Elite: Dangerous or Endangered

Right at the end of 2014 there was a discussion between the Elitists of the Pelit-magazine on the subject of who would write the review. I was quite relieved when we decided that it wouldn’t be me, because rating the game at the time wasn’t a challenge I was looking forward to – and it still isn’t. There wasn’t a grade in this world that would accurately depict both the insane addiction and excitement I felt as well as the myriad of flaws in the core design I had noticed back during the beta period.

I’m one of the untold numbers of gamers whose lives were altered by the original Elite back in the ’80s. The sequels didn’t really hold my interest – mostly due to the rather boring combat – so I didn’t even back David Braben’s new Elite project in Kickstarter. When I heard news about how the design was supposed to be closer to the original with furious dogfights and pew pew lasers I got interested, however, and joined the beta.

Let’s get one thing straight here; I am first and foremost a simulator specialist and I would love to see a “realistic” interplanetary/stellar combat game, but that would look nothing like any Elite game so far. For one thing, there’s no reason for combat starships to have windows. Let’s just say that I’m not against the idea of a realistic space combat simulator, but I strongly believe the decision to go unrealistic with the flight modeling in Elite was absolutely the correct one.

Since July I’ve played totally unhealthy amounts of Elite and it is in many ways one of the most important game releases of 2014 for me, which has prompted me to type down this rather sizable blog post. And it’s not going to be just me trying to figure out all the positive words in the English language, I have what could be described as a rather stormy love-hate relationship with Elite: Dangerous. At least it should make things interesting and this post can be considered a good test: if you have the patience to observe my sanity slowly breaking down all the way to the end you probably have Anaconda-level grinding potential in Elite. So, let’s go.

Elite: Insanely Impressive

David Braben has always been an astronomy nut and his fetish for squeezing as huge a universe into as small a space as possible has reached its climax with Elite: Dangerous. And we are talking about a modern game with gigabytes of data here. The scale is literally astronomical with the entire Milky Way galaxy of about 400 Billion stars in the game and the ones we actually know something about are even supposed to resemble the originals as much as possible.

To share some of my perspective and how I feel about this achievement let’s just say that when I first learned to read, one of my favorite things to do was to read the entry on “Stars and stellar objects” in an ancient (I believe it was written at about the time people figured out how stars actually work) encyclopedia over and over again. Through those pages the wonders of the universe came to life in an enchanting way that was more wondrous than any magic invented by human beings before or since. Later on in my life I would find out that I was no wizard, but that didn’t lessen my awe at the true majesty of the reality we live in and whose mysteries we strive to understand through scientific discovery. My view of science as the greatest human endeavor has been a defining characteristic for me since childhood.

With that said even I think Frontier Development’s choice to go the unrealistic route in space combat was absolutely correct. You can probably imagine how much I also respect and appreciate the detailed modeling of our home galaxy – and when I later suggest it might have been the wrong choice you should understand it’s not because I didn’t consider it a monumental achievement worthy of great praise in itself.

Elite: Dangerous is one of the most beautiful games I have ever seen. Vehicles and structures are modeled nicely, the lighting is done exquisitely and the rest is just raw natural beauty unrivaled by anything created by humans. Usually space games tend to look like someone ate a box of pastels and subsequently vomited on the screen, but not Elite. In Elite the real beauty of the universe is displayed raw and untamed, which makes the man made structures – gorgeous in their own right – stand out and feel so very lonely and vulnerable in the vast stretches of infinity that extends so far beyond human imagination we can barely try to comprehend with the tools at our disposal. There’s of course as always room for improvement in the visualizations and modeling of the universe, but what we have now is already immensely impressive.

Actually flying the starships in the beautifully rendered universe feels rather impressive as well. With flight assist on, the ships maneuver a lot like the atmospheric aircraft of today, making flying in general and especially combat quite interesting. You can get a pretty good feel for what the ships are doing at any given time, but sometimes they seem to do odd things I can’t quite get my head around. It’s not exactly DCS, but not bad either. Of course by turning the assists off you also get the unassisted, Newtonian flight model that seems to work as expected – except of course for the arbitrary speed cap, but there are good gameplay reasons for that so I choose to suspend disbelief as far as the flight modeling goes.

All in all, flying the ships feels nice, but I don’t really have enough experience in high-level combat to estimate if it’s e-sports quality game design all the way. To sum up what it feels like to fly in the Elite universe let’s just say that I’ve done the most repetitive and dull things imaginable for extended periods of time – and still felt that slightest feeling of wonder every time the ship took off and headed for the airlock. Unfortunately, we’ll have to get back to that “most repetitive and dull things imaginable” part later, though, for a number of good reasons.

Elite: Unfinished

Before we head into the abyss let’s make one thing absolutely clear: in early 2015, as I write this post, Elite: Dangerous is in a completely unfinished state. Traditionally it could hardly be considered beta, but today’s world of gaming is a bit different and this kind of thing is becoming the norm. I’ve seen enough game and general software projects to know what it means when patches with massive changes keep coming almost daily just before the official release and even after the game is officially out there are huge changes and bug fixes that touch core gameplay. There are surprisingly few actual bugs, though, and the stability in general is good, but a lot of the design related issues I will touch upon later result from the unfinished state of the game and the seemingly tight schedule the game was released with.

As noted earlier this is the way things work in today’s world of computer gaming and I’m not berating Frontier Development for it. There’s an upside to it in the form of a promise of extensive further development, obvious downsides and last but not least, potentially disastrous consequences due to shortcuts taken during the design and development process. Oh, by the way, in case you’ve been wondering, that’s actually the main motivation behind this blog entry: assessing the future of the game based on analysis of the existing design or lack thereof. This game is very important to me personally and I can’t help but try to figure out where it’s going and speculate on its future.

The important thing to note for now is that we are looking at an unfinished game that is expected to be in continuous development for perhaps years to come.

Elite: Arbitrary

Even more than books, movies or any other forms of narrative, games need to make sense at a very deep level. A large part of the gameplay is figuring out the implicit rules (how the game is actually played) from the explicit rules (describing how the game works). Think about chess for example, it has explicit rules such as “game ends if one side’s king is threatened without possibility of escape” which in turn result in implicit rules such as “it is important to protect one’s king from being cornered”. This is actually the point where all the richness in gameplay is born, the point where chess becomes a wildly successful classic and Elite: Dangerous probably does not, at least for now.

But we’ll get to the conclusions later. For now, let’s just say that one of the biggest problems I currently have with Elite is that it is so full of completely arbitrary things that just don’t seem to make sense. That’s not uncommon for games or any other form of story telling for that matter, but Elite has an unusually high amount of oddities and it’s precisely the kind of game – universe simulator – that so very much needs to make sense at every level. Following is a list of just a few examples through which I’ll try to demonstrate the point.

• A number of randomly spawning AI units appear at Nav Beacons for no reason whatsoever. They don’t do anything at Nav Beacons, in fact it’s not possible to do anything at a Nav Beacon to begin with, their existence is entirely pointless – except for spawning infinite numbers of targets for the player to shoot at.
• Speaking of random spawning; Pirate lords, random traders, bounty hunters and whatnot randomly spawn around the player, hanging around in regular space just sitting there going nowhere (it is not possible to get anywhere in interplanetary space without engaging the frameshift drive).
• Speaking of going nowhere, after entering an economic boom of never before seen richness and wealth and staying there for a while, the systems seem to transition into extremely bloody civil wars. Just… yeah.
• Speaking of civil wars, luckily for the inhabitants of the Elite galaxy, they have no real effect on anything, just an infinitely and randomly spawning group of ships fighting each other endlessly, wasting resources at a level probably unsustainable by the entire inhabited galaxy.
• Speaking of unsustainable losses, let’s do some calculations. A battle worthy Anaconda costs around half a billion credits. To try to make sense of the actual cost we can compare it to the cost of about 55 million tons of gold. I’t important to note that in an interstellar economy the price of resources is probably not comparable to the price of the said resource today. However, gold still seems to be one of the more precious commodities and we are talking about ballpark figures anyway, so when this kind of ships are lost in civil wars, Nav Beacons, Resource Extraction Sites (BTW mining in Elite is a more dangerous hobby than motorbike chainsaw jousting), randomly spawned Unidentified Signal Sources, piracy and last but not least blown to bits by the stations due to parking violations, holy crab that is an incomprehensible amount of waste.
• Speaking of calculations, remember the half a Billion credit Anaconda? Well, imagine this phone call:

“RING RING”

“Blackwater, how may I be of assistance.”

“Hello there, this is the US government! We have a job proposal for you.”

“Ok, let’s hear it then.”

“Well, we have this terrorist group that has approximately half a Billion USD worth of the most modern military equipment manned by highly trained elite (pun intended) forces and they are causing us quite a lot of trouble as you might imagine. We need you to take them out.”

“Ok, that’s quite a task. What kind of compensation are we talking about?”

“$200 000, after and only if you succeed. No costs or losses covered.”

In Elite, you’re fully expected to shout “Hecks yeah!” and eagerly grab the great opportunity with both arms.

• Speaking of opportunities, that is the biggest payout for a combat mission. Like all other careers besides trading itself, combat earns an order of magnitude or worse less credits than trading.
• Speaking of trading, one of the most profitable endeavors I’ve done was to trade “performance” enhancers from one station to a couple of ships parked a planet’s width away from it, with a massive premium in the price. Their combined cargo capacity of 2000t easily handled the millions of tons of stuff people took to them from the station whose supply of 50k+ didn’t budge.
• Speaking of taking stuff from one place to another, the ship’s computer is perfectly capable of calculating the approach trajectory to any station at faster than light speeds, but for some reason it will overshoot every time unless you manually tell it to only use 75% of the throttle it intended to.
• Speaking of stations, why on (or off?) Earth do the massive space stations have a single, smallish entry/exit hole in them, causing constant hazards and traffic jams? I doubt the stations could actually handle the kind of traffic required to supply them and run their day to day operations with the logistics as they are now.
• Speaking of exits, why are there blast deflector plates on the landing pads in front of your ship and why do they need to be lowered before you can take off, but not when you’re landing?
• Speaking of deflectors, why does one sort of a shield give much higher level of protection on a big ship instead of a small one? And vice versa in some cases? Why does it not say anywhere exactly how much protection any given shield provides?

Much of the above was mentioned just for good fun, a lot of more or less important stuff was left out because even I have limits to how much of my brilliant ideas I care to type in one blog post and most of all you would kind of expect things like this even in a good, finished game, let alone one still in the making. Some of the things not, however, and the issue with Elite is that it’s so full of weirdness and inconsistency that it’s becoming an epidemic. And my greatest fear is that there’s a certain, potentially extremely damaging reason behind all the inconsistency.

Elite: Endangered

Someone on the official forums mentioned that Elite: Dangerous feels like a puzzle whose parts just don’t seem to fit. I agree and I also believe that the reason for this is that the pieces were not cut from a single plywood board that had a certain picture on it.

When I look at Elite, I don’t see a coherent design behind it.

The reason it’s not outrageously crazy to expect someone to take out a half a Billion credit Anaconda for 200k credits is because neither the price of the Anaconda nor the reward for its destruction are derived from a coherent design in the background. The reason those few ships can offload infinite amounts of “performance” enhancers produced without limits by the station is because there is no economy; nobody actually produces anything and nobody is dependent on the products, at best their prices are determined by shallow algorithms based on the type and size of the economy. The reason massive amounts of lives and resources can be lost without consequence in the meat grinder we call business as usual in the Elite universe is because nothing is actually lost or produced, in fact besides the human players nobody and nothing exists beyond the short moment they appear as randomly spawned company at one of the momentarily populated instances in the game.

If you were observant enough you may have noticed that I used the rather toxic word “random” above quite a few times. To understand randomness and why I feel it’s so badly abused in Elite, let’s think about what randomness is in reality and in game design. First of all, there’s a big difference between something that is random and something that merely appears random. In the real world you actually have to go to quantum physics to find true randomness (or so we/I believe at the moment), everything else is a result of physical laws – we just don’t know all the variables and even if we did the interactions would form such a complex web of cause-effect relationships that we still wouldn’t be able to predict the lottery numbers or even which number the ball will land on in a game of roulette played for example a hundred years from now. Or if roulette is still a thing at the time or if the humankind still exists.

However, although the future will never be very predictable with human resources, being based on laws of physics rather than randomness has a massive upside. As I mentioned earlier I’m a massive fan of all things science and all of that is based on the fact that the world is based on coherent rules we can try to figure out. It is possible to make observations, figure out what’s behind them and make predictions based on the conclusions. During the last few years this process has brought us planes, trains, automobiles, computers, Elite, Frontier: First Encounters and Elite: Dangerous. Among other more or less useful things. If everything was just pure randomness, it would be impossible to progress beyond the starting point.

Games are a bit similar – and although randomness does play a role in many games its role needs to be considered very, very carefully. Games like chess have no randomness in them at all. Chess is actually a perfect information game, as all the participants have full knowledge of all the rules and events of the game. Randomness has a major role in many successful games, such as for example a famous imperfect information game, poker. Again there is an important consideration regarding the nature of randomness in imperfect information games, however, since instead of being simply random they usually – like poker, Elite: Dangerous or life itself – rely on managing probabilities. Hidden information, even when based on random elements, can provide extremely compelling gameplay elements, but pure randomness has an element of danger to it that threatens to take meaningfulness away from the game.

And that’s precisely the problem I have with the design or lack thereof in Elite: Dangerous at the moment.

Everything in the game can be summed up in a few numbers. First of all, there’s the player’s bank account, which is obviously quite significant for the game’s progress. Second, there are (rather meaningless) figures for the player’s relations to the major and minor factions in the game. The factions also have numbers for their influence within a system that result in a few different outcomes – for example the already mentioned civil war that seems to happen when one faction has completely taken over a system. I don’t know if that makes any sense, but that’s precisely the point: things that are hand-made by designers instead of being the result of higher level rules may very well not make any sense whatsoever in the context of the game.

The main point here is, however, the extremely basic level of interaction between the player and the universe as modeled in Elite. Everything that happens has an effect on a few abstract numbers and even those are fairly meaningless. The last number mentioned – the one about faction influences – kind of affects what happens in the universe, but then again it’s completely out of the scope of what any single pilot can hope to really influence. The level of interaction between a single player and the universe in Elite: Dangerous is ridiculously shallow.

In a game of chess, the first move you make determines the course of the rest of the game, just like each consecutive move after that. The end result is a game that has compelled people for millennia, because although games are somewhat predictable on a high level, there’s incredible amount of room for variation. Then there are games like go that are even simpler to describe than chess but result in so much emergent complexity that not even supercomputers can handle a skilled human opponent.

Elite: Dangerous on the contrary is comprised of short, simple and thoroughly inconsequential loops of gameplay that always immediately return to status quo, only affecting the aforementioned numbers in a relatively insignificant way. The random Unidentified Signal Source events are spawned completely randomly and only exist if a player actually visits them. After the player leaves everything is as if the events didn’t exist (because they didn’t, actually) except for the few numbers we already talked about. The same goes for every single mission and instance type in the game. Every single type of interaction between the game and the player at the level where a single person could actually have significance is reduced to completely meaningless and repetitive grind. Well perhaps aside from exploring, which kind of works the way it should, since you wouldn’t expect very high levels of interactivity from stars or planets in reality, either.

What this means in practical gameplay terms? Well, let’s take for example my fully kitted combat Python that kicks arse and takes names (something between two and three thousand so far, just imagine the loss of life and resources by my hand alone!) like nobody’s business. Yet there’s nothing of any significance I can do in the Elite universe. I can’t even really make money with it unless I load it up with stuff I take from A to B. I’m literally a god among insects among the AI population, but the game’s interactivity offers me no possibilities whatsoever to even turn that kind of power into wealth. I could wipe out small factions, take all their riches, blackmail stations into submission – except I can’t, because all of those entities are represented only by infinitely spawning, inconsequential randomness.

I believe this all boils down to the fact that the design is actually ad hoc to a very high degree. There is no big all-encompassing design or ruleset you could derive smaller gameplay details from, all parts of the game are basically separate and unconnected. From the bounty for killing an elite Anaconda to how much protection a class 6 shield provides to a Python compared to for example an Imperial Clipper to how much a trader seeking “luxuries” is willing to pay for the commodity are singular hand-made decisions without dependencies to other features. What this means is that there is no intricate design for players to discover and more importantly there are no intricate chains of cause-effect relationships to enrich the gameplay beyond what a human being can script in a few moments. Writing every detail and event by hand will seriously limit the amount of variety and gameplay Elite: Dangerous can ever hope to provide, especially since it has to work for a million players at the same time – none of who can become more important than any other, which limits the possibilities even compared to other entirely hand written games.

At times like this when it may seem that I’m slamming the game design rather heavy handedly (perhaps because I am), it’s important to understand that I’ve played the game a lot and actually enjoyed it a least some of the time. I’m not writing about a total lemon here, I feel strongly that Elite: Dangerous is a solid effort with some serious flaws in it, but also immense untapped potential. What Elite also is is a multiplayer game, but unfortunately that side is kind of problematic as well.

Elite: Grievous

There are three main modes of gameplay in Elite: Dangerous; open play, private groups and solo play. Due to my experiences during the beta and my view on the game mechanics I have played the release version mostly in the private PvE group Mobius and my own group with some friends. What I’ve read on the forums has reinforced my view on the subject, but I want everyone to understand that I actually have fairly little recent personal experience with the PvP side of Elite: Dangerous. However, much of the controversy is tied to the general design issues I mentioned earlier and for the purposes of this discussion my experience should be sufficient, so please bear with me for a while as I explain my views.

One of the main topics in the discussion forums for months now has been so called “griefing”, which basically means trolling and generally spoiling the game experience for other users by abusing game mechanics or otherwise acting purposefully in ways that are detrimental to other users’ game experience. This occurs in pretty much all online games and Elite: Dangerous is no exception. For example in World of Tanks you can grief people by shooting and blocking friendlies, revealing their positions or blocking their shots. In general it’s not so easy to find ways to actually grief your opponents, though, because making life as difficult as possible for them is kind of the name of the game to begin with. I’m not counting verbal abuse as griefing here, because I consider it a separate issue entirely.

I find Elite somewhat problematic when it comes to griefing due to the major flaw I mentioned earlier; lack of coherent design. In fact I don’t think there’s currently a functional framework for PvP contact at all. Since the AI are only random generated, utterly meaningless and inconsequential target drones, there has been no need to create a coherent framework for participants of the game world to interact on a more equal footing. Except of course if you factor in the multiplayer part of it all, which you kind of should in a game like this.

To me it really seems that there is no reason whatsoever to engage in player on player action in Elite if you look at it purely from game design perspective. It is by far not the most profitable endeavor (trading is), in fact it might be the least profitable one of all. Especially with the higher-end ships the risks are tremendous whereas the rewards are minimal, in fact due to out of this world operating costs simply interdicting someone can mean that both parties will suffer a loss no matter how the conflict ends. In fact I would estimate that player to player conflicts are a pretty major money sink where on average everyone loses.

Except of course if you factor in the thrill and fun of fighting an actually challenging opponent – who also happens to be a real human being instead of an utterly meaningless temporarily random generated piece of code. That’s of course not griefing, how could it be when both parties want to have a good scrap against a worthy opponent. However, how could it not be griefing, when the other party is not as willing a participant? A trader peacefully running his route has all the right in the world to wonder why someone would pirate or outright murder him when gameplay-wise it would make more sense for his opponent to simply run a trade route like him and make massively more money – or at least engage the meaningless AIs or at least willing opponents. To me this kind of disparity of viewpoints is a direct result of a non-existent framework that would provide context for such encounters  so that everyone could at least assess each others motives and achieve an understanding of when and where PvP are likely or even desired. In my opinion they should also strive for a situation where PvP encounters are desirable and profitable – or at least not detrimental to all involved. An indication of just how bass-ackwards the system is right now is the way highly populated core systems are the most dangerous places of all – if you want to be safe you have to go to the outer rims of civilization. And that makes just no sense whatsoever.

If done right, good fights between players can be the highlight of any game. Multiplayer gaming has shown its immense potential during the early 21st century and it’s not wrong to look for depth and longevity for an Elite game in that direction. In this case I even believe that both multi- and singleplayer enthusiasts would benefit from the exact same thing: a framework that could govern interactions between more equal participants, including contacts between players as well as AI characters modeled as capable and important actors within the context of the general gameplay. However, there are potentially fatal problems in the way.

Elite: The Future

During the last days of the beta stage Frontier Development dropped a bit of a bomb on the community when they announced that there would be no offline mode. A number of people got understandably upset when a promised feature was being cut – and I can understand them, I’ve had internet connection issues and it’s annoying not to be able to play something when your connection is out. I was actually quite upset, too, but not for that particular reason.

Elite:Dangerous has a peculiar technical approach to multiplayer, with everything being run on the players’ own computers. This leads to multiple simultaneous instances of for example our home system, Sol. Unlike for example in Eve online there could be a million people in Sol at the same time but in parallel universes so that only a few of them could see each other. However, all of these parallel realities are connected. Remember those weird numbers I mentioned earlier? They are the connecting factor as everyone’s activities whether or not they are in one of the parallel multiplayer universes or just playing alone in solo mode affect the same abstract numbers. These numbers may well be the lowest possible common denominator between all the different instances. And you know that when someone mentions the word lowest common denominator in the context of computer gaming  it’s not going to be good. And it’s not good this time either, since those numbers being the lowest common denominator might well mean that

everything below that level will forever be by technical necessity meaningless, random, inconsequential crab.

Whether or not you agree with my assessment of the situation you at least probably understand why I’m so worried for the game I think has so much promise of greatness in it. If the scale of the universe is too big for it or if the joint multiplayer universe doesn’t allow for Space Rangers 2 -style AI actor modeling I will gladly switch into a smaller scale galaxy and limited multiplayer in a heartbeat to get a proper Space Rangers 2 from the cockpit -experience. If you don’t know what I’m talking about check out the game, it’s somewhat uneven and a rather weird mix of features, but it does the living universe part just brilliantly. It’s been a while since I last played it, but I think all the AI units are actually modeled doing their stuff in-world and if there’s a bounty on a pirate lord there’s also an actual pirate lord who has done bad things to deserve the bounty out there. And taking him out will actually be an epic achievement, it will mean that an actual AI being is now out of the game and the universe will surely take note of your monumental achievement.

It’s usually fruitless to fret about what a game could have been like, but in this case I feel this is all about what really should have been and hopefully of what could still be. I mean, games like Eve Online, the X series and Space Rangers 2 exist and they have shown what cause-effect modeling or supply chains -based economy in a virtual universe is at least supposed to be like in the early 21st century.

So, when Frontier Development announced that there would be no offline mode because they had this wonderful online dependent shared universe thing that would be needed to provide all the content I couldn’t help but read it “there’s not going to be any kind of proper low-level interactions modeling, we will have to inject all relevant events by hand for all eternity”. And lo and behold, they announced the soon to arrive “community events”, which are, unsurprisingly, the exact same kind of grind based on percentage figures the game consists of today. They are a bit odd, though, since the “performance” enhancer rallies of recent weeks have proven beyond doubt that the game is kind of already capable of generating community-based events like this all by itself.

I would really, really like to end on a high note, but the best I can do right now (early 2015) is to leave everything open, grasping at the fact that the game is still very much in development. I would hope that talks about features such as planet landings (which is probably a huge resource hog and adds little to the game) and especially first person mode (which will bring nothing to the game in its current state) will soon give way to discussions about deep interactivity and proper gameplay mechanisms.

I already explained why I fear this will not be the case, but I’m not giving up all hope quite yet. The Wings update (that is supposed to enable co-operative play) is coming and probably being designed right now. This design process should uproot a lot of the issues with the current multiplayer framework (or the lack thereof), forcing another look at how things should actually work. There’s no question that Elite: Dangerous absolutely needs a robust framework for multiplayer interactions and as I mentioned before, if you plan a way for equals to interact in reasonable ways within the context of the game that should also open up ways for the AI to assume roles beyond simply being target practice drones.

Let’s try to end on a high note despite everything. After all, now that I’m done writing this post I’m about to hop into a starship and go shoot some baddies. At least we know that the team did have great ideas during the development and although not that much of it all made it to the release version, perhaps the designs aren’t buried away forever. Of course things like an actual design framework isn’t something you just inject into a game, since everything depends on it and is derived from it … aw crab, guys, I promise I tried, I really did. Let’s give it one more go:

I firmly believe that Elite: Dangerous is a true love child for Frontier Developments and they will do whatever is in their power to create the game they envisioned years back when they begun the effort. The development experience will help them in their endeavor and they will emerge wiser and more experienced from every setback and painful compromise.

There, that’s much better, right?