Elite Beat Agents - HEAAAAAAAAAAAAAALP!

[Dracos]

Elite Beat Agents, for those who don't know, is a rhythm game iNiS developed and Nintendo published throughout the states, and I think Europe.  It was the american version of Osu! Tatakae! Ouendan, an extremely popular DS rhythm game known for it's cheerleaders that helped people with problems.  It should be stated ahead of time that I came to this game as a novice in the genre.  Rhythm games aren't an area I've spent much time in at all, so while this review will reflect game analysis, it'll reflect it from the role of someone unfamiliar with the conventions.  Elite Beat Agents bears the distinction in this of breaking the wall for Rhythm games to me and, at the same time, convincing me to put it back up afterwards and not bother, to provide the short good and bad of it.

Elite Beat Agents is not simply a port of Ouendan.  It's a completely new game built with the same principles and design aesthetics that made Ouendan a hit in japan and an import cult hit here.  iNiS has very smart folks over there though and realized that if they made a game uniquely suited to the culture, it'd sell better.  Good call there, iNiS.  Elite Beat Agents sports one of the best game motifs I've ever seen.  The whole government beat agents deal is a fantastic and cool setup for the whole thing, mixing men in black with stylistish dance moves.  Beyond that, the theming tends to be along a bunch of semi-classic american tales, from babysitters wanting to score with football quarterbacks to folks digging for oil.  Knowing that anyone who sees it is going to think Men in Black, they even include some aliens at points who are defeated by the power of rhythm.  They keep their motif solid throughout, the game world never questioning the narrative rules they've set forth and the game always working as one would expect there.  In this respect, they've really done a fantastic job.

The animation for the game is awesome as well.  While there's a touch too much hand waving that goes on, I understand the choice as a visual cue.  The dancing in the background is both appealing and kept going along with the rhythm beats you have to hit, which was a good move.  The constant playing story animation gives a reason to watch your own replays while at the same time, the stopping gives both rest breaks and neat story segments as rewards for making it that far.  The art is colorful and bright throughout with a lot of style and emphasis on face shots, bringing the player in and emphasizing how your dance is affecting these folks.  Lots of nice touches here and an animated story telling style that I'm sure is going to be copied by others as it's a very good one to map on top of a perpetual 'puzzle' type challenge.  Way to go iNiS.

The music is pretty good.  I might sound a bit snooty though as I'm going to say it should have been better.  It felt like, at times, there was an emphasis on known American tunes that sometimes took precedence over 'is this tune good for the game'.  I'm being picky as most of them were and mapped well to the rhythm beats the player is supposed to hit during them.  I can't really separate the music selection from the rhythms they chose for them as they were so integral to the game experience and frankly, some felt like they were just genuinely poor choices, 'Canned Heat' being one that definitely caught my eye.  Fans of both Ouendan and Elite Beat Agents have commented that generally the song list wasn't as interesting on this one and while I can't entirely back that, I'll put it out there for folks to consider.  Most the songs are excellent, but due to certain game design choices, you don't get to pick and choose which you are playing until having beaten them.  Reiteration: Good but could be better!

The interface is both clever and... really stupid.  Frankly. it's remarkably fitting usage of the DS interface with no singing.  Yaay.  Way to choose there.  The game is entirely navigatable using the stylus (Yay), and generally the rhythm game naturally works well with the kind of tapping it requests.  Why would it be stupid?  Well, in the latter levels, they start throwing out 32nd notes.  I stopped when I noticed tiny pockmarks forming in my screen from 'hitting' these.  Generally, I think the design assumed the invincibility of the screen and I'm not sure that was entirely a fair assumption to make.  It could just be mine though.  It had right and left handed modes... but it didn't REALLY.  As a left handed player, this really hindered my own experience  on it as I'd find consistently that things I needed to hit would be spawning below my hand.  Easily I could see if I had been doing the stylus with the other hand, it wouldn't have had this issue and given many patterns do mirroring, they could've easily fixed that.

Let's get to the game design and why it's kind of thrown me out of the genre.  Elite Beat Agents, while having many fun levels, is not a well built game.  Compared to others in the genre, it locks all of it's songs until prior levels are beaten in each difficulty mode, which restricts the flow to a mostly linear experience in each one.  At any time, you have between 1 and 3 'new' levels you can complete, of which all of them must be completed to unlock the next chunk.  Each level consists of 2 to 4 'stages' with an animation at the beginning and end to give rest time.  During the stage, you have an elite meter that  constantly decreases which you must keep high by successfully getting 'good' or 'great' on hitting the various rhythms.  The rhythm beats often come out in patterns which getting them all at 'great' gives a bonus, as does getting them all.  Missing the beginning of a pattern generally will end the level (regardless of your current state) since it tends to be very hard to recover before the pattern is done and a mere 4-5 consecutive bads or misses will empty the bar.  If the bar empties, it's game over.  The longer your consecutive combo level, the more the bar seems to refill and the more points you get per next beat.  Perfect level completions involve missing none of the beats and often give scores that are four to eight times as high as a single miss would give.  If the bar is below the 'yes' margin at the end of any stage portion, you get a losing animation, but the game continues, the ending animation for the level depending on how many you won.  Problems I had with this design?  While on surface, it looked good, I found it to be nearly impossible to lose any of the stages and actually beat the level.  It left me going "What's the point?" to that whole mechanic after a bit.  If you lost a section, odds are the level was over.  It was near impossible to see anything but perfect win or almost perfect win states if you won at all.

The elite meter had other problems.  It always decreased, which in the later levels really could leave you in no win situations if you'd done poorly at previous points.  Areas with few or no beats could result in as much as a 25 percent drop in the elite meter even by nailing every beat that came out if you weren't already continuing a huge combo.  It was this mechanic that caused the end of level elite meter graph to be so wiggly since rather than simply dropping as you did poorly or increasing as you did well, there would be elements in many levels where it simply didn't matter, the bar was going to drop some and you couldn't do anything to stop it, only bring it back up a few seconds later.  This felt unnecessary, especially in light of the fact that a missed pattern almost instantly would drain the meter anyway.  I suspect  this contributed to another issue I had.

Just as it was hard to beat a level without winning every stage of it, it was also hard to beat a level while doing poorly at it.  Whatever their scoring mechanic was, there often would be very little difference in rankings between levels I barely beat and ones I completely trounced.  Score would often differ tremendously, but it made the rankings feel silly when a  'bare pass' would often get a B or A, leaving little for a perfect play to go up to.  I got a single 'D' in the entire time playing.  Either the pass guidelines were too high or the ranking guidelines were poor.  Equally, rankings were often unrelated to score, which cared most about how long your combo was.  It was always weird to get a lower B-C ranking tied to scoring two to three times as many points.  Overall, I think they could've provided either a more unified front or a more cleanly divided one.

The way the game depended on patterns was fine...but the lack of any good practice mechanism generally meant that if you failed on pattern on stage 3 or 4 of a level, you'd have to repeat the level up to that point.  That could be a real energy killer at points.  There's only so many times I'm willing to do the first 2-4 minutes of song to try and get just a bit better at the end.  They knew what they needed with a review option, but the review option just lets you see what you did, in it's wrongness, rather than provides a valid tool to help practice getting better at it.  I hope they fix that if they make another game in this series.

That's about all I have to say about level mechanics... but there's also the overall gameflow mechanics...which weren't very good.  It was  entirely possible to halt on a level for hours... and then crush the next 3-4 without a wink.  Which is bad, but not as bad as the difficulties.  The game starts with two of the four difficulty modes  unlocked.  Playing easy mode teaches a player to lose at easy mode.  Playing normal mode teaches a player how to beat easy mode.  I assume playing hard mode teaches you how to beat normal, but I never unlocked it.  What do I mean by teaches you to beat or lose?  Well, easy mode teaches you to play by visuals.  There's lots of visual cues for most of the levels, and getting at all used to using them ensures that you won't be able to beat the last level, which uses none of them.  I thought that was kind of shitty and pointless  in a 'rhythm' game.  Normal mode requires learning to do it by ear from the get go (Good), keeping most the visual cues  off screen, but also introducing...the patterns you might want to know to beat the last level of normal mode!  Amazingly, it might be nice to introduce the patterns in an easier state before demanding mastery of them, and that's just what normal mode does... for easy mode's final level.  When folks talked about Ouendan, they mentioned commonly that they had to move up to the next difficulty level before being able to beat the final level of the previous one.  I can't help but notice this is an intentional flaw in the design and it doesn't impress me one bit.  I don't care that it's supposed to be a finale level, there's no reason not to include some of those patterns leading up other than to provide a frustrating wall.  It's a design scheme that makes it more likely that the player fights to become good rather than naturally finds themselves suddenly able to perform extremely complex patterns without realizing that they've been learning the basics of them all along.  I'm told this sort of spike design is endemic of the genre.  I'm not sure I believe that as other rhythm games I've  looked at since (DDR, Donkey Konga) visibly have no such idiocy about them.  Either way, the thought that it was endemic of the genre pretty much convinced me for a while (until someone handed me more for free; Thanks IGDA) to swear off the genre again.  The designers can ooh and aah about the challenges they've delivered, but in what all too often is most exciting from the performance artwork standpoint to the general audience, they've missed the point.

I come across very harsh on this point...  which is a bit unkind as when they get it right, the game is very fun.  Learning the new pattern is also generally pretty fun.  Getting perfects on known songs is exceedingly fun.

Edit: I forgot the multiplayer mode, which I did try out...and was pretty cool.  My only complaint was that the multiplayer only rules weren't entirely clear.  I had to think about what certain things actually meant and had other designers actually wondering out loud as it was very non-transparent why certain things would happen and whether they were 'bad' or good.  aside from that though, the inclusion of other little competitive story animations (Go iNiS) and versus gameplay on it felt pretty satisifying.  The switch from 'do good or your score drops' to 'do good to raise your score' was also reasonably motivating in a competitive sense and made it feel like you were racing the other player.  Good moves.  I didn't get to try out co-op, but I suspect that'd be fun too.

But overall, the game becomes a mountain without a vista at the top.  It likely would've been better served to unlock all modes after the first time the game was beaten and thus let the natural fun of the experience keep the player going who has already seen all the story.  The story might be the carrot urging the player on early, but after a run through, it's really the gameplay itself that asks the player to continue, and EBA ends up putting a wall between more gameplay and the player.  Unfortunate, given the fun level the game manages much of the time.

Dracos