next »

[phi6]

So I've been working on this small game for a couple of months now, it is basically your R-Type inspired retro shmup. At the moment I am focusing on epic boss battles.

Pretty happy with the general gameplay so far, but I still need to add some structure to the levels and also implement the boring stuff (menus, achievements etc..)

Screenshots:





Early Boss Trailer:




Also my pixelling feedback thread (although I haven't implemented the improvements yet):
http://forums.tigsource.com/index.php?topic=27388.0;topicseen

My question to you guys:
What do you think of modern bullet hell games? In my opinion they look very pretty but everything is scripted so it is much more of a memory game. With mine there is a lot of random behaviour and AI that reacts to the player. Also enemy spawns are random, so the emphasis is more skill based (I hope!)... every encounter is unique.

And I have yet to determine what the final boss will be

[phi6]

Hey everyone, I've picked this game up again after many months of procrastination - a lot has changed in the past few weeks and I've had the graphics completely overhauled. I got a little bored of the monochrome green!







And a mockup of a potential final boss

[Udderdude]

Any chance this is going to be on PC as well?

[phi6]

Ah - just edited my original post. This is intended to eventually be multi-platform (PC, Mac, iPhone and Android).

The game is built in AIR, and GPU accelerated with Starling.

Thanks for asking!

[Gregg Williams]

My question to you guys:
What do you think of modern bullet hell games? In my opinion they look very pretty but everything is scripted so it is much more of a memory game. With mine there is a lot of random behaviour and AI that reacts to the player. Also enemy spawns are random, so the emphasis is more skill based (I hope!)... every encounter is unique.

And I have yet to determine what the final boss will be

I think memorization is a core aspect of general shooter design. It is what allows for mastery of levels, and zen like flow to develop to a large degree.

[Udderdude]

My question to you guys:
What do you think of modern bullet hell games? In my opinion they look very pretty but everything is scripted so it is much more of a memory game. With mine there is a lot of random behaviour and AI that reacts to the player. Also enemy spawns are random, so the emphasis is more skill based (I hope!)... every encounter is unique.

I used to think the same way, but then I realized that the randomness really just lead to a less focused game with uneven scoring opportunities, and had more chances to create impossible scenarios for the player, depending on how difficult it got.

I'm saying this as someone who released a bullet hell game with lots of randomness and eventually removed all of it.

[Tondo]

Mmm, I like what I see. I kind of liked the old graphics, but hey that's just me being a hipster, so you can ignore that.

Also is there some form of demo we can try out for PC? I'd love to try this out.

My question to you guys:
What do you think of modern bullet hell games? In my opinion they look very pretty but everything is scripted so it is much more of a memory game. With mine there is a lot of random behaviour and AI that reacts to the player. Also enemy spawns are random, so the emphasis is more skill based (I hope!)... every encounter is unique.

I don't think the random aspect is a good idea. I do agree that the bullet-hell concept is in its core a memorization game, but I don't think that's generally a bad thing. The strength of the non-randomness of bullet-hell games is that the level designer has full control over how he wants the levels to feel and progress. Personally I'd rather have a very well designed level than a level that's randomly generated and surprises me sometimes.

[phi6]

The approach I've gone for is not completely random though, there's still structure throughout the levels. For example, what kind of enemies, how fast they move, the type of bullet patterns and number of bullets is pre-determined. There is also a clear difficulty progression curve.

The random bit is where exactly they spawn, bullet aiming chance, exact placement of power ups etc... The levels still feel familiar and structured between plays, but the exact positions and probabilities will be random, if that makes sense!

[Gregg Williams]

The random bit is where exactly they spawn, bullet aiming chance, exact placement of power ups etc... The levels still feel familiar and structured between plays, but the exact positions and probabilities will be random, if that makes sense!

As a long term shooter fan, this sounds very annoying. It also likely impacts scoring. Its your game though of course, good luck

[phi6]

Seems to be a common suggestion for sure - when I've finished the first level I'll post it here for you guys to have a look.

Cheers

[phi6]

So I just had a thought. A lot of shooters have enemies that shoot at the player, or let off homing/seeking missles - does this not break the fixed bullet pattern rule? Since the trajectories will depend on where the player is, and that would vary every time.

Anyway, I've posted an early build here:

http://phidinh.com/startau/

You can see a bit more clearly how I've used the randomization here. Really, I don't think it is that noticeable and negatively affects gameplay at all - but perhaps I've missed the point of a shmup... Please give a good try and let me know what you think!

Note that level 2 is only partially implemented - but it does show some further ideas for randomization.

Cheers

[Udderdude]

So I just had a thought. A lot of shooters have enemies that shoot at the player, or let off homing/seeking missles - does this not break the fixed bullet pattern rule? Since the trajectories will depend on where the player is, and that would vary every time.

No, because the player determines where the shots will go with their position, whereas random elements are out of the player's control.

Will try the test version in a bit.

Edit: First thing I noticed is the absolutely insane amount of player inertia/resistance.  That is a definite bad idea.  And it's got a huge lifebar.  You're making a euroshmup in Flash here ..

I'm confused as to why you even asked advice for bullet hell games when this is clearly not even close to that.

[Gregg Williams]

Tried the demo, I can't say much about random elements in that.

In general I really disliked it though.

1. Huge player inertia.
2. I really disliked the blue balls, as I kept thinking they we're bullets more or less.
3. Maybe just my preference, but I really prefer level layouts where with skill and memorization you can kill 100% of the enemies that spawn, its possible I really sucked at this level, but I don't think thats possible given what I saw. 

As to bullet aiming and homing stuff Udderdude really summed that up, player controls that, plus the spawn points of enemies are normally fixed so they're also often destroyed quickly.

[firebelly]

This is looking awesome!

[phi6]

Thanks very much for the feedback everyone, I will try and take your comments on board for the next build.

What exactly is a "euroshmup" and why is that a bad thing?

[Udderdude]

Euroshmup is a term referring to shmups with a few outstandingly blah features, namely loads of inertia, lifebars, huge hitboxes the size of the entire ship sprite, boring levels filled with popcorn enemies and not much else, wimpy ship firepower, enemies that don't have shot patterns but instead spray bullets randomly, etc. etc.

It's so named because there were a shitzillion of them on the Amiga (and to a lesser extent, the PC).

The reason behind the first two being bad is that the inertia makes it harder to control the ship, while the lifebar acts as a 'lazy buffer' allowing the levels to be sometimes impossible to avoid taking damage.

[phi6]

Euroshmup is a term referring to shmups with a few outstandingly blah features, namely loads of inertia, lifebars, boring levels filled with popcorn enemies and not much else, wimpy ship firepower, enemies that don't have shot patterns but instead spray bullets randomly, etc. etc.

It's so named because there were a shitzillion of them on the Amiga (and to a lesser extent, the PC).

The reason behind the first two being bad is that the inertia makes it harder to control the ship, while the lifebar acts as a 'lazy buffer' allowing the levels to be sometimes impossible to avoid taking damage.

This sounds exactly like my game so far!
Must admit I am a little bit disheartened by this...

[motorherp]

I dug up an old post I wrote for someone at the old shmupdev forum who was facing a similar problem to yourself, as in they wanted to write a shmup with a more classic feel but were falling into all the bad euroshmup design pitfalls.  Hopefuly you'll find this useful, and if you have any questions feel free to shoot them my way, I'm a long time fan of shmups and used to run the shmupdev forum for many years so I've plenty of experience with these things:

Indeed, the term euroshmup doesn't so much describe its area of origin but rather has become linked with certain design and gameplay decisions which are detrimental to the core concepts and mechanics of classic shmup games which make them feel broken as a result.  It just so happens that many of these games originated from europe and hence how the phrase was coined.

I dont really intend it to be a dig at your game, but rather just some advice in that you should consider what your target audience is and cater the game to their expectations accordingly.  At the moment it seems fairly clear to me that your game is heading down a casual shooter path which wouldn't appeal to genuin fans of the genre but could possibly be entertaining to casual gamers looking for a 60p iphone app to while away a train journey.  There's nothing wrong with that if that is what you want but then you should probably commit to that path entirely.  If your intention is to try and throw in more features to appeal to classic shmup fans then with your current foundations you will find yourself sat on a fence between both markets and your game likely wont get any recognition from either.

If you do decide that the classic shmup is what you really want to acheive then I really do recommend you do some redesign now before it becomes even harder.  One of the biggest pulls for the classic shmup fan is that the game should feel like you are constantly on the edge of being killed and its only your skill keeping you alive, but when death does come it should always feel fair and avoidable with more practice.  Having inertia and health/shields really detracts from this feeling since you are no longer in full control and not very often at risk and hence skill doesn't feel to play a large part in the game and isn't rewarded sufficiently.

Cool, I'm glad to see you're taking people's advice on-board.  I wouldn't want to discourage you from making the game you want to make though.  At the end of the day, unless you're making this game for someone else or to make money, the most important thing is that you make something you enjoy playing yourself.  Its just that I've seen similar design decisions made many times before and they've always resulted in games that haven't felt very shmupy (new word?) beyond the fact that you shoot stuff and stuff shoots back, and thus they aren't generaly well received by fans of the genre.  If one of your aims is to have your game feel like a classic shmup then its always worth while learning from the short-comings of others and getting in tune with the core ideas and mechanics of the great games.

If it helps, in my mind the great shmups tend to unravel in various layers the more you play them.  The first taste you get is like I described before, that the game is challenging and contantly makes you feel on the edge of being killed and only your raw skills and reactions are keeping you alive.  These are the moments that have you on the edge of your seat and afraid to blink.  Also like I said previously, in order for this just not to be fustrating it also has to feel fair, that you are in complete control and if you die its because you messed up.

The next layer then reveals itself as you gain more experience with the game from replaying it.  Sections which you previously felt very difficult, you start to discover routes through the mayhem and methods for tackling the enemies so you can start to successfully navigate those sections repeatedly with little loss of life.  This is the reward for your repeated efforts and application of skill and you start to feel a sense of mastery.  This is why most shmup games have a largely linear format with fixed level layouts and recongnisable enemy types and formations and their associated bullet patterns.  This repeatable nature ensures that you are rewarded for learning the ins and outs of the game since you start to be able to predict and anticipate the action which increases your chances of survival, and previously succussful playthroughs if repeated will work again.

The really great games often feel like puzzles in this respect.  You must first brave the carnage relying on your reactions alone and often a safe way to navigate the patterns and enemies isn't immediately obvious.  With repeated attempts though you can try to tackle the section in different ways and thus try to puzzle out the safe routes and required maneuvers.

Lastly, after mastery of survival of a certain section or level is acheived comes the scoring layer.  This is when the player is encouraged to take more and more risks to improve their scores and prevents already mastered levels from becoming stale too quickly.  Its important to get a good risk vs reward scheme in place for this final layer since this will be the final differentiator for competition and self improvement.  The scoring is where the real meat and longevity of shmups recides and long time shmup fans often consider this to be the most enjoyable and important aspect, but dont discount the importance of the other foundation layers either.

One of the great things about shmups is that you'll often experience all three layers through-out the course of a single play-through.  The earlier levels you've played repeatadly will be all about taking risks and scoring, the furthest level you can reach comfortably will be all about puzzling out routes and enjoying your growing sense of mastery, and the final levels will be a heart thmuping struggle for survival on your skills and reactions alone.

I hope that helps.

[Udderdude]

I don't want to make it sound like your game is beyond repair and should be tossed out the window.  But you need to take a good, hard look at modern and classic arcade shmups, and see what made them work.

[phi6]

Thanks guys, and motorherp for providing that post. Can you link me to the original discussion so I can have a read?

Well, the good thing is I'm still happy with my engine for the most part, perhaps except the inertia and lifebar but that can be fixed right away.

Okay, I'm going to try creating a level without any randomization with one-hit player death, as an exercise in level design. If I can do this properly, hopefully it will be easier to scale up the lifebar and random elements from this base.