Friday, June 15, 2012

Badge spotlight: GemCraft Labyrinth

Remind Me of the Babe (hard, 30 points) -- Complete the final stage of the labyrinth.

Today I'd like to talk about the length of badges. This is actually a subject I suspect I'll be revisiting a couple of times over the course of this series, but GemCraft seems like a good place to start the discussion.

I'm generally a completionist when it comes to games, and I like getting achievements (which is obviously part of what draws me to Kongregate and has kept me there). On the other hand, there are limits to my completionism; when, for instance, I looked at the achievements for Team Fortress 2 and saw the one for healing a total of one million HP as a medic, I decided I was perfectly content not getting that achievement, since I enjoyed the game, but not enough to play it several months continuously (all as a medic, even). And in general, for games on Steam or consoles, designers don't really have any incentive not to add ridiculously long achievements -- once you've sold the player the game, you might as well try go keep them playing it as long as possible. Kongregate, on the other hand, doesn't have these ridiculously long kinds of badges; it's #2 on Greg's principles of badge design, and I think it's an excellent idea. Part of this is because most Flash games don't support that kind of replayability, of course, but that's not the only reason. There's also the fact that, unlike in the Steam or console ecosystems, where achievements are created by the game developer, Kongregate creates the badges itself, and the strength of Kongregate is in its breadth, not in the depth of any single game. So it's better off encouraging people to play a wide variety of games. In addition, avoiding the super-long badges also makes it seem like all of the badges are attainable, which is a very appealing feature to a gamer like me. Overall, I think this strategy of Kongregate's is one of the best features of its badge system.

However, the flip side of the coin is easy to see. For any game with a well-defined ending (i.e., not counting games where the goal is just to get the highest score or build the biggest empire or whatever), awarding a badge for reaching the ending is pretty much a no-brainer -- the only thing really left to do is decide what difficulty it deserves. That's all well and good if we're talking about relatively short games, which the vast majority of Flash games still are, but what if a game takes a really, really long time to reach the ending? And what if a lot of that time is relatively repetitive content -- or does that matter? Does it still deserve a badge for finishing the game even in that case?

And that brings us to GemCraft. GemCraft is a wildly popular series; the first two installments both spent considerable time in the top 5 of the Kongregate rankings, and while Labyrinth comes in a little lower (currently #44 as of the time of this writing), it's still rated 4.37, which is an incredibly high rating by Kongregate standards. Plenty of people are huge fans of GemCraft, and while I've never quite been one myself (I've always preferred the free-build tower defense games to fixed-path ones), I certainly recognize that GemCraft is pretty much the pinnacle of fixed-path tower defense games. It's very solidly designed and has an extremely polished presentation; Labyrinth even borrows a few elements from juggling-based TDs, as well as eliminating the most annoying feature of earlier GemCraft installments, and it has tons and tons of depth, with a bevy of gameplay modes, additional challenges, and secret achievements all available to the GemCraft enthusiast. The game also does a good job of making a lot of different strategies available to the player, and encourages the player to develop a wide range of strategic options for dealing with different sitautions, which is another hallmark of good design.

But here's the thing: GemCraft Labyrinth is long. The aforementioned badge is for simply completing the game, and after completing the game there's plenty of additional content if you're crazy about GemCraft, but even completing the game takes a long time. There's a total of 165 normal levels, and they're not easy for an average player -- while you may get through many levels on your first try, some will take several, and this being a TD, an average level is going to take several minutes to complete, so you're looking at a pretty substantial time investment. Now, in fairness, you don't have to beat every level to reach the center -- if you follow the optimal path, in fact, you only need to beat 76 levels (but you'd have to know the optimal path by looking it up in a FAQ, since the game doesn't tell you). However, that's not necessarily going to save you time, since beating levels gains you XP, which makes your character more powerful, so if you take the optimal path, you'll be incredibly weak (relative to where you "should" be) by the time you reach the center of the labyrinth, so only a very skilled player (which of course takes a lot of time on its own) or one who has spent a lot of time grinding to gain more levels (which, of course, also takes a long time) could survive that way; I ended up playing through every level and still had a lot of difficulty in the later levels, so there's really no way to reach the end without some significant time investment. (Labyrinth also, to its credit, discourages mindless grinding; you can't just replay the same level over and over and keep gaining XP. Instead, you only gain experience by besting your previous performance, which requires either beating the level with harder difficulty modifiers, or simply playing better.) And of course 76 levels is still not exactly an afternoon's playtime.

So, is completing the game simply too long for a hard badge? This is a difficult question to answer, but my inclination is as follows. I think that the default assumption should be that any game worthy of badges should get a badge for completion, with exceptions only in the cases where the game is not very good (just above the quality threshold for badges) or completing the game is exceptionally tedious. While GemCraft Labyrinth is certainly very long, it is an excellent game, and while there is naturally some repetition involved in completing the game, the designers have done a good job of throwing a wide variety of challenges at you, so it's not horribly tedious. So in this case I think that having a badge for completion is justified. (And similarly, while it would certainly be possible, given the large number of additional achievements in the game past simply completing it, to assign a badge for further activities, I agree that not having such a badge is also for the best.)

That said, though, it's pretty much never that Kongregate doesn't add a badge for game completion to a game for which that's possible, and I do think that there are a few more cases where it would be a good idea -- it doesn't have to be many cases, but I think that more than nearly zero is quite reasonable. I don't want to go into a long list of badge-bashing, but just to provide an example, LethalRPGDestiny 2 is a game that I would have found tedious at 1/10th the length, and yet it still got a badge for game completion -- a much shorter badge I think would have not hurt anyone's experience, and of course people who still wanted to complete the game could. So perhaps a little more aggressiveness on the part of Kongregate in this field would be a welcome idea -- it would cut out some of the most unpleasant badges on Kong without really hurting anything.

Thursday, June 14, 2012

Badge spotlight: Amberial: Nebulosa Realms

Spherical Ninja (hard, 30 points) -- Complete every level with an Ace and a Thunder finish.

I'm directing today's spotlight onto Amberial: Nebulosa Realms (reviewed previously here), but really I want to talk about an issue that's pretty pervasive in designing badges for Flash games (a problem that I'll bet people designing achievements for the PS3 or Xbox don't have to worry about), and that is that the quality of the computers that people use to play Flash games varies widely. Some people are playing games on decade-old machines or woefully underpowered netbooks, while others are playing on rigs capable of running Crysis 2 without any problems. As a result, a game that will play fine for some people may be unbearably slow for others.

For the majority of games, playing the game on a slower computer merely means that things happen more slowly, which (especially for fast-paced shooter-type games) can often give the player an advantage. Indeed, I will admit to deliberately playing some games on a slower computer to give me an edge on some of the more difficult badges on Kongregate. While you might think that this only would affect really graphics-intensive fast-paced games, it affects a pretty wide swath of Flash games, probably due to the fact that writing poorly performing Flash code is pretty easy. It even affects some games you would think wouldn't be dependent on processor speed at all (99 Bricks is my standard example -- although the game appears to proceed at the same pace on slower machines, your tower sways noticeably slower, which confers a significant advantage). And often even a small, nearly imperceptible difference in speed can make the difference between an impossible badge being attainable (at least for me) and truly unattainable.

If that's the only issue to worry about, the solution is pretty obvious -- play a game on a sufficiently fast computer that there's no slowdown (I would assume that Kongregate has such machines) and assign the badges based on that, and just accept the fact that people with slower computers have an advantage. This obviously isn't a perfect solution, but it's clearly better than the alternative of making badges targeted at slower computers which may be impossibly (or at least unreasonably) difficult on faster machines.

Of course, things are often not so simple. For instance, while most games have a well-defined top speed, some will just get faster and faster with computer speed; there are some games from 2008 on Kongregate which are nearly unplayable now due to the speed that they go on an average computer from today. Other games actually get more difficult on slower computers. In Vector Runner, for instance, while the pace of the action doesn't appear to get slower (at least it didn't when I tried on a slower computer), the controls get mushier, so having a faster computer is an advantage. There are even some extreme cases like Typing Ninja Hunter where the game is actually unplayable on a slower computer (the keys are so unresponsive in the boss battles that you simply can't get them to register in the allotted time).

The reason I've chosen Amberial: Nebulosa Realms as the spotlight for this discussion is that it has one of the most frustrating manifestations of this problem of all that I've seen. One of the requirements for its badge is to get a Thunder finish on each level, which requires beating a certain target time. On one level, I had a tremendously difficult time beating the target time, and it seemed to be due to one place where I had to wait a couple of seconds for a moving platform to reach me. It wasn't like there was any obvious way for me to catch it earlier, since merely saving a few tenths of a second wouldn't help; I would have to improve by a significant amount to catch it earlier, and that just didn't seem possible. So I looked on YouTube to see if there was some strategy I was missing. Nope -- they did the first part of the level pretty much exactly the same as I was doing it, and arrived at the moving platform at pretty much the same time, except instead of the moving platform having already left, it was just arriving! Apparently my computer was faster than the one in the YouTube video, causing the moving platform to be just faster enough to make the target time ungettable. So I finally resorted to somewhat questionable means -- I played a video in the background and suddenly, I could catch the platform and get the target time relatively easily. This is the only time I've had to resort to such shenanigans, and it's rather unfortunate that it was necessary.

As these examples illustrate, this is a pretty thorny field, and I don't think there's necessarily a clear-cut answer. I mean, it's clearly impractical to expect Kongregate to support 100% of its users' computers, given the wide diversity, so I think it's OK for Kongregate to sometimes say, "if you really want this badge, you might need a better computer". But where do you draw the line? Personally, I think it's OK to set the line at maybe a slightly below-average machine and accept that machines slower than that may have problems, but this may simply reflect my bias that the machines I've used for Kongregate (except for when I've deliberately played on a slow computer) are generally pretty middle-of-the-road. And this doesn't really solve problems for games like Amberial where the natural progress of computer quality may make the badge ungettable -- I don't think there necessarily is any solution to this problem other than to hope that game programmers take these issues into account when doing their timing routines!

One last note: This fact also tends to skew feedback on badges, and I think it's actually a pretty large factor. When people say that they found a (speed-based) badge very easy, my first thought is usually that maybe they just have a slow computer (or even a fast computer which they were doing a lot of other stuff on). I wish more people would take into account this factor when evaluating Kongregate's badge difficulty, since at least in my experience, it tends to be pretty solid.

Wednesday, June 13, 2012

Badge Spotlight: colourPod 2: dimensionPod

Fear Not the Darkness (impossible, 60 points) -- Complete all 24 missions.

I'm going to kick off this series with what I think is one of the most underrated badges on Kongregate, but first a little talk about impossible badges. As the name suggests, these are awarded for some of the most difficult achievements on Kongregate (although, fortunately, not literally impossible), and they tend to draw a disproportionate amount of attention. While there are (as of this writing) only 44 impossible badges, a new impossible badge, or the possibility of one, is always guaranteed to get a lot of discussion. As a result, there's a lot of pressure to get the impossible badges as right as possible, and this, in addition to the fact that most games simply don't have a task which is really suitable for an impossible badge, has resulted in a gradual decline of the number of impossible badges on Kongregate. This is certainly fine by me -- given that the impossibles, not surprisingly, often require an outsize amount of time to complete, I would certainly prefer having them awarded only to games which are really a great fit for the impossible badge idea rather than trying to make a greater quantity of not-so-great impossible badges.

Given how people like to complain on the Internet, for pretty much every impossible badge on Kongregate, you can probably find someone complaining about it on the forums: it's too easy, it's too hard, it's too luck-dependent, it requires too much pixel-perfect positioning, it takes too much grinding, etc., etc. Some of these complains have some validity, but I'd say that probably most don't. The impossible for colourPod 2, though, I don't remember anyone complaining about, which is in many ways the highest compliment a badge can get! (Of course, part of this is due to the fact that it's one of the older badges on Kongregate, I'm sure, but even still, the impossible for Pandemic 2, which came out at about the same time, is certainly not safe from criticism.)

So, why do I think this makes a good badge? Well, first, a brief description of the game: at first glance, it's a pretty ordinary shooter. You control an immobile turret in the center of the screen, and enemies ("fragments") approach from the edges, and you have to destroy them from before they hit your turret. The fragments can appear in either of two dimensions, so you have to switch between the two dimensions (by pressing the space bar) to be able to hit everything. Colored fragments will also appear, and if you catch them on your "lens" (mounted on your turret opposite the gun), you will gradually gain additional powers, with different colors conferring different powers. As you have some control over the colors that arrive, choosing your colors wisely is a very important skill.

Anyway, this badge has what I find by far the most satisfying quality in a badge: at first (in this case, because it seemed that the badge was so dependent on reflexes, which, as I've mentioned, are not my strength), it seemed truly impossible. Then, I gradually got better, and it seemed attainable, and by the time I reached the end, it almost seemed easy; although, of course, that was an illusion -- it was still hard! It was just that I had practiced so much that made it seem easy. And in fact, there are plenty of places where you can improve your skill -- for instance, since the game only very gradually introduces you to the different color abilities, there's a lot to learn in which ones are best in which situations, and which sets work well together. The other area where I really found myself improving over time was simple -- the game demands that you be doing a lot of things at once, and sometimes it will exceed your ability to do everything (probably true no matter how good your reflexes are, but certainly true for me). But as I went along, I discovered the key was triage -- not necessarily trying to do everything at once, but learning which things were the most critical and which wouldn't hurt me too badly, and this skill was definitely something I could see myself improving at as I played through the game.

As you can see at the top, the badge requires you to complete 24 "missions" (the game's name for various achievements), which is also a quality I generally like in a badge -- I tend to prefer badges which require a variety of tasks rather than one extremely difficult thing. Many of the missions simply boil down to playing the game well, but there are also many which require you to use or not use a particular color or colors, which adds an interesting layer as well -- they get you to experiment with abilities you may not have used normally, so you really get a chance to see the whole game.

In conclusion, then, this is why I think the badge is a solid one: The game itself was good (obviously a necessary prerequisite), the badge requirements were challenging but attainable (and you could work to get better at it), it really felt like it rewarded your ability, and it provided a good variety of different tasks rather than being completely repetitive. In some ways, in fact, this is the badge I'm proudest of; it's not the most difficult or rarest, but being in a category of game that I normally do poorly on, the fact that I was able to learn, gain experience, and improve my play to the level where I could get the badge really gave me a positive opinion of the game and the badge.

Tuesday, June 12, 2012

Announcing something new!

So I've been playing games on Kongregate for a while now, and in that time have accumulated a fair number of achievements, so naturally I've thought for a fair amount about the topic of achievement design. The achievements were what really got me into Kongregate in the first place, and I think the general quality of their badge design has been a large factor in keeping me there. In 2009, Greg McClanahan, the man who, at the time, was solely responsible for the badge design on Kongregate (now, he at least has someone to assist him with this task) and arguably the most visible person on Kongregate as a result (you will still find people who are under the impression that Greg founded and/or owns Kongregate), wrote an article for Gamasutra on the subject of badge design, and I think that the principles he lays out are really excellent ones -- I think it's contributed a lot to the success of the Kongregate badge system, and conversely, I think in the cases where he has produced bad badges, it's a result of (hopefully inadvertently) violating these principles. But other than that article, it doesn't seem to have been a system that's been thought about too much (as one can see from the wildly differing approaches to achievements on other platforms, for instance).

So I'd like to share a few of my thoughts with whomever is around to read them. I think the best way to do this is with a "badge spotlight": each post, I'll take a look at a badge which is particularly interesting or controversial, and talk about the issues around that badge's design, hopefully illuminating some more generally applicable principle. I've already written up a few, to ensure that this project gets somewhere, and will probably run to about 20-30 posts before I run out of ideas (or steam).

Naturally, the bad badges tend to get the most attention, and of course it will be instructive to look at cases where I think the badge process has erred, but I don't want to get completely into a negativity trap, especially since I think that would obscure the fact that the vast majority of badges on Kongregate are indeed well-chosen. So I'll try to also shine the spotlight on badges that I enjoyed and were particularly well-designed os that we can get some examples of what makes a badge good, as well.

Finally, a few words about my own preferences. I don't expect everyone to agree with these, of course. My own gaming strengths lie more in persistence and cleverness than speed, so I tend to prefer games which reward thinking and learning over pure dexterity. I really don't like achievements which require you to do something perfectly over a long period of time, since I find there to be nothing more frustrating than being great for five minutes and then screwing up at the end and losing it all. I'd much rather have a task with a lower cost of failure (even if it is overall harder), especially since that ties into my first point -- if dying only costs you a few seconds, but you have to try a large number of times, it gives you a lot more opportunity to learn and get better than if you have a smaller number of longer attempts.

Anyway, I hope this series will prove interesting to whoever reads it, and I also look forward to any feedback or other comments you may have!