A Very Particular Set Of Skills

By Sam Quirke, 5 months ago
10,000 BCE.

The world of Oros is a brutal, unforgiving place. Mauled by sabre-toothed tigers, bashed by rampaging cannibals... Takkar's life is simply a case of kill, or be killed. As he wrenches an arrow from his shoulder and wipes the blood from his club, Takkar sees a mysterious symbol right before his eyes:


Takkar must now face the most fundamental question in all prehistory...

... should he spend his point on 2-for-1 arrows, or save up for a cool-down reduction for his owl?

Me Takkar! Me warrior! Me collect enough shiny XP rocks to see yellow plants in super-vision!Me Takkar! Me warrior! Me collect enough shiny XP rocks to see yellow plants in super-vision!

Skill trees have been around in video games for a long time, taking an important role in many RPGs and strategy titles. Their primary purpose, one can argue, is to put the development of the hero or their team in the hands of the player. This makes perfect sense in an RPG - we play them with certain expectations. There will be a heck of a lot of choices to make, character elements to customise... many menus will be involved. But that's OK -- it's part of the experience.

In the best of these games, skill trees are crafted with as much care as any other element. The complexity of the skill progression matches complexity of the universe it inhabits. The player knows their choices will have an impact -- on gameplay, NPC interactions or even the direction of the narrative. With all that going on, spending a lot of time in a skills menu is part of the experience.

So why does a game that explores the visceral brutality of pre-history have such a distracting, bulky skills matrix... when we can neither customise the hero nor the choices he makes?

Mirror's Edge Catalyst is not even out yet, and has already come under some fire for locking up skills that were available in the first game. The franchise's unique selling point is in the purity and freedom of sprinting across skyscrapers at breakneck speed -- so why dilute that energy with a glorified flowchart?

The recent over-abundance of skill trees might seem a strange thing to get upset about. After all, I believe Far Cry Primal is a decent game (despite what I have to say in the next few paragraphs), and skill trees are not new to the series. Equally, a skills matrix developed properly can be an enjoyable part of the experience. But like many mechanics traditionally limited to RPGs, skill trees are contributing to that exhausted feeling I'm getting whenever I try and get stuck into yet another open-world adventure.

To explore this idea further, here's my manifesto for building a decent skill tree.

1. Don't build one.

'Is this adding any value to the experience?' If the answer is no, then get rid of it.

If the answer starts with, 'It forces the player to...' stop right there. Locking up a skill in a tree is a bad way to make a player do anything. Skills, implemented during gameplay, should speak for themselves. The developers behind Mirror's Edge Catalyst have argued that the skill tree will encourage players to use skills that were often overlooked in the first game. As far as I'm concerned, if a skill wasn't used much last time, it's because it wasn't implemented very well, or is simply not as necessary as the designers have assumed. Locking it behind an arbitrary skills tree isn't going to suddenly make everyone want to use it.

Scrapping the skill tree isn't the same as not having any skills in your game at all, or that there shouldn't be limitations and progressions. But if your game is not primarily an RPG, and especially if you want your game to be immersive, there are other ways to implement those progressions without shunting another tab onto our Start menu.

To use Primal as an example again: there are several skills which unlock the ability to tame more beasts. Each require both a certain number of beasts to already be tamed, and a bunch of skill points. Why do we have this extra 'points' requirement on top of the work we've already put into reaching the quantity target? Scrap that, and let us know after taming the right number of beasts that we can take on a different one. Maybe Takkar can mutter, 'Guys, I totally feel like I could tame a bear right now...'

Similar examples can be found in most non-RPGs with skill trees - it's a missed opportunity to integrate skill acquisition into gameplay. Skills can be obtainable when completing certain missions, which also act as a tutorial. This actually happens - only twice - during Primal, when you are introduced to the owl and the grappling hook. Can't all these 'specialist' villagers mooching around your homestead actively train you in the rest of the skill set, rather than just being a static face next to a bunch of icons? If that's too much work, you could at least merge several skills into the already exhaustive crafting and village upgrades systems, reducing three menus down to two.

Let's assume our fictional developer is still insisting that the game needs a skill tree. They've heard RPG elements are justification for copy-pasting gameplay over a large world. They can technically say the game lasts 90 hours and is worth $60. This excites them.

Fine. What can we salvage?

2. Skill trees should be pruned to the bare minimum.

Too many games have too many skills. It's like developers are worried that gamers are going to turn their nose up at another pointless menu system, so they flood the page with equally pointless perks and ambiguous upgrades just to trick us into thinking it's worthwhile. In fact, if I must see a skill tree, I'd like to see a simple one with obvious progressions. In this regard, the Mirror's Edge Catalyst skill tree doesn't look too irritating. Totally unnecessary, of course, but at least not a headache to look at.

Mirror's Edge Catalyst skill tree, from the recent Beta.Mirror's Edge Catalyst skill tree, from the recent Beta.

Making up skills because you want every branch to be even, or the whole tree to be in the shape of a smiley-face, is not a good enough reason to add it in. Equally, if you've got a cool idea but you can't find a logical space to shove it in your skill tree, find somewhere to introduce these in the gameplay instead. The Witcher 2: Assassins of Kings has a big skill tree befitting its RPG nature, but it also has some great tongue-in-cheek skills gained while playing -- like fire resistance after standing in too many bonfires like a fool, or magic resistance for turning down hot-tub tomfoolery with a witch. Putting bonus perks in the game itself not only subtly encourages us to play around with the environment and decision points, it leaves the Skill Tree itself to be well-balanced and meaningful. Speaking of which...

3. Make the skill tree mean something. Choices should have consequences.

Tacked-on skill trees that can be completely filled out in one playthrough do nothing except stall the player experience. If it's perfectly possible to get all the skills eventually, the only input the player has is grinding XP, often by hunting down collectibles. Like shiny rocks.

Hah, hairy elephant! You were goring me two seconds ago, but I used shiny rock XP to buy your loyalty!Hah, hairy elephant! You were goring me two seconds ago, but I used shiny rock XP to buy your loyalty!

If you're going to borrow this mechanic from RPGs, at least act more like an RPG in the execution. A skill tree should be impossible to fully complete in a single playthrough, because the only reason a skill tree should be there in the first place is if picking one skill over another will make a difference substantial enough to want to play through again to see the consequences. This can either be in gameplay or narrative. The Witcher 2's fairly slow levelling rate means you pretty much have to pick between swordsmanship, alchemy or magic -- or risk being a mediocre jack-of-all-trades.

The only consequence of buying one skill or another in Primal, meanwhile, is just how much more skull-smashing/rock-picking you'll have to do before you can get the other one. Skill points become just another collectible, dressed up as something more meaningful.

The worthless grind can turn a great game into a merely OK one (or an average game into a terrible one), because immersion wears off when you're running around after XP. Not only is there no incentive to start the game again, but chances are we'll be too bored to bother.

4. Make your skill tree something to be proud of. Make it fun, or at least interesting to look at.

At least dress the blasted thing up. Make it look like part of the game. Inject it with the same tone, humour and aesthetic as the rest of it, and make it a menu page we quite like visiting. There are a handful of great examples out there.

Skyrim - colourful, clever design in keeping with the lore.Skyrim - colourful, clever design in keeping with the lore.

Borderlands 2 - witty, lean and purposeful.Borderlands 2 - witty, lean and purposeful.

Fallout 4 - Possibly too much going on here, but it's fun, functional and animated. Fallout 4 - Possibly too much going on here, but it's fun, functional and animated.

Far Cry Primal - It kind of looks like cave paintings... but mainly it looks like a brown grid with brown things on it.Far Cry Primal - It kind of looks like cave paintings... but mainly it looks like a brown grid with brown things on it.

Alas, our skill tree developer has put their foot down. They want a bloated, boring, consequence-less tree with exactly 4x4 branches so the menu looks neat. Surely they can listen to one last plea for sanity?

5. If nothing else, please don't punish players with an 'all skills' achievement so they have to run around post-game pick--

Far Cry PrimalExpert WenjaThe Expert Wenja achievement in Far Cry Primal worth 111 pointsLearn all skills.

… Ah. I see.

Come along then, Takkar. Let's go get you some more shiny rocks.
Sam Quirke
Written by Sam Quirke
Sam has been gaming long enough to know the pain of a failed DOS install on the fifth floppy disc. As well as writing news, he also looks after our Community Interviews. Send him a PM if you are interested!