Eisydian Saga – A Rookie Developer's Post-Mortem Analysis

I'm writing up this rather in-depth analysis on what was essentially my first serious crack at making a flash game in hopes that it might help budding developers get a feel for what's important and what to expect when making their first few games; of course, if more experienced developers are also able to glean something from reading this, then rock on. :)



Overview

Eisydian Saga is a top-down shooter/shmup/bullet hell game that I released on March of 2013. As far as such games are concerned, it's pretty much a basic biscuit, but IMO that's acceptable for a first game. After all, shooters are a fairly easy first project because mechanics-wise, you really only need to worry about detecting collisions between players, enemies, bullets, and upgrades.

If you're unfamiliar with ES, you can check it out here if you're interested:

http://www.kongregate.com/games/Aesica/eisydian-saga

Just don't expect an award winner.



Story

The story revolves around a girl named Alma who was once a devout follower of 'The Church of Asua.” After the first level, she learns of the church's true desires and turns on it as she tries to prevent Erathiel, the Asuan spiritual leader, from ultimately ending the world.

The good:

  1. Has the potential to be a decent story (bearing in mind that different people appreciate different types of fiction) but that's really it. Literally any story idea can either be amazing or terrible depending on how that story is told.

The bad: There was actually a lot of bad in terms of this game's storyline:

  1. I made the fatal mistake of assuming that people playing a shmup didn't really want to deal with very much in terms of story material, but based on feedback I've received from players, I couldn't have been more wrong. What I actually presented in terms of a story in ES was nothing short of awful. If you're going to include a story in your game, either give your audience the full deal or don't even bother.

  2. People quickly drew parallels between “The Church of Asua” and the various faiths of Abraham (Christianity, etc) and rightfully so, because that's where the inspiration came from. The critical flaw here is that the people who follow these religions are probably going to be offended, stop playing immediately, and rate the game as low as possible. I'm not saying every story you include in your games should be watered down for the sake of being politically correct or anything. Just be careful around the more sensitive subjects, because you don't want to alienate members of your potential playerbase if you can help it.


Graphics & Sound

Was there anything good about the graphics in ES? Not especially. One thing a lot of programmers are not good at is creating artistic assets aka your game's graphics, music, and sound. Tragically, these things are arguably the most important parts of a game. There are very few games that can truly get away with skimping on graphics and sound, and even that is the result of having enough artistic sense to arrange minimalistic assets in a way that is still aesthetically pleasing.

Lesson 1: Keep Artistic Choices Within Reason
Having said that, ES wasn't intended to be minimalistic in terms of its artwork, however since my artistic ability only goes so far, it really showed. Painfully so. If art isn't your strong point, then try to avoid building your game in a way that demands high quality assets. For example, I could have just as easily styled ES more along the lines of the Frantic series and still produced a top-down shooter. All 3 games in the Frantic series don't have stunning graphics, but that's forgiven because they don't try to. Instead, the presentation is simplistic and clean, making for enjoyable gameplay, even if gameplay is all the series offers.

Having said that, if your artistic skills blow but you want to create something that requires artistic talents beyond your own, either suck it up and pay an artist, or at the very least, fish around online for free assets.

Lesson 2: Boobs
I made the horrible mistake of assuming that most flash gamers were going to be boob-obsessed teenage/young adult males, so the outfits my characters ended up wearing were perhaps a little too risque. All those seedy Chinese fly-by-night MMOs seem to be enjoying success, right? So why not create sexy characters? Several reasons, actually:

  1. First, and most importantly, most of your female audience isn't going to be happy if every single female character they meet in your game looks like a Halloween party stripper. It's easy for male game designers, as well as female game designers who do like such characters, to not see this as a big deal, but believe me, unless you just don't care about the majority of your potential female audience, tone down the sexiness. (Or include some sexy males? :D No seriously, just go easy on the sex appeal. You're making a game, not porn)

  1. If you dress your character like a sex object, that's all most people are going to see her as. You can have the coolest personality imaginable for your busty, scantily-clad heroine, but most people will never be able to see past her sex appeal. The perfect example came to me when I tried searching for ES fanart. I only found a few by only a single person, and surprise, they all had NSFW written all over them. Oops!

  2. It's insulting to your male audience too when you think about it, and to your game as a product. Sticking boobs on the cover of something is pretty much akin to saying, “this is the best my game can offer, please play it. I'm dangling boobs in front of you because I know you'll click on it!”

Lesson 3: Music
Finally, a point of success for ES! I'm pretty picky about music, so I spent awhile cruising newgrounds before I found some tracks I liked. Nearly all of the players I heard from seemed to enjoy the tracks, so the extra time spent and pickiness involved in screening was worth the effort. I realize it's often something developers gloss over because most people just hit the mute button, but think of it like this: What if your game's soundtrack is actually good enough to deter them from muting it? When I think back to my favorite games, the music is what I remember the most. (I'm looking at you, Saga Frontier and Final Fantasy!) About the only thing I could've done better regarding the music was a little level variety. As much as I like the music that plays on every level, it probably shouldn't have been the only tune that played for every single level. At least the final boss had her own (incredible) theme, however I doubt very many got far enough to hear it.

Lesson 4: Sound
Invest in a good (royalty-free) sound effects package, either through an effort to find one for free, or by purchasing one. While ES had great music, the sounds (all of which I made using SFXR, a very basic sound effects tool) were...mediocre. As such, I can't help but wonder how many people went for the mute button after hearing the repeated enemy explosions or coin pickup plinks, despite the music quality.

Lesson 5: Art Style and Consistency
Art styles: Choose a style: pixel, hand-drawn, 3d, realism, cell-shaded, etc, and stick with it. ES was a rather large mess of several and it made the game look like some sort of patchwork nightmare. When creating your art assets, take your time! I rushed through a lot of ES's graphics because I'm impatient and at that point, I wanted to just get it done. I suspect the game would've done a lot better had the graphics not been such a mess.



Gameplay & Mechanics

ES is for the most part, a basic top-down shooter with upgrades, bad things to blow up, crap to collect, and bullets to dodge. There's really not a lot to say about it in that regard, however there's a few design decisions I'd like to highlight:

The upgrade system
A few of the upgrades were fun, such as those that unlocked additional shots or other abilities. The problem however is that there are also a lot of “upgrades for the sake of providing something to spend your gold on.” Hit points, attack power, etc. Attack power in particular was poorly done because there was both a general “make everything hit a bit harder” upgrade and additional upgrades per shot type. At least with the weapon upgrades (fireball and withering darkness, anyway) the attack graphics changed as they were upgraded.

Another problem was that, after unlocking an ability for the first time, every subsequent upgrade level for it was nothing more than “more numbers for this ability” A lot of games do this, but that doesn't mean it's interesting design. When I reflect back on my playtesting of ES, there was a lot more excitement when I first unlocked a new ability (Felfire Infusion, Soulfrost, Chilling Touch, etc) than there was with upgrading each one further as I progressed.

Make your upgrades interesting! Make them matter! There's plenty of other, better ways to implement generic power growth.

Level design, progression, layout
I don't feel like I presented the level selection screen to players in a very interesting way. Rather than stale rows and columns, level selection could've been offered through a world map or something with more interaction. At the same time, that approach is more artistically demanding, so take this thought for what it's worth.

The last 2 levels are also hidden until the player finishes the first 8, which I think was a nice touch for the few who actually managed to get that far, but at the same time, it might lead players to initially believe there's less content than there actually is. “Dr Wiley” style endgame levels can make for a fun addition, but it's definitely not something to overdo.

Watch out for difficulty spikes! Play your own levels, and play them repeatedly! Then get friends to play them and give you feedback. It took a lot of playthroughs before I managed to balance the level and boss difficulties in a way that felt acceptable, and even then, it wasn't perfect. It's very easy to make one level too hard, then the one right after it too easy. If you want your game to be decent, you're going to need to spend a lot more time balancing your levels and dragons than you will simply making them.

The actual length of each level was a bit much, especially since dying means starting over at the very beginning. Had I split the levels in half, I could've doubled my apparent level content, eased the burden on players, and focused on offering a few really hard enemies instead of a crapton of fairly easy ones. I think I started to gain a better handle on level length when I made the bonus levels.

Scoring
Scoring players based on how many dragons they slew and how little damage they took on a given level seemed like a great idea initially. When I look back at it, all it really did was encourage players to max out their upgrades, then go back through the early levels and mow everything down.

Backgrounds
Simply put, I should have reused many of them create more level content. I made a silly mistake and assumed that each level needed its own unique background when actually, I should've used the same forest background for several forest levels, etc. The way I did it instead resulted in the player progressing through a bunch of really weird terrain types on the way to accomplish something that was meant to be straightforward. Forest → Dark Forest → Plains → Beach → Underwater → Ice Cavern → Normal Cavern → Lava Cavern → Void/Space → Pseudo-heaven. Yeah WTF? All the player really needed to do was stop the bad guy from blowing up the world.

Gameplay
Overall, I'm pleased with the way the game played once you got past the lame upgrades, dull story, and amateur artwork. I originally designed the game to be played with a keyboard (ideally mapped to a game controller) but quickly found that mouse movement was just so much more powerful and accurate for bullet hell games.

The Good

  • “Focus” was a mechanic that allowed the player to switch between a more spread-out shot pattern and a more focused/linear pattern at will, based on what they were dealing with. While originally meant to be a mechanic for keyboard users to down movement for more precise bullet-dodging, it ended up being more interesting as a shot pattern switch, especially since mouse-based movement (better for bullet hell games anyway) allows for precise or fast movement based on how you move the mouse.

  • Shot variety. Each of the player's shots did something different and meaningful, and that was further enhanced by unlocking various upgrades. One helped recharge energy faster, one put damage over time effects on enemies, one pierced, one could slow down enemy movement for a short time, and the special shot, Entropic Sphere, plowed its way through enemies for massive damage until it was destroyed (prior to maxing it) or moved offscreen. I felt that this worked especially well in conjuction with focus.
  • Boss mechanics. I was actually pretty pleased with how each boss ended up overall, even though I know I could make even better ones now that I'm more experienced. The health bar that doubled as an enemy position tracker was a very good thing, because the player could always monitor it instead of risking a quick glance up to see if the boss was nearly dead or not. Even though boss movement was extremely basic, moving targets made for a greater challenge, especially when trying to properly utilize Entropic Sphere.
  • Enemy variety. While poorly drawn, I was still pleased with how most enemies truly felt different from one another in how they behaved rather than just having bigger or smaller numbers. That's not to say that having an occasional “red slime” (old Dragon Warrior reference—red slimes had only slightly higher stats than normal slimes but were still basically the same enemy) in your game is a bad thing, but you don't want to make every enemy feel the same either.


The Bad

  • Random drops. Whether or not a player is able to clear a certain level could absolutely depend on whether or not several shield, health, or bomb powerups dropped back to back. While I think I kept the actual enemy RNG to a minimum, random drops and how much power they provide needs to be approached with extreme caution. More than most people realize.

  • Energy. Energy was a resource that filled up over time and was meant to regulate the use of Entropic Sphere. Entropic Sphere was also regulated by the fact that only one could be visible onscreen at a time, and as long as you kept your energy upgraded as well as the Energy Drain upgrade, there was never any real issue with being unable to use Entropic Sphere. With only 1 ability to regulate, energy was pretty much a worthless bar that meant nothing. I should've probably had either Entropic Sphere on a cooldown, or more than one ability tied into the energy resource.
  • Boss movements. Aside from the tutorial boss, every single boss was either statically positioned or moved in a slow, side-to-side, up-and-down bouncing pattern. Thankfully, this wasn't really a big deal because the shot patterns and appearances are what made each boss feel different moreso than the movement. Still, I felt this deserved mention.
  • Bullet graphic designs. There's a reason most shooters use really bright bullets—they're very easy to see! While I wanted to style them more after the enemies that shot them (slime blobs from slimes, splinters from tree stumps, etc) a few really didn't contrast with the backgrounds very well. For best results, use very bright bullets and darker backgrounds.
  • Backgrounds. With the exception of a few, the backgrounds were too busy. The original ones were even busier, and some people told me they even got headaches after a few minutes. Keep backgrounds for games like this simple enough that they don't interfere with the gameplay. They also probably didn't need to scroll so quickly, in retrospect...


Marketing

I poured a lot into ES, so of course wanted it to have some sort of payoff. I messaged a few of the big name sponsors to see if they were interested. Nothing. I put it up on FGL, realizing I'd soon have all these sponsor offers to sort through and choose from! Next day...nothing. Next week...nothing. After about a month, I began to realize that either sponsors were all insane for passing up this “amazing gem of a game,” or perhaps the issues I discussed above were curbing their interest. (Spoiler: it was most likely the latter)

So I opted for the self-publish route. For those who don't know, self-publishing is basically just wiring in an advertisement service plus a link to your developer page (which you should always do anyway), putting your game out there, and then hoping for the best. It's not very good income compared to an actual sponsorship and should only ever be a last resort.

There was, however, one success story here:

Viral distribution and the clever use of advertisement services. Kongregate has a policy that prevents you from displaying ingame ads on their site (because you get ad revenue from ads shown on your game's page instead) however, that does not mean the game can't contain any code for displaying ads. Using only a few lines of sitelock code, I told the game to suppress ads when played on Kongregate, but to show them elsewhere. Because there are a lot of seedy portals that auto-harvest from Kongregate's servers, I ended up netting a good amount of advertising revenue (I actually made more this way than from ad revenue through Newgrounds, which surprised me).

In the end, this game's income sources were ad revenue, a single sitelock sale, and a weekly contest from Kongregate. Not great by professional developer standards by any means, but acceptable for a hobbyist and great for a first game.



In Conclusion

Making a game with depth is no small task, especially if it's your first try at doing so. Don't expect your first game to be a masterpiece, because it probably won't be. Hopefully though, some of what I've talked about here can help you avoid a few stumbling blocks along the way. If your first game does poorly, don't consider it a failure. Consider it a learning experience and take that experience with you to make the next one even better! Good luck with whatever you may be working on, but most importantly, have fun with it!