Retrospective: Deus Ex, Part 4

To Augment or Not to Augment…

System Shock (1994) now there was a game that I remember with fondness from my teens. Forget Doom (1993) that was crap as far as I was concerned – let’s face it, the AI is hardly “stellar” and gameplay “diverse.” And yes, I know that System Shock was hardly perfect itself; the FPS mechanics were awkward to say the least – Doom had better mechanics and that is saying something.

But along with a strong sci-fi cyberpunk story driving the action of the game through it’s survival-horror atmosphere that is the isolated Citadel space-station. System Shock, was also a title that did not rely on any dated RPG stats for its gameplay – well, none that were visible and exposed directly to the player at least. Rather it had objects.

Want an ability? You found a neural chip for it. Wanted to be better at that ability? You find a better version of that chip. You didn’t have to progress from 1 through to 5, either. You could based on the choice you made in a relatively nonlinear game world, go from 1, to 2, then stumble upon – with great difficulty and perseverance mind you – 5. All the rest would then be useless without it feeling like the game had become too easy because of it; it was your choice to do so after all and it would have been a hard fought reward as well.

It was an object-based action RPG system; no stat leveling with experience points required.

In Invisible War it appears they’ve tried to do something similar in that they removed the RPG systems, or at least, made these far less “exposed” to the player. But they’ve taken a step too far. This is evident in their “augment-biomod” system, which is analogous to System Shock’s object-based “neural-chip” system – these are “powers,” which RPGs generally have in some shape or form, such as spells, potions and enchanted items.

First off, I played Invisible War from start through to end like I do with any game I play, on the normal difficulty setting. To me this represents the most balanced version of the game and how the developers intended the game to be experienced. Anything else would either be a watered down (easy) version or an artificially contrived difficult (hard) version.

For the most part, I found the game to be too easy via normal and I doubt that playing on “hard” would have helped in making much difference; in that the reason it was so easy, was to do with the game systems utilised and it’s “basis” in the game world: biomod canisters.

Having played the majority of quests (roughly 20 hours worth) these are the number of biomod canisters that I had managed to find with ease that I had left after making my mod choices, upgrading them to full, and, sticking with them until the end of the game:

• Augment canisters: 11
• Blackmarket canisters: 9
• Grand total canisters: 20

Number of times you can switch and upgrade a mod: 6.6.

That’s at least 4.6 times too many. That’s not just switching mods to the lowest level, which would then make it 20 times that you could do this. But rather, switching and upgrading an augment to its fullest strength. You can do this so many times to make the importance of choice within the game seem next to irrelevant. This contributes to the game being too easy by providing too much “choice” to the player.

Who exactly did they imagine would be playing this game?!

Deus Ex with its augmentation – power / ability system – required you to make a choice. You got a specific augmentation canister that had one of two choices for a specific area like the arm. Once you chose, that was it. You couldn’t just replace that area with another augmentation. You’d made your choice and would have to live with the advantages and disadvantages of that choice.

Now, you don’t have to be that anal about it. After all, if they are “nanite” augmentations, then, I don’t see why you wouldn’t be able to make one or two changes – switch your arm augment from strength (carry-weight) to power (melee and throwing). The premise supports you being a tad bit flexible. However Invisible War allows far too much flexibility which merely undermines gameplay.

If you’ve chosen a style of play, then you may gripe but you will want to stick with it and make it work and overcome the game in that fashion. Being able to make one or two changes, that are still fitting to your style of play, allowing you to progress without making the challenges irrelevant from ease. I think can be a good thing in moderation.

Essentially, they took it too far. Like abolishing the weapon skill system based on experience points, because, you no longer have experience points – it’s all object-based. Forgetting momentarily that an element of this is provided by the weapon-mod system – which was also a part of Deus Ex – they could have provided a more progressive alternative to go with their progressive systems. One that requires the player to invest more directly in that skill, and feel, like they’ve made a choice in doing so – by not allowing them then to use the “experience” to develop unrelated skills.

Personally, I think that NOT having to award skill points to upgrade (weapon) skills, is a good move, in that it’s a dated RPG mechanic that there are alternatives for. It’s quite simple really in how it works, in that you start off at the most basic level for every skill, unless, you’ve picked a class (assassin) that has an affinity for that skill (sneak) – if you have character “classes” that is. Then the more you used that skill, the better you will get at it – the more “experience” points will be awarded directly to it – and eventually these skill “ranks” and their benefits are unlocked to the player.

Logical isn’t it?

“Sandbox” titles: GTA San Andreas (left), The Godfather (middle), The Warriors (right)

Though utilised in Morrowind for some of its skills, the example that sticks in mind from a title around when IW was released is action adventure (open-world / sandbox) GTA: San Andreas (2005) – it was also the same with another action adventure title: The Godfather (2006); and also to a lesser extent in the underrated beat ’em up sandbox The Warriors (1995) for some of its player stats.

The more you utilised a weapon in the game, the higher in experience you’d then gain with that weapon and at set experience point intervals – like 25, 50, 75 and 100 points; or alternatively, every 100 kills with that weapon sees a new rank “awarded” and the weapon kill-meter set back to 0 – you’d unlock a new skill level with that particular weapon, like graduating from Rookie to Footsolider.

A progressive system like this, found in sandbox games, would be a better alternative for an action (sandbox) RPG title. However, you could question as to whether in a firstperson shooter, or even thirdperson shooter, whether you’d need to implement such a system at all. The more the player plays the game and gets used to how the mechanics (interface) of a particular weapon handles, they’ll naturally get better at it regardless.

Still when you have to simulate weapon “physics” like how bullets will spread over distance, and how a gun will kick back and up from recoil – an SMG might be best used in tight controlled burst for that reason. Then you’ve got some good “excuses” to implement a skill system where the higher you get in rank with that weapon – “newbie” to “hitman” or whatever you’ve gone with – the more improved various weapon stats (mechanics) involved with how it performs, will become for the player.

As no matter how good the “interfaces” that we create for players to interact with the game world are, they’ll never be as rich an experience as the “real” thing – at least not with today’s technology, not even with the: Wii, Move and Kinnect.

You have to give players some concessions so that they can then interact with the game along the lines of the intended experience without the limitations of the interface getting in the way. And an extension of this is that you sometimes have to provide them with “concessions” so that “reality” doesn’t get in the way; why most games let you take a couple of bullets to the head before game death kicks in.

These “mechanics” allow for elements to be “modeled” in order to create a more immersive experience.

So in this way a weapon-skill system allows a player whilst playing the game, having already mastered the mechanics as best they can. Still feel like they are “progressing” in skill, advancing in ability, by providing them with “skill” levels that will then alter various mechanics so as to aid them as they’re pitted against tougher game challenges. With each skill rank, the tighter their shots will become, the less effect running-and-shooting will have on shot accuracy, the better control of the recoil they’ll have and the quicker they’ll be able to reload a weapon due to familiarity with it.

That one in particular is a good example, in that we usually press a button to reload and have no control directly as to how quickly the whole operation takes to execute. A skill-system changes this as perhaps does a weapon mod system, but, perhaps in another way – an extended clip increases ammo reducing the amount of times you need to reload, for example, but does not alter how quickly you do reload a weapon.

This to me is a far more intuitive and logical system, as well as, one that creates a more rich gaming experience for the player – especially when it works hand-in-hand with a weapon-mod system in creating a robust action RPG experience. Also, I’ve used weapon skills as an example, but, it’s also applicable to all skills such as lockpicking and hacking, and, even attributes like stamina (sprinting) and agility (jump height and falling distance without damage).

And if you want to maintain a method to “level-up” a skill without actively using it – experience grinding. Morrowind and other RPGs allow you to read books and the like that will give you a once-off increase to a skill – another reward and incentive you can provide players with for completing objectives. It doesn’t have to be oldskool in the form of a book as well; with a neural interface, you could just as easily find (or buy via the blackmarket) a datacube – or whatever, like cracking a military skill-bank or “training” via a commercial terminal by paying the required fee – so as to download the knowledge from and gain an immediate “rank” in skill.

No direct “leveling” of stats with points required of players; all achieved either via a passive system that tracks the player’s actions or via the player using an object. All far more intuitive for it whilst creating (not undermining) player choice.

Universally a Bad Idea…

Something that’s always bothered me is that System Shock offered an object-based RPG system yet the developers of the sequel saw fit to implement a largely traditional (dated) RPG system instead. I know, it boggles the mind… that is until I began to compare Deus Ex with Invisible War.

Warren Spector is the guy most mentioned when speaking of System Shock and Deus Ex. It seems logical to expect that he’d carry on what was started in System Shock over to Deus Ex, but instead, the team implemented various RPG staples like a skills system based on experience points. Then it seems that with Invisible War they tried to do something similar as with the original System Shock. But, it failed.

There’s a reason why System Shock 2 also implemented the dated RPG mechanics, it’s because they provide something that the technology (like the AI) just wasn’t able to on its own in order to create the compelling action RPG gameplay players expected – the “sandbox” techniques just weren’t developed enough yet; EDIT: that, and Irrational Games didn’t have the luxury in way of time and budget to develop such systems.

I believe (or like to) that the reason for this boils down to the need to effectively create “choice.” Too much choice is a bad thing, and a system like the one in System Shock, meant that areas of the game could potentially suffer.

When you can “hot-swap” abilities with ease this causes issues of balance – making the game seem too easy and somewhat generic. It wasn’t obvious at the time because of all the novel 3D technology and the originality of the concept that was System Shock – also, you didn’t miss out on any powers in favour of others; by the end of the game you had all available abilities and all developed to full.

For the later released Deus Ex it had to implement older RPG systems, otherwise, it’s limitations would be far too “obvious” to the player. As in the case of Invisible War in that the technology was nothing that a number of other titles did just as well, if not far better.

And as far as story, concept or premise – however you want to look at it – of Invisible War, it had been done before by Deus Ex. Without these to help “cover-up” the ugly parts, and distract the player from the reality that is the generic playing AI and maligned RPG systems, that provided too much choice and therefore undermining  player choice and eroding their immersion into the game world. Well, you need something to provide that variation in player experience, something to artificially create the gameplay, to create the illusion of choice for the player.

Why Deus Ex and System Shock 2 implemented dated RPG skill systems, and where Invisible War, should have provided a progressive alternative – better suited to the action “sandbox” RPG expectations of players – along with better versions of their weapon-mod and augmentation systems. This then would have created the illusion of choice by providing clearly defined (limited) choices to the player to chose from.

However, the lack of a weapon-skill system is not the worst of it – especially as an object-based weapon-mod system when done well, largely makes up for it. Though you can gripe about a lot of things to do with their weapons systems, like the fact that weapons seemed to have NO reload – it effectively has an “infinity” clip. Invisible War actually had a lot of features in way of game mechanics that I quite liked – primary and secondary fire modes for weapons so that the SMG can launch a flash-bang grenade, for example. Nothing that hasn’t been done before, but still, I appreciate the thought. But the one contentious element, the immersion deal-breaker, universal ammo, was simply a bad idea.

I can see to a degree how by providing “universal” ammo it can on the surface appear to be liberating in that the player can then choose which weapon, which tool to use, to overcome the obstacle they face. However, I see how by doing so how you actually undermine tactical play and diversification of play.

You’ve got an SMG, it’s out of ammo, so you pull out your silent toxic dart gun to stealthily take out key enemies so that you can complete your mission. You’ve only got one rocket left, do you use it against that bot in your way, or, do you find another way? You decide it’s not worth the investment as rockets are hard to come by, better saved for a tougher, more worthy target. You try sneaking past instead or setting up some mines that you’ve picked up but never found a good reason to use, that is until now.

With universal ammo, every obstacle is pretty much taken care of with your “favourite” hammer. Every challenge runs the risk of playing exactly the same as the last one because of it. And there is no spur of the moment “oh shit, I’ve run out of SMG ammo, what am I going to do” situations when you have bucket loads of ammo to spare, the kind of situations that force the player to be inventive, creative with what they do have – “ah, but I do have a gas grenade and baton… and that guard patrolling at the far end will definitely have SMG ammo and, can be easily ambushed when he wanders into that warren of cargo containers.”

Things become generic quickly with universal ammo as the “freedom” it provides the player in way of choice doesn’t provide the player with any incentive to chose to play any differently from their preferred style of play, with their weapon of “choice,” as they easily progress from the promising start through to the bland end.

What’s more, I don’t think that the developers (and/or publisher) were being very truthful with themselves let alone others. Was this decision made so as to provide greater choice to the player, or, so as to make lives easier for themselves? If you have multiple ammo types then that is more “balancing,” more playtesting that needs to take place. More decisions to be made and more stress from having to make them – like knowing you have to make major changes that will affect the schedule which could slip and require more funding to get the title out the door.

It’s like the Invisible War mission designs having only one objective with shallow “options” of how to go about completing that objective. A mission with multiple objectives, each with thought out balanced options of how to go about completing them… is a lot of work, a lot of stress. So, why not just leave these “decisions” up to the player and reduce your work – since they’re never happy anyway – and leave it up to the player to decide how they want to play the game…

Hmm, because the player ultimately doesn’t know what they want, only thinks they do, and ultimately, they may be knowledgeable fanboys but are hardly experienced game designers!You need to shape the experience for them, whether they gripe about it or not.

Anyway, not to be too critical of the developers in that this could just of easily been due to the pressures of a publisher wishing to cash in on a successful title by taking “shortcuts” to get a sequel out – relying on the goodwill of the previous title to bare the burden of selling the next. Like the 80s film industry that created poorer imitations of successful films, sometimes, a whole series of progressively worse sequels in order to cash in and bleed a license dry. Something that film publishers have and haven’t learnt from – with their contentious license “reboots” – and perhaps you can say the same with game publishers.

Getting back on point, in order to make “balancing” a bit easier and provide greater flexibility of choice to the player without undermining it. You could have implemented a set of undifferentiated ammo for the level designers to place along with predefined (eg shotgun) ammo objects. With the predefined ammo making up 70-80% of the ammo placed in a level, with the 20-30% remaining being the undifferentiated kind. These undifferentiated, or “universal” ammo objects – that would come in 3 flavours: small, medium and large – would then “randomize” into predefined varieties when the player starts that level.

You can take this a bit further in that if you were keeping track of various player statistics you could place greater chance (emphasis) on certain ammo types because of the player’s preferred play style and weapon. So, that medium universal ammo object will “randomise” into a clip of tranq darts as the player has a predilection to play as a stealthy character that uses the dart gun to overcome enemies in tight situations.

In this way you guarantee that at least some of the ammo found in a level will actually be useful to the player, especially in critical situations – you could also extend this to other consumables, like those to do with health and performance.

However, I do use “random” with some apprehension in that I do not mean it as most developers, especially the coder variety, take it in meaning. In that I see that most cases of “random” (procedural) systems as being nothing more than the lazy way of developing games, and also putting too much emphasis, on the coding side of development. Usually these systems result in generic playing games, that are either too easy or too hard, rarely just right.

So, I warn thee to take careful consideration prior to implementing random procedural systems. It may make some things easier, for you as the developer, but it could also end up leading to a universal fubar… Make sure to read: part 5!