then having each area, which made of few rooms, as you shown above. Therefore, you can clear each area individually.

You could consider areas as stages. Or sub stages.

Another thought is, you could have multiple rings of your base.
The center, inner rings and outer ring, with entry hatch.
Each ring as described earlier, would be decided into areas, and interlinked strategically.

Clearing areas would require some strategic thinking, to gain bonuses, upgrades or switching areas features. For example backup electric power area, which allow access to other areas.
Or collect tools area, so you can access other area. Or gain enough upgrades, to be able withstand enemies boobs in other areas. Allow to go back to previous areas.
Keep them cleared, or allow to complete them again, to allow get more XP. Need experimenting with that. But clearing again, should be less rewarding. Or allow for better achievements, when coming back to clear it.

Allow some secret, or special areas, with extra challenges and bonuses. Some areas may be harder to complete, without specific made progress by the player. But may allow for more rewards.

Regarding center, either make it to be Station core. Or make it to be a command center.
Then next set of stages and areas Could be at lower deck.
That could be like clearing set of stages, before progressing to the next stage. Meaning new set of rooms.

Or, do as you please

 

Yeah my reply is toward the level you play in the current build linked in EternalAmbiguity's post. I'm not sure if that build has anything to do with the circular room idea.

 

Billy4184's level design is more typical, and for a reason -- it provides interesting pathing choices.

But the wheel idea could do some unique things, such as rotating (since it's a space station) to change the layout during play.

 

Personally I wouldn't go that route. Seems unnecessarily complexity, for little to no gain.

 

Sure! My suggestion wasn't specifically a response to the post (which focuses on the circular room) but rather feedback for the current build that was linked, which features a non-circular room.

I'm assuming there will be a lot of levels, and not all of them will be circular.

Overall though, I think it's more important now to focus on creating solid core gameplay rather than the circular room, which is a bit of a game design issue on its own. Fleshing out the different kinds of AI, and determining the interesting and varied tactics that the player will need to defeat them, is where I would suggest putting a hard focus. For a game like this with a top-down view and abstract visuals, it's begging for more tactical and skill-based gameplay in my opinion.

 

With this hub design I've been coming at it from the perspective of "realism," or realistic at least. Just building out a design that makes logical sense for the space station. While the hub here is circular, as I mentioned before I'm planning on these 12 (for now at least) modules coming off of the hub, and those will be more rectangular areas with more traditional rooms. They'd still support something like Billy's layout. And most of the "goals" will be related to those modules, so i should consider things from that sort of layout.

I really like some of what you present here.

Based on the module types I list in my spoiler in the first post, which would you remove? Right now I have specific story or gameplay related events planned for Hanger/Inprocessing, Research labs, ATM, HOT, COL, Security and Power, and Bridge. The others I don't have anything specific for, but since I'm going for a realistic design I feel like they need to be included. Though I like the idea of splitting them off into being connected to one another and not necessarily the hub.

I am planning for the ATM, HOT, and COL modules to require player upgrades they won't have at the start, and I'd also like to have health and "energy" upgrades as well just scattered about.

Heh, so here's the thing: the conversations I had in the last couple of FFs (before your last one of course) led me to think I'd need to work on the level design to really tease out the enemy AI stuff. The behavior of the enemies* and the room modification system * and how the player interacts with those feel like elements that really require a decent level design to experiment with them properly.

It might have been better to pick the hanger/inprocessing area to start out since it'll be more traditionally shaped, but the idea here was to get rooms and layouts that I'd have in the real game, and then play around with the gameplay.

* for behavior of enemies, there's a lot going on. You have the blob "grower" who just grows constantly in all directions. You have the blob melee character who will follow the player and try to attack when next to them. You have the blob shooter who will move around randomly and fire bio bullets. You have the "machine" faction turret that sits in one place and fires, and you have the robot who patrols and shoots enemies with energy bullets when they're in view. Plus a certain other enemy. The blob and machine factions attack one another.

* For the room modification system, there are several modifications you can make (toggle power, make hot, make cold, fill with poison, fill with antiseptic) that all affect the player and enemies in different ways.

So there's really a lot there (and all of that is present in the demo online), but I felt like I needed a better level design to play with those interactions.

 

I definitely agree that the level design and AI are a chicken and egg problem here. Right now perhaps the enemies are too clumped together to be able to deal with them uniquely. Do you think it might be beneficial to create some test levels that are each designed around fighting one specific enemy? That might help clarify the relationship between an enemy, the player, and the level.

By the way I did like the turret room in the top right, that's where I think rooms shine because you can focus each one on a specific enemy for distinctive gameplay and variety.

 

So another thing I forgot to mention is that the melee and shooter "blob" enemies grow from the the blob grower, so in a sense it makes less sense for them to ever be truly alone...but it's still possible for them to move around, so perhaps there will be situations where the player just encounters one of a given type.

For these test levels should I go for the kind of detail you presented with your layout?

 

Not necessarily I don't think, each test level might be a small one designed for a specific enemy (or group of symbiotic enemies) to clarify a sort of 'distilled theory' of combat with that enemy type.

Do you have a specific thing that you want the player to do with each type of enemy? For each of the ones you mentioned, can you describe what unique way you think the player would best handle them?

 

Hmm. I hadn't planned out the player's offensive actions at that level. Right now all I have for them are different types of bullets they can shoot, but it's still just pointing and shooting. I do have a little bit for defensive actions though.

Blob grower - nothing really, they're immobile and can't attack.
Blob melee - avoid them as they track and close in on you
Blob shooter - use level geometry to hide when in range of them
turret - save as above, though they're fixed in place so there's a bit more freedom
robot - hide from being shot, as well as move to avoid them closing in on you
(I was thinking about adding a stealth element where the robot can only see one direction, but that seems contrived)
secret enemy, corrupted robot - RUN

For the bullets - there are "bio" and "energy" bullets that each faction has resistances against, so the player has to choose between those, but that's not a strategy. The different enemies have different health levels, which of course affects the encounters, but I hadn't thought about that in terms of informing a player's strategy.

The room modification system is another layer on top of that - certain modifications kill enemies outright, some affect their movement or growth...but the suggestion I got before was to keep that on the backburner and focus on direct combat.

 

Consider writing down some "magic moments" of gameplay that typify what a player in an ideal playthrough would experience. I feel like there are lots of good ingredients here. I just don't have a sense of how they really come together into unique gameplay moments. The act of actually writing down some hypothetical watercooler story moments may help us understand the overall experience you're trying to give us.

 

I feel like on some level I'm subconciously restricting myself to situations that are already supported in some way, but let me know if these seem interesting or if any of them need more detail or player actions.

This one seems most lacking to me, and implies there should be descructability or maybe turrets should be able to summon robots. Needs something more.

 

Don't force yourself on being "realistic". There isn't really anything realistic about your game. Me not being sarcastic or anything. You got for example some blobs which are anything but realistic. Just keep it entertaining and enjoyable. And having sci-fi elements is perfectly fine. You don't need toilets for example, which may be more important, than other type of areas

Like with gravity discussion, that we had previously, you may be over thinking again.

So my point is, any of these elements can be perfectly excluded and replaced for anything else. They don't destroy meaning of the game.

You can always change numbers of areas, even make them uneven if needed, 5, 7, 9. It is just a bit harder to draw.

 

I like the watercooler stories that were close calls. They didn't need to be so narrative, but it doesn't hurt. Second to the close calls are the ones where the player feels clever for having planned out a good strategy. I think those are the kinds of moments that it will help to plan the gameplay around. If you can make the whole game feel like a push-and-pull of close calls and victories due to clever planning, it will be exciting to play.

 

Alright, thanks for the help folks. I'll take another step back and look at the gameplay for each enemy individually and then go from there.

 

This is cool.

You say 'avoid' the clouds, what about having clouds that drift across the room slowly, making the player have to dodge and keep moving.

Right now one issue I think with the poison clouds is that due to the number of enemies and the clouds hanging there for a long time, they become a static obstacle and make the level feel more restricted without introducing excitement.

I like this one. What about creating a level with, for example, a lot of melee enemies in the center, but around the edges there are doorways, corridors, and various obstacles that you can draw them into to create an ambush/trap like you just described.

If these enemies are 'dumb', maybe you can lead them into electricity/power saws etc.

Also, if this is a space station, I'm not sure if you've considered the idea of having airlocks you can open and suck everything out of?

How would the spitters move around? What patterns/objectives would they use?

What about being able to throw something inside to distract the turret, and then run in and hide behind something?

This is very good, and has the potential to build atmosphere rather than just going in shooting. What stealth aspects have you considered? It seems like this is a very good setting for that.

For this scenarion, you could have a 'server room' with aisles of computer terminals, giving the player lots of opportunity to play hide and seek and do a lot of hit and run moves to wear down the robot.

Another opportunity for stealth and atmosphere. Is this a boss that you would hide from over and over until you had some opportunity to kill it (sort of like Alien Isolation)?

Overall I think these 'stories' already create a variety of mechanics and enemies to start fleshing out the game with.

 

Since there's a holiday tomorrow (for Americans at least) I uploaded a build. The main menu allows you to select an enemy type, and then there's a small map with a number of obstacles and rooms. Should give a chance to engage with enemies individually and consider improvements.

Interested in any comments or suggestions for improving them again on that individual scale.

Same link: https://eternalambiguity.itch.io/ca-shooter

 

I like these ideas, especially the traps (especially the electricity).

The spitters are just random walkers, before and after encountering the player. Playing through encounters with just this enemy makes me realize they're pretty boring, they need more...something. Adjusting their movement in some way seems like an easy change.

Good idea for the turret. I've been thinking about a "bomb" type of bullet or weapon, just a projectile that would land at some specific location rather than being shot in the air. This sounds like an opportunity to introduce that.

Yeah, for the robots right now they'll only attack the player if he's in view, if there's a direct line of sight (so the player can approach them if he stays behind cover). Right now the robot can 360 degrees around; I was considering reducing it to enable more sneaking around, but that seemed contrived.

The last enemy is meant to be very powerful, yes. It's a "corrupted" robot. I was considering making it systemic, where the blob can combine with a regular robot to create it and thus it can occur during regular gameplay. But I also like the idea of something like Alien Isolation or Metroid Fusion where it's a specific enemy you have to build up to.

 

I think this is a great idea! But I had some problems with the demo.

- Selecting the Grower didn't spawn anything in the rooms.
- Selecting the Shooter spawned correctly, but seemed to turn into a Grower when shot? Shooting it caused it to multiply and drop poison clouds.
- The Melee enemy was extremely slow (moved like 1 tile in 3-4 seconds).
- The Turret spawned correctly, is it killable? Seemed to take an awful lot of hits.
- Choosing the Robot basically dropped the fps to 1 or 2, completely unplayable for me.

But I do think isolating enemies and experimenting with them this way is a great move. I'd suggest spawning them in different locations (e.g. corridors, rooms etc) that seem relevant/interesting to kick off some discussion, rather than spawning one in the same place in the corner.

Incidentally I've been doing the same thing with my enemies, might have something to put into next week's FF.

 

For the grower I think I mapped it to the wrong state in my backend simulation so it wasn't producing the enemy properly. I fixed this.

For the shooter, yeah currently I have it set up such that when it dies it "bursts" into the grower enemy. That could be removed.

For the melee, yeah I felt the same, and I changed things so it's much faster now.

Turret is definitely killable, though keep in mind that there are two different bullet types (LMB, RMB), and the enemies are susceptible to one or the other.

For the robot - yeah something's up with the parallelization for finding paths, I'm currently trying to fix that.

 

@EternalAmbiguity looking forward to the updated demo. I'd also consider giving options to set stuff like the melee enemy speed, so people can play around and get ideas.

Gonna lock this one, see you in next week's FF!