Want design feedback for your new game? Then you’ve found the right thread! Post here to get precious player feedback. Discuss until next Friday, when we’ll lock the old and start anew.
How To Ask For Feedback?
Show - Be Interesting! Pics, videos, or best of all, a playable game!
Be Concise - Who’s got time for Wall 'O Text? Less is more
I started a little side project a couple weeks ago. Some of you know I am a fan of more interaction and better AI far more than I am a fan of awesome graphics. I’ve often thought it would be fairly easy to make an interesting scrolling shmup simply by focusing on the AI and interaction.
I figured just some basic stuff would make a big difference. You know in shmups you basically just fly along obliterating a mass of enemies who seem to be little more than mindless drones, heck they could often be just a chunk of asteroid floating through space (and often in fact are just that). They just float across the screen happy to be nothing more than cannon fodder.
So, I’ve always thought why not give them a bit of intelligence? I mean surely these enemies would at least attempt to defend themselves at the very least trying to move out of the way of the players projectiles. And I think a player would feel much more satisfaction from obliterating these somewhat intelligent enemies than they get from obliterating masses of what could basically be driftwood floating down a river.
So, this project is an experiment to test this theory.
I only have 18 hours into this project so far and much of this time was spent just knocking out some simple graphics (notice the veggie-inspired space ships?) and building the parallax scrolling playfield (tile maps) just to get that stuff out of the way.
So let me present Simple Scrolling Shooter:
These green enemies are your standard shmup enemy. Just move in a pattern across the screen. However, there is some slight difference with interaction. Engage to find out. Nothing major. These dudes will be the ones releasing powerups eventually.
These red enemies are smarter than your average shmup enemy.
There is only about 4 minutes worth of gameplay in here.
After that time no more enemies will appear and the playfield will continue scrolling until the end (I think there is about 12 minutes worth of stage in this level, not sure now to be honest) and then it will wrap back to the beginning.
So, my question is… what do you think about the gameplay experience?
Do you find it more satisfying to engage these slightly intelligent somewhat more interactive enemies… than it is to engage the normal shmup enemies?
The best shmups do have enemy AI. The demo is not bad at such an early prototyping stage but it needs more development to see where you’re going with this. Often context makes a difference, at the moment one isn’t really engaging with the ships much, the gameplay in it’s current state is just too short, the enemy AI just too simple. The ships dodge then run off and can’t be followed. Could the gameplay be more satisfying? yes, could interactivity provide this? yes. Does it currently ? No.
The idea is definitely a good one, the AI likewise for what it is, but it’s just too short in length and too simplistic to really judge
From what I can tell people like the boss battles, the crazy insane bullet patterns dodging the bullet hell. I don’t like shumps myself But I watched slowbeef play a bunch of them.
Ha ha! Thanks for the feedback. Yeah, I made this mistake in the past throwing out a very early WIP for feedback. This is just a side project I am messing around with but I thought I had enough in to show the differences between the standard shmups enemies just moving in, stopping, firing and flying off seemingly oblivious to the fact they were even taking damage. BUT… it is probably just way too early to show that at this point.
Anyway, thanks for giving it a play. I appreciate it.
See that is perfect example of a standard shmup. The screen is filled with crap. It is basically just an overload of graphics and FX going on. Half the time I cannot even see what is going on in that video other than things are blowing up. The other half when the view is clear I can see the same ole enemies moving across the screen seemingly completely unaware they are in danger of being destroyed. Even the bosses seem to be completely unaware of anything other than their pattern.
Basically all shmups are like this so I don’t see the need to make another like them and want to explore a different way. I will be back to my main project this weekend but maybe in a few weeks I will be able to throw out a new WIP that better illustrates what I am exploring.
Yeah I guess the reason for throwing out so many bullets is because the enemies are so dumb that thats how you have to do that to make it somewhat of a challenge (but I have to wonder if players like that sense of feeling like they are so good at dodging the bullets along with the “pretty bullet patterns”). I think it could be interesting take on it, imagine if you had bosses for example that actually learned how you play rather then I am just a dumb boss doing my dumb boss pattern.
Hi everyone! It’s been a while. As promised, I have perhaps the smallest possible demo I’ve ever released, but with arguably the most important question.
Sara the Shieldmage - First 30 Seconds Demo Unity Webplayer (Works on IE)
Controls that Matter:
Mouse - Click Buttons
(WASD still works, but there’s nothing to do with them in this demo.)
The Big Question:
Click “New Game” from the title screen. Does the first few seconds of this demo give you any reason to want to see more of the game?
If not, could you go into more detail about why? Alternatively, what would you like to see instead of the opening as it exists in this demo?
I went back to play a little gradius for comparison (since it seemed more in line with the style you’re trying to ape), and you’re “more interactive” enemies aren’t there yet. They still seem to be pretty flat as their patterns are pretty one dimensional. Right now it seems like you’re basically trading complex movement patterns for simple interaction patterns.
I agree with your sentiment, to me I would change the context from a normal shmup where possibly the player is a bad guy and is attacking a convoy and can choose the best moment to attack, and then the other ships would evade and have adaptive AI which tries to save as many of their ships as possible, which might mean the player is basically the space equivalent of a wolf pack during WWII. That could mean being able to scroll as far back as forward so the player could stop any fleeing ships, overcome escort ships etc, maybe allow for space debris traps etc. As the demo now stands it has very rudimentary AI but gives no indication of what the idea really is. The point of shmups is to fill the screen with crazy bullet battles etc. If you want to change that normal gameplay then I think you will need to flip things around a bit, not only change the ship AI but maybe experiment with the format, keep the side scrolling aspect and the shmup format but switch things around within that format to use it more as a battle arena, maybe going to the extent of flying from right to left first to lay one shot traps that will all be about the tactics of the situation. I don’t know, really. Shmups litter the screen with enemies and bullets, at which point it becomes difficult for a player to then deal with ships that may also dodge his bullets unless the shots are timed right. Might be one task too many.
It certainly worth experimenting with the way shmups do stuff normally.
Not really. Why? because you tell, and not show. Why not have a camera shake and a few falling rocks in a corner, then shorten the dialogue about how dangerous the place is. Anyway, that’s just opinion and everyone has one. There’s nothing intrinsically wrong with the setup. I would suggest you go away get further with it and let an idea for a grabber start scene gestate in your head for a while, then I have no doubt a couple of weeks down the line you’ll have a flash of inspiration that will bring it all together and give you the perfect start point that will sell the game.
My recent game I’ve been working on for 15 months and two weeks ago I got that spark for the opening scene. Unity makes those sorts of changes easy to implement so I would just go away and continue working and have a bit of confidence that the opening you want will come along after the game is much further along.
There’s quite a bit to like about the scene, I personally thought the art style was distinctive and the walk animation was really fluid and smooth and quite pleasing. It’s all good, so just get further with it and revisit the start scene from time to time until you figure out how best to do it.
Ah… Nope. The first issue is it took me a while to figure out what they were even talking about. You definitely told me something just happened. The other issue is who is talking? Sure, you only have one important character, but more indication of who is talking will help a lot.
That is absolutely spot on. I didn’t want to bring that up but was hoping someone would mention it. You answered a main question I had. I wondered how much interest in generated from the movement patterns in shmups. I think it is an important part of the games and wondered if perhaps even enemies with some intelligence would seem no more interesting (and perhaps not as interesting) than truly “dumb” enemies moving in interesting patterns.
Thanks for giving it a play and providing your feedback! I appreciate it and will be sure to combine movement patterns with the AI going forward.
It works fine for me. I am venturing into some unstable ruins in search of the Feldhold Codex and have two guards with me. I know it is an ancient relic of great importance because this relic is physical proof of the existence of a Goddess who is apparently quite well known to these people although I have no clue yet who it is.
I ran down the hall and when the fadeout occurred I was looking forward to exploring.
Some suggestions for improvement:
Like @RockoDyne mentioned, I think some indicator of who is speaking is important. It does not have to be anything fancy. Just a little ! exclamation mark or something over that character’s head would work fine for me.
An intro before this that is simply a paragraph or two of text introducing the Goddess and the impact she had on this world including a very simple mention of the artifact (when was it first seen, etc) would help a lot to set up your current gameplay opening.
Try centring the text on the player whos talking like they did in FF7, but by that point in the game everyone would already know who’s talking anyway. So maybe also an arrow above the talking person’s head.
@GarBenjamin , I gave that your shmup a try. The red enemies were a little harder to hit, but I found the best way to attack them was to fire one shot as soon as I was lined up with them, then line up with their new position and spam the X and Space keys until they died. If you continue with this, perhaps you could make them move a little earlier and a little faster.
@Asvarduril, the main problem I had with your demo was that the development console threw up lots of errors about SkinnedMeshRenderers that partly hid the text. However, once I got that, it gave me enough context to know what I was doing in some unstable ruins. I’m also going to repeat what everyone else said about not being unable to tell who’s actually speaking, even if they are nameless redshirt guards. Perhaps you could try rotating them to face the other as they speak (i.e. Guard 1 turns to face Guard 2 when speaking, then turns to face Sara when finished).
If you want to add a few more layers to it, some more things you might want to consider are enemy-enemy interactions, especially behavioral ones like they might act differently in formation, and a “wave director AI” that would be controlling and planning the waves to maximize the occurrences of effects.
Great stuff! Thanks. I was considering the idea of some kind of global AI to oversee the big picture instead of just relying on the local AI. So far there is only local (individual enemy) AI. Mainly because it was super easy and quick to implement. But, more than that, I also considered having the enemies communicate with each other: if player is damaging enemies the enemies would call out for assistance from the other enemies (including of different types) who would then respond to the call for help. I am still deciding between these two different approaches but actually… having a sort of General (Master AI) could mesh well with this. Basically, enemies would be dialing 911 and the Master AI receives the call and it then could dispatch help by contacting one or more units and asking them to change their behavior.
I am not sure how much this would be noticeable within the structure of a shmup as @sicga123_1 mentioned. In my opinion, a big issue with shmups is they have evolved to basically be bullet hell and nothing else. So I know I need to address this first by adding some depth in general. I really think the enemy force needs a bigger objective than go forth and destroy until you are destroyed. That greater goal is what most shmups do not have.
I think it may start with the player objectives. So what I would like to do is give them some purpose beyond just blow crap up as fast as you can and stay alive as long as possible. Capturing certain enemies could work well. Destroying certain structures is a possibility . Deploying troops is another objective that could work. Guarding a civilian vessel is workable. Once the player has a concrete agenda so all of the blowing up of enemies becomes a means to an end the enemies will then be in a better position to make use of some intelligence.
However, this can be applied to the enemies as well. Maybe they are the invaders and you must protect buildings or perhaps they are mining and stealing crystals or something. Some kind of reason for their existence beyond just serving as cannon fodder
As you can see I have some general ideas but this is all in the very early stages and nothing is in concrete yet.
Everyone - That was some awesome feedback! I knew it wouldn’t pass this round, but the reasons were an interesting surprise. Here’s what I took away:
Not enough is happening - There’s no feeling that the party just survived a cave-in, everything is too calm ( @Gigiwoo would say my scene isn’t very juicy; he’d be right.) The characters don’t move enough to tell anything about them, which dovetails nicely into…
It’s hard to tell who is talking - I need things that tell the player who’s currently talking. Sara’s easy - she’s the only one with two X chromosomes in this scene; Guard A and B are a bit harder to know about, more so that I haven’t began coloring sprites (or, even creating most of the lineart.)
Too much talking is occurring - I would do a better job of selling the “Oh, Crap!” moment with less text. I would do a better job of framing the scenario with less text. Less, but more effective, words, are going to help this scene.
As a result, the first few seconds aren’t “grabby” enough - 'nuff said.
@sicga123_1 - I appreciate the notes about the visual style, and that there was a lot of good in the 30sec demo. You mention that there’s nothing intrinsically wrong with the setup, but I get the sense that if you were the one doing this, you might do some things differently (above and beyond the whole Do > Show > Tell thing.) Keeping in mind what I’m going with for Sara being sort of an adventure archaeologist who can use shielding magic, who is on the hunt for the greatest religious find evar!one!!™ in this game world, how would you structure the intro? I ask because, as noted in another thread - storytelling isn’t my strong suit. In fact, the whole reason for this demo was because I’m doing some I’m not good at, and I know it. Spare no detail with your explanation; assume I know nothing about videogame storytelling (you’ll be generally right if you do.)
@Aiursrage2k / @RockoDyne - I was thinking having a speech bubble appear above the character’s head; I’ve seen such a thing in Radiant Historia (which, funny enough, also helped inspire/validate my mixed 2D/3D visual approach.) In your mind’s eye, would that be helpful? (And, @Aiursrage2k I haven’t forgotten about diagonal movement. I’ve had enough gripes between two projects now to do diagonal movement that I’m going to have to cave in and implement it sometime.)
@GarBenjamin - I actually don’t want to talk overmuch about the Goddess in this opener bit. The fact that you expressed wonder about who she is means I dropped just enough information; a player wondering who the Goddess is is a player I suspect would be more likely to explore the world, which is what I’m going after - Sara the Shieldmage is an adventure game conforming to Eastern RPG mechanical tropes. I want the player to wonder about some things right off the bat; they will be explained/set up before the end of the Feldhold Ruins introduction sequence. Other things of course won’t be explained until you start doing sidequests.
