Elite: Dangerous Lives Up To The Hype

Tis' the season, and Steam's Holiday sales are in full swing.  The Humble Bundle has been having a sale for a few days now, as well.  However, I have beyond too many games right now.  I bought one final game this season, full price, and that's Elite: Dangerous.  That's enough!

Elite: Dangerous is pretty damn good, but I have a hard time putting into words as to why.  It is not because I am particularly lacking in eloquence, but rather because the main appeal of the game is overwhelmingly tactile.

What do I mean by that?  Well, you may have heard game reviewers discussing a first person shooter game, saying that the guns need to have a good "kick" to them.  A combination of the animation, sound effects, and overall presentation that makes what's essentially a simple projectile spawner in to something that's fun to use.

In Elite: Dangerous, EVERYTHING has that kick.
  • The very MOVEMENT has a kick.  You move forward, and you will be treated to an overwhelming sense of acceleration as your view bobs back to the g-forces.  Every action causes the engines to roar in engagement, each ship having a unique sound effect for movement.  Even slightly nudging in a direction makes a different sounding maneuvering rocket noise, my Viper sounds a bit like a tiger purring as it rotates about.  Sometimes, the ship's frame creaks from a subtle movement, and you come to anticipate the sound as associated with that specific make of ship.
  • Interacting with the environment has a kick.  You run into a cargo canister or shard of asteroid, you will be treated to an appropriately loud KLANG as it bounces off your hull.  You bump into most things, you will hear your shields short-circuiting against it.  You enter the main entrance of a station, and you can hear the pulsing energy containment field on the station as you pass through it.
  • Traveling has a kick.  There is a very real-feeling simulation of what should be a thoroughly unreal thing: you are in a space capsule that is capable of folding space around it in order to make travel between the stars and planets, with reasonably realistic distances, much quicker than it would be possible with present-day technology (needless to say). The developers' incredible vision for spaceflight brings this to you.
  • Your ship's cockpit has a kick.  There's a wonderful number of tone-perfect beeps and boops, in addition to voice queues, for various activities you are undergoing in your ship, from landing to cargo scooping to fuel scooping.  You hear the fine grating of the nano scrubbers as they repair your hull damage at the station.  You hear the fuel as it is loaded aboard your ship.  You can freely look around your cockpit, which is fully modeled (including the fidgety pilot sitting in it) and the Oculus Rift is supported.
  • The stations have a kick.  Some stations are just platforms in space, but many of them are gloriously-modeled habitats that you fly through and watch ground traffic as you hunt for your landing pad.  Landing is compellingly nail-biting until you're good at it.   Once you are on the landing pad, you can choose to be moved underneath, inside of the hangar (and must be in order to have your ship outfitted).  Outfitting your ship treats you to a cinematic view of your ship's exterior as you swap your equipment, though it does stop just short of having mechanical arms come down and do the work.
  • Space itself has a kick.  The custom engine renders outer space the better than any game that I readily recall ever playing.  The suns look like shimmering, vaporious balls of burning hydrogen.  The asteroid fields look as brooding as you would expect for a bunch of dead rocks floating in space.  Hyperspace sends you plunging through limitless nameless nebula.  The planets... well, actually I guess the planets are a bit of a work in progress (making them landable is on the "to do" list) but they do not look out of place.   Overall, if this is not the best looking space game there is, it should be in the top 1%.
  • And yes, all the weapons have a kick to them as well, from the rolling metal of the multi-cannon gatling guns to the rapid pew pew of the pulse lasers.  Some have more kick than others, true enough, but they are all supplemented with a fine feel of being mounted on the ship in deployable hardpoints.  Much of what makes up this elaborate combat engine is basically all of the above combined.  An enemy damages my windshield, and what happens next is incredible.
In short, it simply has to be experienced to be appreciated.  Maybe, in time, the wonderful feel behind everything will lose its luster as the players simply extrapolate the whole experience down to what it takes to win, a natural process of learning.  Until then, I am in awe at the pains taken to make Elite: Dangerous perhaps the most immersive-feeling space simulator ever made.  I am entertaining thoughts that this game may well have finally unseated Volition's Descent: Freespace series for genre king.
In my previous entry, I was worried about how Elite: Dangerous does not have any of the sandbox aspects of the likes of Space Engineers or StarMade.  At most, the sandbox aspects in this game are currently a very watered-down implementation of what EVE Online does: cells of space in which you may encounter other players, but there is no crafting in Elite: Dangerous so the whole player economy that is the backbone of EVE Online is not really there, just the space and other (potentially aggressive) players.

With no easy way to see who owns what, nor a way to see where the most players are, the online mode in Elite:Dangerous is not all that different from playing it solo.  This is because the universe in Elite: Dangerous is huge, modeled at 1:1 scale to the real universe, and (in the spiral arm galaxy it simulates) there will be hundreds of thousands of solar systems, many rich enough that a player could choose to stage their whole spacer career there without ever leaving.  So, once you get far enough out of the starting sector (of which there are about 50 or so) what are the odds you will ever bump elbows with another player?

What I am suggesting here is that there needs to be better mechanics for interacting with other players over vast distances in order for the online mode to work.  For that matter, even playing a game with friends is hard, because grouping up at the same system takes a lot of coordination to avoid getting separated.  If you have a different enough ping than your friends, you may be excluded from interacting with them.  (Don't hold me to this, though, I really haven't messed with the "group" mode much, but from the sounds of things it is basically just a "solo" game that you invite friends to.)

Overall, my lead critique of Elite: Dangerous is that it could probably use a better end game.  Aside from PvP and grinding credits for bigger ships, what is there of worth to do in the Elite: Dangerous universe right now?  Well, if you complete certain missions (and maybe other things), then this would seem to lightly influence buffs and debuffs on NPC factions.  It is a start, but a really bare-bones implementation of player consequence that the game does not communicate clearly as making much difference.  Unlike EVE Online, there's no player faction space holdings yet, but there have been developer interviews that established it is something they have put some thought into.

Despite this, I am not complaining, as Elite: Dangerous is well worth the price of admission simply for this gloriously atmospheric, well-balanced, plays fantastic, space exploration/combat simulator.  However, as it is currently much ado about nothing, I wait with bated breath for what additional aspects of purpose the developers come up with and actually get around to implementing.


I still have not quite got around to download libGDX and beginning my baby steps in a new IDE.  I know a bit about programming already, so jumping aboard Java wholeheartedly is not so very intimidating.  Understanding variable scopes and object orientation is the hardest part of getting used to that language, and I am pretty much past the greater bulk of that.  The intimidating bit is learning a whole new library.

I estimate that, if I worked really hard on getting into libGDX, I could catch up with what Construct 2 would allow me to do now (and surpass it) in in a time frame that could be as early as two months or as late as two years... and that later bit is when $130 for Scirra's game-making-made-easy software has a certain attraction.  I would not need to code my own pathfinding routines or tile engines, and so on, because those are already done.  However, I would have to settle for games that run in a browser environment...

One must walk before they can run, but am I holding myself back by not breaking into a sprint considering programming is probably my strongest skill to begin with?  Says the person with little math skills.  Very well, then, let us procrastinate some more.

Now Playing: StarMade.

If anyone ever tells you that Space Engineers or Starbound are essentially, "Minecraft in space," you can sit them down with StarMade and say, "No, THIS is Minecraft in space."  StarMade is about as literal of an interpretation of this as you could ask:
  • Everything is made of full-sized cubes that come up to about half the height of your character.
  • The planets are essentially many-sided dice (12?) made of cubes.
  • You can harvest any cube you see and it goes into your inventory for later placement.
  • StarMade was programmed in Java for some reason?
The main difference between StarMade and Minecraft is that whole, "in space" business.  Instead of a world that stretches on virtually forever, you have a universe full of solar systems, planets, space stations, asteroids, and wormholes.
StarMade: Having lost my previous ship to something inexplicable (a bug?) I have decided to create a new one with an interior cockpit.
A secondary and significant difference is that you refine raw minerals into usable minerals in StarMade.  Picking apart a derelict stations will get you scrap metal which forms the basic raw materials.  Raw materials make the basic factory blocks.  Basic factories create standard factories, and standard factories crate advanced factories.  The more advanced the factory, the more advanced things you can create, but this will require increasingly more specialized  forms of raw materials, most of them taking the form of "capsules," the ultimate refinement of ore (which can only be found on asteroids and planets).

Good insight can be found by comparing StarMade with Space Engineers, as they are essentially two games of Minecraft in space described above; the only goal the players are given in both games is to find raw materials, mine them, refine them, and then use materials to build bigger and bigger ships and stations.  So where do they differ?
  • The primary difference is in the scope of the two games.
    That imperceptable little yellow cube in the middle is the 50km square sector I am in right now.
    • StarMade is all about having a scope of an epic-sized universe with planets, stations, ships, and asteroids. 
    • Space Engineers focuses on a small patch of local space with a few asteroids to mine.
  • Graphically, the approach is different.
    • StarMade, like Minecraft, works in cubes.  Everything is cubes, including the terrain.
    • In Space Engineers, only the artificial blocks are cubes, and some of them are premade parts that are not cubes.  The asteroids are not made of cubes, they look like real asteroids (except perhaps not as pockmarked).
  • The philosophy behind ship building differs in fidelity.
    • In StarMade, you can build epic-sized ships, but it is very crude.   You build them block-by-block, putting down weapon blocks and slaving them to a weapon computer block, where the difference between a big gun and a small gun primarily comes down to the number of blocks you use.   Thruster blocks largely only go in one direction, they boost your acceleration, where the size of your ship decreases your acceleration.  Turrets are actually smaller ships you build and dock onto larger ships.
    • In Space Engineers, the bigger your ship, the more complicated it gets.  Conveying cargo between parts is not handled automatically; you need to install a conveyor system to move cargo between parts.  Weapons and turrets are all predefined parts, so you will not be building larger scale guns by putting down more of them.  Thrusters only work in one direction, so if you want your ship to have the full range of maneuverability you need to place them on the back, front, top, bottom, and both sides.
  • There is a difference in natural threats in the games.
    • In StarMade, there is an NPC "pirate" faction that will occasionally dispatch fighter ships to harass you.
    • In Space Engineers, there is no NPC opposition.  You can turn on meteor storms to pepper the local space periodically, and there is the occasional auto-piloted cargo ship passing silently through the sector.
  • Both games largely depend on player versus player for any form of end-game activity.
    • In StarMade's case, you can create your own faction and then designate a planet or station as your main faction base, and that base (and all docked faction ships) will be invulnerable, which gives them a good place to fall back to in order to keep an ongoing conflict going with other player factions.
    •  In Space Engineers, the maps are too small, there's no real room for politics.  In a way, PvP is more akin to a traditional first person shooter deathmatch setup, except for the building elements.
I would say Space Engineers is a more refined product, and StarMade is a bit bugger, but what's the point of that?   They are both in early access status, and little said here is permanent.  It would wager to guess that it will probably remain that Space Engineers is more about the local sector scope while StarMade is more about the entire universe scope, but that's about it.

Considering playing: Elite: Dangerous.

Elite: Dangerous is out, no longer in early access, and I'm tempted.   I could pay $60 for this game and end up with a highly-elaborate, cool space-shooter with some open-universe aspects to it.  What's more, the attention paid to the sound effects and overall presentation make it perhaps the most atmospheric space fighter game ever.
What's the holdup?  Well, I've already played Elite before, and this is essentially the same thing, modernized, with some more ships from later games in the series.  I know how this goes: you go out there to earn money for bigger ships in order to earn money for bigger ships, then you get to the biggest ship and there's nothing left to do.

Even when made massively multiplayer, even with the finest multimedia presentation there ever was going for it, does that grind sound like a worthwhile endeavor to you?  Games you can get for $10-$20 have better sandbox aspects than Elite: Dangerous, which wants me to pay $60 for a universe that never changes.

The creators, Frontier Developers, have mentioned they are going to add more sandbox to their elaborate space fighter simulation now that it is released, but I've seen such promises broken in the past from different parties, so only time will tell.  In the meanwhile, I am feeling mighty hesitant to slap down so much money for a 1984 game made current.

In The Dark, Per Usual

"If this is not windy enough to knock out the power," I said, "I would be surprised."

Sometimes, I hate it when I'm right.  I was in the middle of a game of Cataclysm:Dark Days Ahead that likely would have consumed the rest of my weekend when a power outage ejected me rudely from a virtual simulation of a town without power and into a real one.

In my case, five hours without Internet.
2 hours after lights out, I'm bored, the best thing to entertain me being an Android tablet, and most of its content is sitting uselessly on a cloud.  However, it so happens I do have this nifty blogging app installed on it, and it is as good of time as any to recap my weekend.  [When the power came back the next day, I added some pictures, links, ect.]

Well, despite wanting to mix things up more, the entire 4-day bizarro weekend was consumed by Cataclysm:Dark Days Ahead.

I mostly tried to create characters who had ninjitsu, as this was the most interesting-looking of four martial arts available in the game that are usable with a melee weapon.  The main benefit of this style is it makes no noise when performing a melee attack.  Secondary benefit being a nifty damage multiplier on a "precise" strike.  Aside from that, ninjutsu is pretty useless compared to many unarmed styles of combat in C:DDA, as it does not consistently stun enemies.  A stunned enemy is an enemy prone to free attacks in this game.

Using a melee weapon was key for my characters because my previous survivor was a "tiger style" unarmed combatant who butchered zombies with incredible ease.  He consistently did with his hands what high caliber guns could not.   While it is realistic that being good at a martial art should largely nullify the threat of idiotic human-shaped belligerents, unarmed combat in C:DDA is a tad overpowered for this same reason.  Another reason why my "tiger style" character had such an easy time of things was because I had set the zombies at half their normal speed, with the exceptionally-dangerous "special" infected turned off.

Now, I was putting myself up against full speed (but still mundane) zombies, and over a dozen characters died to me not respecting how much harder the game was now. 

The core of the problem (and a brilliant bit of balancing) is that fighting your garden variety "green" zombie is a war of attrition.
  • Even a one-on-one fight has a chance to hurt, especially in poor conditions.
  • Little hurts add up to big hurts.
  • Big hurts slow you down.
  • Slow down enough, and the horde trailing you (perhaps attracted by the noise of your scuffle) is going to swarm you, bringing about a typical zombie movie bad end.
  • This is to say nothing for the foes that could outrun you even at your best.
Even before this latest run of unfortunate characters, I had played enough of C:DDA to know this, but knowledge alone was not enough.  My characters kept dying, time and time again, mostly for having the audacity to think they can just stroll through a zombie-infested town and outrun the ravenous natives.

In one memorable instance, I had smoothly infiltrated a town under the cover of night and was pulling a shopping cart full of delicious swag away to safety.  Then, with no warning, bear appeared out of the gloom within swatting range.  I do not know what I did to anger it, maybe I simply got too close, but I did not survive that night.

Eventually, a luckier survivor stumbled upon the ultimate solution for the inexperienced melee character: a .45 caliber hand cannon by the name of an,"L2031 Enforcer."
Judging by its overwhelming firepower, this fictional gun is probably intended to resemble something like this Colt 1851 Navy Revolver.
I learned that, even in the hands of green survivors who are literally unable to hit anything with a pointy stick, having a powerful sidearm at their disposal will almost ways blow the heads clean off persistent enemies at point blank range.  For example:
  • A Z-9 (zombified police dog) catches me, its superior speed making running impossible.  BLAM!  
  • A couple of times, a bear came lumbering out of nowhere, thinking I would be an easy meal.  BLAM!
  • I thought I could get some archery practice by shooting some kind of overgrown lobster mutant on the other side of a chain length fence, but they tore right through it and were faster than me.  BLAM! BLAM! (Hey, you armored SOB, where do you think you're going?)  BLAM!  
Getting into these situations is usually a game finisher right there, an appropriate punishment for pushing my luck too far.  Instead, this gun saved my character's incompetent hide several times over.  (No wonder the NRA loves these things!)

Even now that my character has survived long enough to get some decent melee skill and armor, I keep this gun in an ankle holster as the, "Oh, shit!" option.  This is the right way to use them in this game: bullets are rare and the noise will attract unwanted attention if you use them indiscriminately.  Yet, when you are backed into a corner, they often can blast you right out of it.

What I should have been doing:

That said, playing C:DDA this much was craven procrastinating and I knew it.  Every time I stopped, I found it was either that or come to grips with the choice laid out in the last entry: to make my game the easy way or to make it the ideal way.

The easy way would be to shell out $130 for Construct 2 and compromise my game vision by doing away with dynamically generated spaces and instead have static scenes defined which could still have some level of dynamic change.  Hey, it works for ASCII Sector, and that's one of the game ideas that I want my own to play with.  (For that matter, probably 99% of games do not require the functionality of creating new space and swapping it in and out dynamically.)

The ideal way is not to use Construct 2, or any of those other hobbyist game development kits, and instead pick an integrated development environment (IDE) that already has fantastic Intellisense (such as Eclipse or Microsoft Visual Studio or basically any professional-grade IDE).   Contrary to what I was saying in the last entry, it may not be a matter of reinventing the wheel, considering there's quite a few cool looking application programming interfaces that at least give me a rolling start.

Along the lines of the ideal solution, I did a bit of research, and the current main contenders look like this:
  • It is currently looking to me like libGDX might be the best way to go, judging wholly by most projects made it cataloged on indiedb being rather good.  It also has surprisingly good performance, it is perhaps the best overall API now that Microsoft XNA is being retired.
  • Monkey X translates to many native platforms and is so simple to code in that I would say it is a better first programming language than BASIC.  Unfortunately, even as amazing as that is, few examples exist that demonstrate it can do much, yet.
  • If I am willing to go as far as performing that much coding, there is really no reason not to give Unity3D another spin, and that engine positively overflows with proven examples it works and works well.
I still have yet to decide, but I probably would not turn down a Construct 2 sale right now, just as a last ditch effort to see if the easy option will truly wash with me.  At the very least, it should make for an effective prototyping tool.

Unfathomable Shades Of Days Ahead

Though my last post was pretty much a rehash of what's been going on over the past few years, it was an important milestone in that I had reached a slightly higher level of perspective as pertains to my game development endeavors.

Moving forward, Construct 2 was looking like a good way to go because was it the only one I tried besides BYOND to include an interactive tile engine (the tiles in Game Maker being non-interactive) and the best automatic code completion mechanics of any of the engines mentioned here.   I have said it before and I will say it again: Construct 2 is a joy to work in.

Yet, when I asked a question on the official forums about dynamic map creating, loading, and saving, I got no answer.  This confirms my suspicions that this is one thing it cannot do and I am not entirely sure a Javascript extension could get around that given the HTML5 sandbox limitation.  This throws Construct 2 into being just as limited as any of the other choices in my eyes... but it's still a lot more fun to use.

If there is no reasonable middle path to be found, then there is only two ways to go from here:
  1. The easy way: Compromise my creative vision enough that I could use any old engine to make it happen.
  2. The ideal way: Make my own engine so I can do all the weird-ass stuff I want to do with it.
It is not as easy of a choice as it looks. We would all do the ideal thing in an ideal world, but mortal limitations often intervene.  In this case, the limitation is time, creating your own engine is reinventing the wheel and taking responsibility for making sure it can roll, even before going where I want to go with it.  I may not be a complete slouch at coding, but I do not want to invest that kind of time unnecessarily.

Of course, making a game is hard, regardless of your methods.   But I am not so sure it is the effort that is proving off-putting so much as the fear that all my time will be wasted because the thing I want to make cannot be done on a given platform.

I am anticipating Construct 2 will go on sale sometime around the end of the year, what with all the holiday festivities to be had.   Why spend $130 today when it could be half that tomorrow?  Steam has trained me thus.
Such frugality is proving costly.  Prior to this, I had been working hard to stoke the fires of creativity.  Now, that energy is failing to find outlet and being absorbed by my psyche instead, destabilizing it, perhaps out of remorse about my failure to realize something.  To take my mind off of it, I am procrastinating, which makes things worse because I am wasting more of my preciously limited mortal coil time to not making a game.  When Construct 2 does go on sale, it will probably be around 33% off, is that really worth the wait? 

Now Playing: Cataclysm: Dark Days Ahead.

This is not the first time I played this glorious post-apocalyptic zombie survival roguelike, perhaps the only one worthy to bear the title.  However, with version 0.B Brin out, there has been quite a bit of changes.

For me, the foremost change is that the temperature system has been balanced enough to be survivable.  I can also see a lot of cars on the streets of cities, which means there is a lot of parts available to work with, and some of those cars may be able to be driven immediately.

They also added a bunch of cool new start options.  You can generate worlds without the more dangerous monsters, monsters that are "fast" or "slow" compared to the default, and so on.  You can choose to take a profession, which is basically an archetype for skills and equipment you have when you start.  For bonus points, you can choose to start in perilous situations such as a burning building.  Alternately, if there is a favorite kind of store you prefer to start in (military surplus stores are mine) then you can choose to do so now.
Though Cataclysm: Dark Days Ahead is typically a very hard game, it allows enough freedom to create monstrous twinks like this.
The game seems a little harder than I remember because static spawns are enabled by default and, with the special infected mobs, this will make cities absolute death for most starting characters.  Zombie Hulks move faster than you can run and smash you through walls with the power of their blows, making them so incredibly powerful that I opted to turn off completely the special infected in my world generation.  Even without, the Z-9s (infected police dogs) are pretty likely to kill most starting characters.  Some of my characters survived all this only to blunder into firing arcs of murderous military automated defenses!  In these ways and others, Cataclysm puts a heaping helping of "unfair" in the mix of its survival simulator, but what do you expect?  It's a roguelike.  

In the long run, "static spawns" seem to backfire a bit because, if you defeat enough zombies and chop them up into bits so they can't revivify, your opposition disappears, and you get free run of the town, and that is sort of boring.  All that city loot without risk leads large surpluses of food and water and long, monotonous days reading skill books.  Eventually, boredom will claim you, you will go and explore some abandoned labs or something, and run across some kind of hideous post-apocalyptic monstrosity that will end you.  Then it's back to scrabbling for basic necessities with your next wasteland newb!

Ultimately, I am thinking Cataclysm needs to turn up the heat some more.  If you simply hole up and live out the rest of your days in relative safety, it is kind of boring, especially with static spawns (since that means you will not be encountering any more opposition once it has been eliminated).  Cataclysm also needs to provide an end-game goal, some way to do more than "survive," to "beat" the game, thereby giving the player a greater goal.  Of course, it could be that all of these things are already in the game, but I have yet to encounter them.  I will say that Don't Starve provides a good example of how this could be done.

My Divergence Of Lost Purpose In PC Gaming

Being a computer gaming enthusiast for over three decades, I love my free time.  I would rather have free time than I would a wife, kids, wealth, fame, and other bothersome things that severely bottleneck a man's free time.  I am an evolutionary dead end of loving my free time... it's a shame I don't seem to know how to use it.

The problem, as I see it, is this: I'm primarily a gamer, and most games suck.  Of course, this is subjective, but when you have been gaming for as long as I have, you will have burned out from nearly every kind of popular genre there is.  For this reason, when I buy discount games off of Steam wish lists, it's usually a waste of money: most of them will go unplayed, particularly the clone genres whose core gameplay fundamentals I played to death long ago.  Ergo, I seem to not know how to spend my free time.

The main genre I can still enjoy is gloriously elaborate escapism simulators, whether they be massively multiplayer role-playing games or open-ended environments like Minecraft or Dwarf Fortress (but not so much DayZ because I am a total carebear).  Perhaps procedurally generated games along the lines of FTL and Dungeon Of The Endless qualify, because I enjoy those games too, but they have to be done right: procedural generation is not implicitly a good game in itself.

If you ask me, this is where the bar of PC gaming has risen, take a look at the kinds of games played on the Yogscast channel, this is why.  Tastes will vary; maybe you like MOBAs or Counter-Strike clones, but wide-eyed gaming nerds such as myself want better escapism simulators.
Point of proof: Starbound earned over 1 million Kickstarter dollars and twice that in sales, even if it never played as good as it looked.
Unfortunately, I am bored of the best games in that genre that I know of.  Mostly it is because I have exhausted everything worthwhile to do in the ones I have played.  This is to say nothing for where my personal taste may have varied from the developers'.  Yet, even if a game completely matches my tastes, there is no way for it to be an everlasting jawbreaker.

What's a perpetually unsatisfied gamer to do?  Take up a new hobby?  Man up and face reality?!  Of course not!  I am not a computer game geek for nothing!

I have come up with a different solution: make my own game.  The appeal of this solution is manyfold: it assures my game matches my tastes, it overcomes the everlasting gobstopper paradigm by ever-building from my creativity, and making a game can be even more fun than playing one once you are in the zone! To these ends, it is something I have been doing on and off for about a decade now.
  • The first engine I tried was BYOND.  I learned a lot about developing in an event-driven engine from it, but ultimately I decided to abandon the platform because a few core reasons:
    • I encountered some critical issues when performing file I/O operations on a particularly ambitious project, and this left me doubting the engine could do what I truly needed it to do. 
    • My yearly subscription (only really needed to host games) lapsed for the second time and they wanted to hike the rate.  I decided to try my luck elsewhere.
    • Barely anyone is willing to install the BYOND software on their computer, so I would have hardly any audience to appreciate my games.
    • A general lack of basic operations, such as projectile routines.  The BYOND community shared a lot code that worked around this, but it is often better to have standardized, optimized solutions built into the engine.
    • The lack of fidelity in floating point operations lead to unpredictable behavior.
  • The second engine I tried was Game Maker.  I regretted abandoning the ease in which I could create multiplayer persistent state environments in BYOND, but I could not deny that Game Maker was extremely flexible and powerful.  Unfortunately, I soon developed reasons I wanted to try my luck elsewhere:
    • The way YoYo Games crammed drag and drop functionality into the development environment is extremely ponderous and annoying to experienced coders.  I have to click three or four buttons just to select an object, open up its event sheet, select the script call, and open the script... I had better hope the code I was looking for was there, because Game Maker expects me to track it down from the event that calls it every time!
    • The room editor is very awkward to use.   Particularly difficult is manipulating multiple objects at once.  The room editor's graphical user interface is quite unorthodox, so your usual habits will just make mistakes.
    • The way most of the code executed is restricted entirely to rooms is awkward to deal with for an escapism simulator.  The bulk of available room functions and the instance functions suffer many omissions that render me unable to do common things I was able to do just fine in BYOND.
    • Tilemaps are wholly decorative; this is not really a tilemap engine, and does not actually have interactive tilemaps unless you fake it.  It is technically doable in Game Maker to have a huge room built out of objects emulating tiles, but this will slow down performance a bit.
  • This week, Clickteam Fusion 2.5 was on sale, and I decided to risk it despite knowing Construct 2 is generally more flexible, because Clickteam Fusion 2.5 is certainly more powerful in terms of end product.  Unfortunately, having dabbled with it since Monday, it looks like Clickteam Fusion 2.5 will probably not work out for me.  Yes, if you know my history on this blog, this is probably not a great surprise to you, but I was surprised at some of the things I did not really notice until I rolled up my sleeves and tried to use it seriously:
    • There is no tile engine object.  Tilemaps are important to 2D escapism games because they allow for a more comprehensive simulation of space.  There are community-made tilemap extensions, but I was unable to locate one sufficiently powerful for my purposes.
    • Unlike Game Maker, I cannot "fake" tiles with instances of objects because, in Clickteam Fusion 2.5, there is a limit of 20,000 active objects per "frame" (frames being essentially what Game Maker calls "rooms").   This would limit me to a room of roughly 141x141 tiles (141 x 141 = 19,881) and that's before adding any active creatures or objects to the room!  The usual proposed solution would be to use cosmetic "backdrops" instead, but that is no good for me because my game concept needs every tile to be potentially interactive and elements off-screen to be actively simulated.
    • The event system in Clickteam Fusion is actually a lot more comparable to proper coding than advertised, which is great for potential flexibility.  However, coding is a bit of a chore.  Users of this software are forced down some very rigid paths of what they need to click on in order to pull functions they need.  More importantly, those functions have very limited accepted arguments, though you can often learn to work around this.  
    I think there is actually a very good reason for much of the limitations in Clickteam Fusion: engine execution efficiency.  In other words, the user is probably restricted in the way the data is entered and how many objects they can use because it lines up nicely with engine optimizations, producing better performance. 

    However, coding in Clickteam Fusion 2.5 is surprisingly hard work; it takes me about four times as long to do most operations in Clickteam Fusion that it does in Construct 2!  (Although probably less with more practice using the engine.)   In a more clear-cut example, there is no such thing as families in Clickteam Fusion, which means if you have several different types of enemies to shoot, you basically have to code projectile interactions for every single type of enemy.  Inheritance, where art thou?!  (In fairness, Construct 2's "families" are not true inheritance either; "families" look to simply streamline coding functions by duplicating the related lines for associated objects.)
Maybe if I did not take over year-long breaks from development here and there, I'd be significantly better at doing significantly better at having completed a game by now!
Honestly, I think I could have figured this out without months of dabbling some ten years ago.
I'm getting old.  Sitting around on my butt all day probably isn't helping the neurons work any better, either.
Anyway, my going through engines like this relates to something I said not too long ago: the engine you choose to develop your game in has a lot to do with the type of game you plan to make.  This entry's important realization is the importance of fully-simulated tilemaps to my game concept.  Along these lines, none of these engines were exactly right:
  • BYOND was closer to the type of game I wanted to make than any of the other engines I have dabbled with because it was, from the start, designed to simulate virtual spaces.  Honestly, I never brought up my file I/O issues on the forum, so it is entirely possible there was a solution or a fix that could have happened.  Consequently, this platform's main problem remains the difficulty of reaching an audience.
  • Game Maker could potentially make the type of game I want to make, a tile-based escapism simulator, but not without the impediments mentioned above.  I already noticed its performance drops significantly with lots of objects-emulating-tiles on the screen, though hundreds of independent entities can be simulated readily enough (and the YoYo compiler may excel at those kinds of operations).
  • Clickteam Fusion 2.5 could only make the type of game I want to make if I were to make some massive concessions in my design, probably abandoning entire rooms of interactive tiles, or else coding my own extension to make it work.
I have determined Construct 2 could definitely make the kind of game I want to make and do it in an unusually hassle-free manner, because it does have a tilemap object, I was able to get it to work with the pathfinding behavior, and I procedurally generated a tilemap in no more time than it took me to do in Game Maker.  However, it is still not ideal because it is not custom-tailored to my game concept, consequently I would need to keep things reasonably small scale, with liberal instancing, because once you get a couple hundred active objects the HTML5 performance starts to tank.

I worked exceptionally hard in my game development dabbling this weekend and now my path is clearer.  If Construct 2 goes on sale soon, I will probably buy it, as I can see this as realizing a serviceable prototype with less work than it would take for me to accomplish the same thing in as Game Maker.   Yet, maybe a serviceable prototype is not good enough for me in the long run, in which case I may want to look into the likes of Cocos2D or Unity3D despite them being considerably harder to use.

Exploration Over

(Please pardon that this is the third repost of this entry.  I have been using an Android app to perform edits, and keep accidentally hitting the 'save as draft' button instead of the 'publish' one.  Whose wise idea was it to make the 'save as draft' button look like a 3.5" disk?  I keep thinking "save changes" when I see that.)

Now Playing: Lots Of Stuff, Maybe...

In a few hours, the Steam "exploration sale" will be over, an event that went from November 26th to December 2nd.  I am fairly certain what they set out to "explore" was the depths of my wallet.  Since I already have so many games on Steam that I require a day planner and an abacus to get around to playing any of them, I kept my purchases pretty moderate this time around:

I don't care if this is YouTube bait, putting a goat in a tudexo and tophat is worth a few bucks to me!
Goat Simulator - With the additional content provided by the "MMO expansion," this half-baked deliberately buggy silly game has become a fully-baked deliberately buggy silly game.

Sid Meier's Civilization V: Brave New World - Civilization V is one of those games that proves to computer gaming cynics that turds actually can be polished to diamonds with enough time and effort.

It released in 2010 a rather buggy mess that played coherently but suffered from performance issues, spotty artificial intelligence, and lacking in many vital features we came to expect from prior games in the series.  Over the next couple years, they patched out the greater bulk of the bugs and left us with a thoroughly alright game.   Major improvements were released with their first expansion in June of 2012, Gods And Kings, which also introduced mechanics for religions as well as brought back the missing feature of spies.  Then, in July of 2013, they released further improvements in Brave New World, which brought back the missing feature of trade caravans as well as revamped the "cultural" victory condition to be a bit less arbitrary. 

In my limited experience with this latest expansion, I will say that Brave New World makes the game a whole lot more interesting because all those trade caravans mean a lot more gold income, which means you can maintain a lot more of a standing army.  During peacetime, this is necessary to protect your trade routes, but this doubles as epic resistence potentional for when diplomacy fails, and basically just a lot more units for mischief in general!  A very welcome change overall.

Moonbase Commander - A 2002 game from Humongous Entertainment that remains among the best strategy artillery games ever made.

Might & Magic X - A 2014 fan indie revival of one of the biggest names in the turn-based, first person dungeon crawl genre, X has harnessed modern day technology to make things compellingly 3D, but stays relatively true to its roots, and is as brutal as any Might & Magic game ever has been.

Thief (complete series) - The first Thief game was released shortly before the turn of the century, and breathed life back into the first person shooter genre by taking the focus off of spastic gunplay and instead shifting the focus towards stealth, where our light-fingered protagonist is a poor fighter who needs to avoid being seen or heard.  The resulting series has delivered some of the most atmospheric experiences ever to grace computer or console, thanks in part to a setting that combines magic with steampunk.  The last two installments, Deadly Shadows (2004) and Thief (2014) fell a little flat compared to the original two, but this nevertheless remains a rather cool series in many gamers' minds.

Octodad: Dadliest Catch - A goofy physics game of far greater production values than most of the genre, this one casts you as a octopus trying to pose as a human without attracting too much notice.  Much hilarity ensues as even the most basic landlubber activity is rather difficult for a octopus controlled with a very literal control scheme.  Even more incredibly, our protagonist is quite worthy of sympathy, he's just trying to be the best octodad he can for his human family!

Dragon Age: Origins - When this game came out, it was so slickly done that it looked like Bioware had successfully revived and modernized the kind of real time strategy roleplaying game they did much to invent with the Baldur's Gate series.  Dragon Age: Origins remains a glorious game, thick with lore and quality gameplay, no doubt about it.  Unfortunately, its sequel was a massive disappointment to fans by largely failing to hit the same notes, and that makes it hard for me to want to get involved in playing this one.

Planetary Annihilation - Latest game from the creators of Total Annihilation and Supreme Commander, I always found this formula of no-nonsense energy and metal generation to lead to a real time strategy experience every bit as good as StarcraftPlanetary Annihilation is the product of over 2 million dollars of crowd funding and years of development.

How did it turn out?  I played one campaign map against the AI... the AI commander built about a half-dozen structures and stalled, so I spent 48 minutes carefully building my way across three planets, expecting resistance at any time, only to find there was absolutely no opposition.  Apparently this "finished" game does not have working AI. Really?  REALLY?  Oh well, I got it for 80% off, I can't complain too much.
Really Big Sky - A surprise favorite of mine, Really Big Sky is a gloriously psychodelic side-scrolling shoot em' up rife with special effects and clever gameplay innovations.  It's the kind of game I would make if I were to take Clickteam Fusion 2.5 and try to implement every trick the engine gives me into a single game. Incidentally, this was indeed the engine the game was made in, and the result really says something for its capabilities!

Gimbal -  I mostly snagged this one because I often see my brother playing it and it was on sale.  Well, okay, it's a top-down space shooter where you build your own ships from prefab parts, with some allowance for real physics so you need to have your ship control realistically, a pretty decent start. 

Unfortunately, all you can do with the designs you have built and tested is participate in old-fashioned multiplayer arenas, a grossly obsolete form of gameplay.  When you get there, you are probably just going to run up against a guy who found a way to mount dozens of guns and blows you away because there is not a whole lot of balancing factors in the game to prevent them from pulling stuff like that. 

Good thing I got it on the cheap, as games like Gimbal tend to taunt me by being really cool core ideas that are being leveraged ineffectively.

Now Developing: Something In This Fancy New Engine I Bought.

The last thing I bought on this Steam sale was a personal license for Clickteam Fusion 2.5.  There was quite a bit of cognitive dissonance behind this choice.  

Recall that, on the heels of my last entry, I had decided that I needed some good "prototyping" software to help me iron out my designs.  (From there, I could move on to realizing a finished version on a more specialized custom engine if needed.).  It is quite possible that Construct 2 is better at prototyping overall due to having more pre-programmed behaviors and a slightly more user-friendly development (its methods are very similar to Clickteam Fusion 2.5, but a little more advanced).  However, it was also quite possible that a given design might need performance in order to be an effective prototype, and Clickteam Fusion 2.5 puts out optimized games that put the opposition's performance to shame (except in HTML5).

In the entry before the last, I did a more thorough breakdown of comparing the two engines.  It was a very close battle between these two products.  Performance is important, flexibility is important, and the difference in ease of use is minor enough to be no tiebreaker.  At the end of that entry, I had speculated that I may well end up choosing the one that went on sale first.  
Well, Clickteam Fusion 2.5 was on sale for 50% off this "discovery," Steam sale, Construct 2 did not deign to be on sale at all during this time, so we have our winner!   In retrospect, I feel I made the right choice because, with Clickteam Fusion 2.5, I am not only getting an IDE that is almost as easy to use as Construct 2, but I am also getting one that has been proven to realize some extremely capable games.  Construct 2 has yet to prove itself outside of the HTML5 platform.

In fact, Clickteam Fusion 2.5's performance is so good, I may well not require another engine after the prototype is complete inside of it.  Who knows?   Maybe I will end up buying Construct 2 on sale for the purpose of prototyping a game to make in Clickteam Fusion 2.5!

I am all out of excuses now; okay, Mr. Hotshot Wannabe Indie developer, you got your engine, now you need to try to make things.

Invested Gambits And Infested Planets

As much as I enjoy my time off, I was naive indeed to believe I would get Thanksgiving weekend to myself, as I am about to be up to my ears in familial obligation.  Sometimes, my introverted nature makes me think I want to live in a fallout bunker and lock the entry hatch. 
Just provide me with adequate nutrient paste and a computer, and I'm set!
However, as much as I fear I am not making adequate time to slake the thirst of my unexpressed creativity, I already made some pretty good progress this week by realizing something I should have realized long, long ago.

What was the problem?  Well, if you look over the past few game development entries, you will see me quibbling extensively over wanting three things in a game engine:
  • Ease of use - How quick easy it is to translate my ideas into the game.  In the long run, ease of use is secondary to simply exerting sufficient will to make it happen... but it is still a real time saver.
  • Flexibility - The accuracy in which I can realize my ideas in the game.  In terms of choosing an engine, flexibility is important only if the engine does not implement the desired behaviors by default.
  • Power - The ability of the computer to simulate my ideas as integrated in the game.  Power is mostly only a concern if the game requires the computer does a lot of "thinking" to work.
The problem was that I was having difficulty picking an engine that would have a satisfying mixture of all three.  Complicating the issue, different game concepts are going to require different measures of each.  I just could not come up with a satisfactory choice!

So, what was my breakthrough?  Use more than one engine.  
In retrospect, this was obvious.  I must be getting old.
Specifically, I need two engines:
  1. A prototyping engine.  This engine focuses mostly on ease of use in order to facilitate speed of integration.  The purpose of the prototyping engine is to see if my the game concepts works out in reality as well as I imagine they will.  The faster I can bang out mockups, the faster I can make up my mind.

    Construct 2 should excel here, as I can get around any lack of flexibility with JavaScript extensions, with the exception of being restricted to "Scenes."
  2. A dedicated game engine.  This is the main shortcoming of premade game engines is that they are usually made to create specific types of games, and can only extrapolate inaccurately from there, hemorrhaging power and flexibility as they go.  It is a noble endeavor to try to make an all-in-one 2D game engine, but 2D games vary radically in concept and execution.

    Here is where I would probably turn to cobbling together a custom engine from existing APIs like monogame, possibly cutting corners by using an extendable multi-platform software delivery system like Cocos2D.  This is assuming a specific game does not run just fine on the prototyping engine, which is possible.  Honestly, for a lot of concepts, Unity3D or GameMaker will work just fine, but personally I find them just slightly too awkward for fast prototyping.
So I will have my ease-of-use where I need it (the prototyping stage) my power where I need it (the final, dedicated game engine), and flexibility throughout. 

At the moment, I am waiting to see if Construct 2 goes on a Black Friday sale.   If not, I will snap it up anyway, it has a pretty generous price by default.  Anyway, my weekend ends after Thanksgiving, so that puts the kibosh on getting anything meaningful done this week.

Now Playing: Infested Planet.

A lot of really quality indie games coming out these days are basically just such radical remixes of older game concepts that they become significantly novel (whereas AAA developers are typically constrained by flighty investors who do not even allow them even that much freedom).  For example, when I played Space Run, it was basically a form of tower defense game turned on its head by taking place on a cargo ship running through blockades of pirates and asteroids.

Infested Planet is a novel retake to real time strategy genre where you control a small handful units (mercenary space marines) that freely respawn on destruction and your opponent is never-ending spawn of aliens.  This differs considerably from your average real time strategy by being less about resource accumulation and having a foe of a wholly different nature than your own units.  However, there is a "build point" allocation that limits how well you can equip your marines as well as limiting available support structures, such as turrets, and labs which mostly unlock various passive buffs while built.

Another similarity Infested Planet shares with Space Run is that I have surprisingly little to say about it despite the game being so much fun to play because they are both simple concepts executed well with a nice presentation and smooth gameplay.  Actually, I will say that Infested Planet might not be quite as well balanced, because I find some of the upgrades you are provided with to be next to useless in most situations, but this is a fiddly kind of balance critique considering the overall balance is just fine thanks to careful weighting of build points.

Overall, I would say that Infested Planet is not quite Indie God Tier, like FTL or Dungeon Of The Endless are, but it is good enough to be a near miss, and I do not regret my time spent playing it.  I completed the campaign in one 8-hour sitting, but there's still plenty of additional modes to play.