Breaking Down the Design : Arkanoid

[svero]

I'm often talking about subtelty in design and what makes a good game different from a great game or bad game. Often I think it's the little things. (like pacman slowing down when he eats dots) There are certain genres that have been done to death now and anyone who's created a game of that genre has faced similar problems and solved them in different ways. Arkanoid is an example. For instance... The last brick problem. It can be boring to try and hit the last brick on the screen if you can't get the ball's angle quite right. Different arkanoid style games solve that problem in different ways. Super DX ball 2 blasts the brick away with lightning. Other games give you a "pass the level for free" bonus after some time. But what if it's 2 bricks? What is the best way to deal with it.

And what are the other elements of arkanoid that are important. Its easy to launch bonuses when you hit bricks. But how many? Should the ball speed increase as the level plays? How important are the various standard bonuses? Multiball? Magnet/sticky? etc...

If this thread works out.. then maybe we can discuss the details and problems that need to be solved in other genres and how people like to approach them. So who's made breakout? anyone have any specific opinions on the subtle details?

[luggage]

The spin on the ball is another element which takes a fair amount of fine tuning.

The pickup frequency is something that could change per player. I'd try (not sure if it would work) the following.

Create a frequency value of between 0 and 1. 0 should be they appear quite rarely and 1 should be they appear very often.

Start the first level at 0.6 for example and see how long the player takes to complete the level. Compare their time against an average time (you can log this throughout testing), if it's quicker then nudge the frequency down - if it's slower nudge it up.

The game should tune to their skill level. Whether it can do this quickly enough I don't know though! You could split it for good and bad pickups or even use it to tune ball speed, etc.

It would also be quite interesting to save that value in the registry so your other games can access it. You could take a peek - see that Arkanoid got a value of 0.9 (ie. lots of pickups so they can't be too good at the game) and start your new game a fair bit easier.

[dntoll]

the problem with few bricks left is a hard one:

In TimeBreaker we solve this by dropping weapon bonuses when there are a few bricks left... This is both good and bad, the good thing is that you allow the player to end the level by themselves instead of just ending the level, the bad is as my girlfriend put it: "Now the game thinks I´m a looser again, and drops those guns!"


(the screenshot is from another game I played today I think it was superdxball)

[Anthony Flack]

I think the problem is with the assumption that the object of the level has to be to destroy all the bricks. Only one breakout game that I know of (Poing) has moved away from this assumption, which is slavishly reproduced from one game to the next.

There are all kinds of other ways you could run it. What if the object was to destroy as many bricks as possible within a certain time?

As for having the ball speed up as you go - it's probably a casual game no-no, but I think it's fundamental, and its absence is one of the biggest boredom-makers in most shareware breakouts. I like it when the ball gets faster and faster

[luggage]

Can an Arkanoid game be too easy in the casual space? They can certainly be too hard.

[Nikster]

IIRC Arkanoid opened up a door on the right hand side which appeared when you collected a certain pickup. So it was optional whether you skipped busting the remaining bricks, but I guess you'd have gotten better bonuses for destroying all bricks..

On the note of two remaining blocks, depending where they are, as in, the side of each other you have a bigger target to hit so it's relatively easy to hit them, but if they're on the opposite sides of the screen you could treat it like a split in 10 pin bowling, and because you know that this is a problem when you have destroyed the previous brick that caused this situation you could have that last brick you destroyed supply a pickup that would make the level easier to complete, be it a shoot'em up option, a free pass, multiball or whatever new concept could be had, I guess you could use the same predictive method to decide on what pickups are given and where.


EDIT: I just sold one of these gadgets here (http://www.griffintechnology.com/products/powermate/) which I used sometimes to play arkanoid/breakout games, uber retro, the way arkanoid was supposed to be played ;)

[Nexic]

My solution to the last few bricks problem would be for them to slowly expand, if there are no adjacent bricks. This might not fit with the 'feel' of the game, but I can certainly see in working in most casual adaptations.

I think things such as ball size and paddle size play a large factor. A nice feature might be for the ball and paddle to slowly shrink as time goes on, thus making it harder.

Hmm, I think I've just invented a new game, Morphanoid!

[Pallav Nawani]

Arkanoid is an example. For instance... The last brick problem. It can be boring to try and hit the last brick on the screen if you can't get the ball's angle quite right. What is the best way to deal with it.

Last brick problem isnt a problem if the paddle aiming is a good. At least that's what I think, and so I did not bother to implement something to fix it. Whenever I played Dxball, I never liked it when lightning struck the last few bricks away, so my game Riotball does not do that :)


And what are the other elements of arkanoid that are important. Its easy to launch bonuses when you hit bricks. But how many?

Make a level editor, add the bricks, playtest the level and then decide. Most arkanoid games don't launch bonuses at random, but you put them in, play it, and do what feels right. You might make some special levels which require some particular bonus to complete it. If you do that, you then need to make sure there are more that one of that bonus around, so if the player misses one (cannot be helped sometimes no matter how good the player is) there is another.


Should the ball speed increase as the level plays?

I do that in Riotball, but then I never thought of doing it any other way at all :)

[Martoon]

I think the problem is with the assumption that the object of the level has to be to destroy all the bricks. Only one breakout game that I know of (Poing) has moved away from this assumption, which is slavishly reproduced from one game to the next.

There are all kinds of other ways you could run it. What if the object was to destroy as many bricks as possible within a certain time?
That's a good point. But I think there's a reason that almost all breakout games have you get every last brick. It's because there's something compulsively satisfying about getting every last brick, or eating the last dot in Pacman, or shooting the last alien in a Galaga level, etc. When I play an adventure or RPG with an automapper, I walk to the very corner of every room so it gets filled in on the map, even if there's no good reason to.

I like the idea of a breakout game that wouldn't require you to get all the bricks to complete a level, but that was structured such that you could get all the bricks if you were good enough. Your suggestion of a time limit would work nicely. If time runs out, you score what you've scored and move on. If you get all the bricks, you get bonus points for any time remaining, and move on.

[DGuy]

Last few bricks dilemma:

Make it a requirement that the player only have to break 80-90% of the bricks before being allowed to move on (via a key-press or RMB-click) to the next layout, but increase the points awarded for breaking each of the remaining 10-20% of bricks.

This will let players enjoy each board, but allow them to move on if things drag near the end. It will also allow more hard-core players to earn "bragging" rights and big points by hanging around and completing 100% of the board.

A good thing would be to keep track of the % of each board the player has completed and display this % with each board on the board-selection screen.

Maybe also during the last 10-20%: NO power ups/downs (and remove any that are currently in effect) and increase the ball speed. If the player wants to take out the last few brinks and win those extra points, they are gonna' have to EARN it!

In a snake-clone I'm working on (not for commercial release) the game is broken into rounds, and the player only has to play each round until the snake reaches a minimum length at which point they can chose to move on to the next round. BUT if they decide to hang around and play until the snake reaches its' maximum length for the round, points awarded for collecting items are doubled. As an added challenge, though, the snake speed increases. Also, each round has a unique bonus item that only appears as the snake approaches its' max length. The special items are worth big points and also get displayed along with the player names in the high-scores list, as a way of showing all the "metals" they've won.

[soniCron]

...there's something compulsively satisfying about getting every last brick, or eating the last dot in Pacman, or shooting the last alien in a Galaga level, etc. You may be right, but notice the differences between each example you cited:

Pacman (http://www.bucknergarcia.com/video.htm) - (Satisfying) Getting the last dot is easy. Just navigate to it without getting eaten by the ghosts.
Galaga - (Mildly Annoying) Shooting the last guy is a challenge because he flies at you kamikaze style, so you've got to shoot and run. It's far easier to die trying to get the last guy.
Arkanoid - (Highly Annoying!) Influencing a bouncing ball with a 64px wide "paddle" using a mouse (or worse, digital input!) to hit the final block, all the while you're stuck guessing and praying you hit the ball in the exact right spot that it ricochets off the paddle in the perfect angle.


Now, it's significantly more time consuming to get "every last one" in a game of Arkanoid than Pacman. Because of this time consuming nature to complete the level, the annoyance factor raises significantly, especially since we're dealing with a method of control that is actually a method of influence, rather than direct control.

Breakquest has conquered this tactic in an excellent way with the "gravity" button. The player has far more control over the ball than just chance, and this works for a more satisfying gameplay.

[svero]

I think there's definitely something to be said for the satisfaction of crossing every t and dotting every i (figuratively) -- I know in my game beetle bomp to make the end easy I threw in some rainbow colored balls that matched any color. People complained. They felt cheated at not being able to finish off the line themselves.

[luggage]

Sonicron: I think you've missed the point there slightly. It's not so much whether it's easy or hard it's purely that people like to do it. They feel a sense of achievement at completing it and may feel robbed of that if it's too obvious when they've had help.

An example are all the stars in Mario 64. Some of them are fiendishly difficult but people like the challenge. As Pokemon says "gotta catch 'em all".

-- edit --

When I say that people like doing it I don't mean every player. Just that Martoon was referencing a group of players who do like to get every brick.

Personally I hate having to get the last brick.

[soniCron]

Sonicron: I think you've missed the point there slightly. It's not so much whether it's easy or hard it's purely that people like to do it. They feel a sense of achievement at completing it and may feel robbed of that if it's too obvious when they've had help.

An example are all the stars in Mario 64. Some of them are fiendishly difficult but people like the challenge. As Pokemon says "gotta catch 'em all". If everyone liked waiting 3 years to get that last block, there'd never be talk about "last block syndrome." ;) That said, in a game like Arkanoid, there's only 5% skill involved when trying to get that final block. The rest is just luck. People love luck, when it's on their side. When it's not, you may as well be pouring boiling oil on them! :)

When you're trying to get the ball to hit the paddle on the exact pixel it needs to ricochet and clear the board, you're certainly not doing them any favors. If you can involve the player more than relying on that chance, the player becomes a part of completing the level and not just waiting to finish the level. That is not to say, of course, that it should be easy.

[Bmc]

i agree with sonicron on this one.

one method I've seen used to cure "last block syndrome" is to let the player move onto the next level once there are only so many blocks left... or keep playing until they are all gone.

[Martoon]

@Sonicron:
I totally agree that it's annoying trying to get the last brick, and I didn't in any way suggest that the "last brick" issue wasn't a problem. But I still maintain that, at least for many players, consistently not getting the last few bricks , and just skipping to the next level, would be unsatisfying.

The "last brick" issue is a problem, and we need creative solutions. But just saying, "You got 90% of the bricks - meh, close enough" and going to the next level gives the player a rather hollow sense of accomplishment.

There needs to be some skill-based solution to let the player get the brick. Like you, I really like the Graviton in BreakQuest. In fact, I think it's the single greatest advancement in breakout games since Arkanoid first hit the arcades.

[soniCron]

The "last brick" issue is a problem, and we need creative solutions. But just saying, "You got 90% of the bricks - meh, close enough" and going to the next level gives the player a rather hollow sense of accomplishment. Again, I don't know where you got that I said we should skip the level altogether. In fact, I suggested the very same thing you are suggesting right now. Please reread my first post on the subject (http://forums.indiegamer.com/showpost.php?p=52113) very carefully. The quote is below:

"Breakquest has conquered this tactic in an excellent way with the "gravity" button. The player has far more control over the ball than just chance, and this works for a more satisfying gameplay."

Please make sure you're actually arguing with me when you argue with me in the future! :)

[Abscissa]

I think the problem is with the assumption that the object of the level has to be to destroy all the bricks. Only one breakout game that I know of (Poing) has moved away from this assumption, which is slavishly reproduced from one game to the next. Pop-pop also moved away from that assumption.

[Bmc]

@Sonicron:

The "last brick" issue is a problem, and we need creative solutions. But just saying, "You got 90% of the bricks - meh, close enough" and going to the next level gives the player a rather hollow sense of accomplishment.



it's worked for games like magic ball... but then again that did awfull, real awful......... *crickets chirping*

[badjim]

you could let the player catch the ball and fire it straight up, say three times a level or however many times you collect a certain power-up.

That would let the player destroy the last few bricks easily but there would still be some satisfaction from completing the level.

[luggage]

Or once you get down the last few blocks a black hole effect starts forming and gradually sucking in the blocks and ball. Very very subtle at first and then growing in strength as time ticks buy.

You'll increase the likelyhood of hitting one and eventually be guaranteed to get rid of them.

[wazoo]

What luck that I just watched "Warrior's Gate" recently... :D

How about the constricting-universe approach?

Instead of doing something with TLB, why not start shrinking the actual gameworld itself?

The player would still pass the level, plus it would add an extreme sense of sheer panic...

[Larry Hastings]

I've been working on my Arkanoid clone for very nearly three years. My main solution for the last-brick problem was suggested by my pal Hanford Lemoore, and it's the "last couple bricks are optional" approach. I think this works incredibly well. You can adjust the percentage based on the difficulty level, so "Hard" level is required to get all the bricks, but "Easy" can skip past if they're mostly done and get bored. However, my game is structured such that it's up to the player when they go to the next level, so if they want to get every last brick, they certainly may.

In the original Arkanoid, there was indeed a powerup that let you skip the level. It opened this sort of static-y hole in the wall on the right side, and if you moved the paddle over there it would slip through and onto the next level.

For the record, there are plenty more assumptions you didn't mention that nearly ever Arkanoid games makes, several of which which I've done away with. But rather than post my big long list here, I'm just gonna make you wait for the demo ;)

[dima]

If I were to make Arkanoid game, I would concentrate on the skill aiming, like in Ricochet, where the paddle is curved and you can get a feel for controlling the ball. Also, when game gets to the last brick or 3, I would kick in a timer, say like 5 seconds, and if within that time you dont break the ball, level ends anyways, but if you do get the last ball, you get some juicy bonus, like score. This approach still relies on skilled and pressured gameplay at the end if you want to feel the accomplishment, but at the same time it end the level if you are having a hard time, leaving you with the sense of losing the last 5 second bonus and wanting to do better next time. In the end, the important feature in my game would be the proper aiming implementation.

[DGuy]

Also, when game gets to the last brick or 3, I would kick in a timer, say like 5 seconds, and if within that time you dont break the ball, level ends anyways, but if you do get the last ball, you get some juicy bonus, like score. What if the ball is on it's way to hitting the last brick and times runs out? I would feel cheated!

If the whole game where time based, like someone suggested, break as many bricks in the time allowed, then, yea, time runs out, time runs out: try better next time.

But to have a game which is NOT time based, become time based for the last few moments of each round, potentially ripping from the players grasp the goal they've been working towards (100% of brinks gone), that's a recipe for frustration.

[tentons]

I think the problem is with the assumption that the object of the level has to be to destroy all the bricks. Only one breakout game that I know of (Poing) has moved away from this assumption...
Make that 2 now. :) In Little Gods, the main goal is to get to a certain number of points (usually 100) to end the level. If you break all the bricks before you get to the score requirement, whichever player has more points wins the round. But there's no pressure to destroy all the bricks.

[Bmc]

one thing that I noticed in some of the newer arkanoid type games coming out in the casual space, is that they help you (very subtly) by breaking bricks for you during the level. For instance Bricks of Atlantis, has little fish that are bust up bricks when they swim into them etc.

I think the ultimate solution to the one brick left problem would be to predict where oprhaned bricks may occur and having "helpers" take em out

[cheese_phantom]

as my girlfriend put it: "Now the game thinks I´m a looser again, and drops those guns!"

Aren't there breakout games that allow you to earn these guns gradually, say one gun when you clear half of the bricks, so when you come to the last few bricks, you are not a looser and the game drops you guns, but you are a winner who earned the guns throughout the game and now uses his rewards?

I think the last brick problem correspondends to a problem in screenwriting. When the conflict is basically solved, screenplays use to have a short wrap-up scene that is recommended not to be kept too long since the dramatic curve has reached the top and now, after the climax has been reached, goes downwards. This means, the more you continue, the more you allow the curve to go down, which means to extend the story needlessly and this kills the joy of the climax. You want to leave with the good feeling, not being held there any longer. The last brick is something like that: You had your tough fight and the last brick means you've in the wrap-up scene. If it takes too long to get rid of it, it becomes a needless extension. And if you send some "external help" like suddenly a gun dropping from the sky, it's like something outside the game world comes to aid you. As someone before has put it, you don't feel that you've earned your victory.

If you make the player earn the guns during the game, you allow the player to solve the problem with his own skills, not with sort of a "deus ex machine", which is an external intervention and means that the game seems to break his own rules and makes you feel that it does not recognize that you earned your victory.

[dntoll]

but you are a winner who earned the guns throughout the game and now uses his rewards?


Maybe we should just have a text saying:
"You have earned the gun!!!" or "Bonus Gun" or "weapon drop" and flash the score a bit so it seems like you reached a special score level to get that bonus? While in fact the logic dropping the gun is based on how few bricks you have...

I think the concept of dropping weapons works pretty well... nothing the testers have complained about...

Btw, "Super Mario Cart" is really good at giving bonuses depending on your position.

[cheese_phantom]

Maybe we should just have a text saying:
"You have earned the gun!!!" or "Bonus Gun" or "weapon drop" and flash the score a bit so it seems like you reached a special score level to get that bonus? While in fact the logic dropping the gun is based on how few bricks you have...

I think the concept of dropping weapons works pretty well... nothing the testers have complained about...

Could be like if you hit three bricks of the same color in a row, or succeed to clean a column of bricks completely you get the bonus gun... which is announced then, as you suggest. This means that you make aimed shooting a more importing thing in earlier parts of the game and not something that gets of importance at the end. The effort that you show for aimed-shooting during the game rewards you with cleaning the last few bricks with the guns you've earned during the game.

I imagine something where for instance the last brick in a column starts to glow or to blink which catches attention and invites the player to aim to hit that brick. Might be also random bricks blinking for a certain time. If you hit them while they blink you get the gun.

Most of this must have been tried though.