Gamasutra: The Art & Business of Making Gamesspacer
Untold Riches: The Intricate Platforming of Tomb Raider
View All     RSS
October 21, 2014
arrowPress Releases
October 21, 2014
PR Newswire
View All

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

Untold Riches: The Intricate Platforming of Tomb Raider

March 1, 2013 Article Start Page 1 of 3 Next

What game design gems lurk in the unexplored nooks and crannies of the original 1996 Tomb Raider? With the franchise now rebooted, designer Hamish Todd takes a close look at the series' first game, and the possibilities its constrained platforming allowed.

The classic Tomb Raider games are some of the most commercially successful of all time. They were formative for many of today's gamers and game designers. They've become most famous for their varied texture work, and for the iconic Lara Croft, with her associated sex appeal. But, at the outset of the franchise, playing Tomb Raider wasn't about getting to know Lara Croft's character. It was about searching, shooting, and platforming -- mostly platforming.

I wanted to take a close look at the platforming challenges that classic Tomb Raiders offered us. I looked through their many levels, and I was surprised by what I noticed: all of the most eye-opening things were to be found in secret areas of the 1996 original. In this article I'll describe my more beautiful findings, and ask what makes them so fun.

Natural Teaching

To give you a reminder -- or an introduction! -- to the controls: Lara is a "tank," so with four buttons you turn her left and right, and move forward and backward. She can also jump and grab ledges.

Additionally, Lara can slide down slopes; when sliding, the only thing she can do is try to jump off the slope. This is not in the instruction manual. The first engagement I want to show you is a short challenge from the game's very beginning which introduces jumping from slopes.

Here, the eye is caught by a small, elevated alcove, and a squat, sloped block faced toward it. Getting into that alcove will be the first short-term success of the whole game -- it contains a health pickup.

Try to jump and grab the alcove's ledge, and you'll find it to be just out of your reach. Turn your attention to the sloped block behind you, the only other thing around. You've seen that it's slightly less elevated than the unattainable alcove ledge. You go over and jump up onto it... then you immediately, unexpectedly, slide down it and fall off, plopping you back at the front, facing the wall. This block is a bit of a troll -- it's the first mischievous piece of design of many we'll look at.

You'll be irritated but intrigued by that slide. You try doing it again, and you realize you can jump. After a few tries, you catch the ledge of the alcove, pull up, collect your prize.

This is a motivated, forgiving, uncluttered introduction of a semi-novel move; no textboxes required. A move that can be pretty exciting and, as you'll see, surprisingly deep.

The Water Slide

This is a rather fascinating challenge. Essentially it's about manipulating momentum.

You enter this room on a slide (0). You're moving, and you can't do much about it. There's a secret on the opposite side of the room -- that's what you want. However you can't make it across by jumping or falling -- not from the top slope, which has brought you here.

What you want to do is jump from that lower slope, but it's not clear how you can get to it. The top slide has given you a lot of momentum, which is causing you to fly over the lower slope, straight into the water. There are a few methods for getting to the secret, and all of them require the player to plan around the platforming controls.

Here's one method (1). If you know what's at the end of the slope, you can send Lara down backwards, which allows her to grab the ledge when she falls off. Drop down to the lower slope, slide a little more, backflip at the last minute and you'll get to the secret.

But this is very unreliable. That final jump backwards needs to be bang on the end of the lower slope -- but you're sliding backwards! Knowing when to jump is guesswork. Getting to the secret this way is time-consuming and dull.

Here's an improved method (2). Catch the ledge of the upper slope, then pull back up onto it, then allow yourself to fall off again. Falling from a standing start, you get exactly the right amount of speed to take you directly to the end of the lower slope. It's nice that there's no timing involved in this method...

...but this third one seems the most elegant to me (3). It can be thought of this way: ordinarily when you fall off the top slope, your forward velocity causes you to sail over the lower slope before your downward velocity can get you to land on it, so you land in front of it. Our last method works by increasing your downward velocity while you're above the lower slope.

Bear in mind that as Lara falls, gravity increases her downward velocity. The longer she's been falling, the faster she falls -- this is true whether you've fallen off a ledge or if you're at the top of a jump arc. All this means that if you start a jump that peaks before the fall-off point, then by the time you get to the fall-off point you'll have gained more vertical velocity than if you had just fallen off, so you can get to the lower slope even though your forward velocity hasn't changed.

This strategy can sometimes be used in Canabalt to avoid hitting the sides of buildings...

...and, I think, tells us something about real-world ballistics.

If you're doing a jump like this, you have to know your arc well and make sure it's not going to hit the corner. This is a fascinatingly unusual aim to have while doing a jump. It's also interesting because jumps are normally about changing your position, whereas here they're about changing your velocity.

Smile for the Camera

There is a very sophisticated three-jump sequence in this room. You go from the floor, to the slope on the lower right of this picture, to the slope in the middle, which you immediately spring off to grab the ledge above.

What makes this such a splendid engagement is a huge camera shift that takes place in the jump between the slopes. The camera rotates more than 180 degrees, and the two slopes rotate Lara by 270 degrees. Things move fast in this enclosed space; it's almost as disorientating as being flung through a portal. You have to make a plan.

One problem is that if your second jump is from the lower part of the slope, Lara will just end up banging her head on underside of the ledge. To be jumping from the correct part of the second slope, you have to jump from the correct part of the first slope. Your starting angle and location affect these; to work out what they should be, you must see through the rotation of Lara and the camera.

Note that all this was enabled by Tomb Raider's much-maligned automatic camera. It puts a strict limitation on you, motivating a creative approach. The same thing happens in a couple of the game's block puzzles.

By the way, look back at my picture of the room and notice the diamond-shaped slope in the far right corner. If you were to jump onto that it, you would witness an unabashed glitch in the slope-sliding. I don't think it was an accident that you could access that glitch in this room. By the end of this article I think you'll agree with me.

Article Start Page 1 of 3 Next

Related Jobs

Zindagi Games
Zindagi Games — Camarillo, California, United States

MOBILE Art Director
Treyarch / Activision
Treyarch / Activision — Santa Monica, California, United States

Senior UI Artist (temporary) Treyarch
Treyarch / Activision
Treyarch / Activision — Santa Monica, California, United States

Lead UI Artist
Treyarch / Activision
Treyarch / Activision — Santa Monica, California, United States

UI Artist/Visual Designer (temporary) Treyarch


Phil Maxey
profile image
Such a great game. Apart from all the innovations, it's focus on doing simple things well is what makes it great.

Vincent Hyne
profile image
Fantastic article.

"Rewarding obedience". That's true, that's what those games are about to a great extent.

Muir Freeland
profile image
I really enjoyed this article. I love how it explores limitations as something that breed creativity; this is something that I think about a lot in my personal level design work, and it was great to read theories about how it played into Tomb Raider.

Christian Nutt
profile image
I'm a fan of games that really explore a set of limitations and make new, clever things out of a limited set of tools.

Albertus Agung
profile image
Wow, This is what I want to read. i cant believe you wrote this article clearly. I love 'Natural Teaching' just like what I did back there playing any kind of old games.

Gaetan Brisson
profile image
Tomb Raider was a niche product that happened to become very popular for all the wrong reasons and for the worst fate as far as we, the original players, were concerned. For us, it was praised for what it gave and what it didn't. I'm convinced Core guys knew damn well what they were doing and didn't improvised much in their games. The rules of the game were given, they were clear and at the beginning you were provided an exercise terrain to get a grip on your puppet, learn how to navigate in the geometry and a few hints "who" your avatar was supposed to represent. A short animation succintly explained the context and you bailed out in the Tomb Raider world without any further explanations. Up to you to explore this strange world that had been created by the clever designers for your pleasure. No maps, no words, no nothing but your ability to reason, observe and act autonomously. The pace was slow, you had to think and check carefully the given geometry versus the allowed moves and the risks of lethal mistakes were all over the place. As working people we hadn't much time to play and the game took ages before we finished it but that was precisely the point, long hours of fun but if you had to interrupt the session you just saved where you were and that was it. That's the main reason you could save anywhere and the designers were aware of that. You mentionned the blocky textures. It always appeared to me as an interesting artefact due to the technical limitations of these days that indeed contributed in a significant amount to the difficulty in navigation but simultaneously benefited the puzzling aspect of the game. These "ugly" textures reminded me of those images used by the psychologists as examples of brain activity involved in perception, like these colored numbers to evaluate color blindness, depht and size cues room or the black and white Dalmatian sniffing the ground under a tree etc. In the same vein, what couldn't be given visually or otherwise was left to the imagination of the player.The few hints or objects disseminated throughout the games triggered questionnings and speculations about this "world" the creators had in mind, letting the player to "fill in the blanks" (Why this flying saucer in area 51? What has Area 51 to do with Atlantis? Will I find the answers in this installment of the serie or will I have to wait the next one to see my supputations confirmed?) I loved this serie and anticipated the next installments. I bought new titles eyes shut and was always satisfied. I also remember there was a strong fan base online, mainly composed of adults in the beginnings at least. Loads of dedicated sites like Stella Lune you mentionned. Discussions forums that offered people a place to share their passion for the serie. Well, all of this ended abruptly when they gave the developpment to other people who threw out the baby with the bathwater. They put the beef of the game, exploration and platforming, to the bin, they focus on Lara Croft's sentimentality (we couldn't care less at the time, just found comical their choice of a booby avatar), shove down our throat their QTE, took us by the hand to the end of their show and to top it now they made it having someone over your shoulder whispering the answers of your crossword? What is left for the player in these sort of games? And something similar happened with other series. Just look at Resident Evil to name it. Btw, sigh... Now retired, looking at my consoles, collecting dust, I tried finding good games but just couldn't find much products up to my expectations. Maybe Bioshock and a couple others. So I decided to unearth my old discs. Found out many are still interesting, almost fresh, ten, fifteen years later. I'm now buying back those I had the bad idea to let go in exchange, those I missed when I hadn't the time to play or wasn't aware of when they where released. Sometimes at premium prices, because they became collectors items. That is really sad. I would like to put more of my money in the actual market but I just can't find games that fit my kind of profile and I can tell you there's a lot of persons like me out there who felt let down by the industry and just turned their back to video gaming. Well, so long folks, that was my two cents.

Stephen Chin
profile image
Great article and a good case study on how less than a handful of very simple mechanics (jumping, sliding, walking) can spawn very clever and creative gameplay.

Shaun Friend
profile image
It's great to see an article on earlier Tomb Raider's controls on a high-profile site that goes beyond instantly dismissing them because they aren't immediately accessible, but some of the things on the third page make this feel a little "close, yet far" in getting to the heart of the strengths that it has over newer action/adventure control schemes.

The major strength of the original controls is their consistency in both function and interaction with the world. If you position right all the situations in this article can be "solved" 100% of the time, and Lara having very clear limitations to understand means that if something seems like it should be able to be done, it can be, including things not intended. Outside of the boundaries of the world and rare overzealous mesh hitboxes you never have to worry about invisible walls or similar stopping you from doing something that should be clearly possible. Modern action/adventures might be far more accessible but it's pathetic how little interaction with the environment they have in comparison; the only one I've seen capture anything like the original TR's is Mirror's Edge, which has a few issues in a couple of areas but actually feels like the closest thing to a natural evolution of the original TR controls and environmental interaction (including unintentional interactions that nonetheless add depth and exploration).

Saying the games are lacking depth just because properly understanding them isn't required to finish the game normally is a little questionable when there are many facets to the way you can interact with things that allow you to do things far beyond what the developers obviously intended even without glitches. There are also the various ways you can interrupt animations and use certain geometry to your advantage that lead to so many nuances for things like speedrunning that new things are still being discovered to this day.

As an extension of that, the dismissal of the sequels and certain things they bought also sells things short; the mid-air twist was already mentioned in this article but undersold on what a massive change it makes to the amount of flexibility you have (with the example given in the article being just one way it can be used), as well as allowing for some interesting and complex mid-air roll+air control navigation. In contrast to the crawl, the sprint that was also added in TR3 is also interesting in how the big speed improvement allowed the developers to make larger levels while minimising the amount of extra backtracking and time spent in empty space (assuming you don't get too stuck anyway, which is admittedly a pretty big assumption in the context of TR3's absurd difficulty). The sequels also generally make more complex and interesting use of space and the moves provided.

Given the bringing up of games like VVVVVV I wonder if the article writer would be intrigued by certain custom levels made for the Tomb Raider editor; while frequently incredibly difficult and complex they allow you to save anywhere (avoiding much progression loss) and many explore the kind of scenarios presented in this article to a much greater degree, from bigger scale versions of the situations in this article to crazy stuff way beyond anything demanded/explored in the main games (such as the previously mentioned mid-air roll+air control combos).

Hamish Todd
profile image
Ok first of all, this is an excellent comment. It has made me feel good about coming to Gamasutra; my previous articles have been on consumer websites, and while everyone is very nice about what I write, my words aren't expanded upon in this knowledgeable, civilized, challenging way.

Paragraph 2: This is a very good assessment - consistency is completely crucial! In retrospect I wish I could have talked to you before doing the article, I would have slipped something in there. Consistency is why, looking for the kind of coolness I've described in this article, you would be better off looking among platformers than action games in modern times.

You would perhaps be interested in my analysis of a battle in Shadow of the Colossus
f-the-colossus/ - I have a little soapbox about sublime consistency at the end. SotC is cool because it manages to combine "cinematic" action-game-events with a lot of consistency (i.e. you often have to think about the physics engine). However, like many cinematic-action-games, certain "ideas" in the game generate inconsistency that feels horrible. I'd be interested if you have any thoughts about what I wrote in the last section? In that article, by the way, I'm also more willing to talk about *stuff the developers did not intend*.

I have two statements and I would like to know if you would agree with them. First: I think Tomb Raider's consistency was born of technical limitations. There is a "one-off" feeling to certain puzzles and events (like the midas hand that turns you to gold). This suggests to me that the team were saying "If we can do something cool and it means being inconsistent, we'll do it, although we often find we can't do things for technical reasons."

The other thing I want to say is that it seems that "cinematic" action-game-events are opposed to the idea of consistency. If you want to have something REALLY bombastic that feels

Paragraph 3: I only partly agree with you here. One problem is this: in a manner of speaking, all games are infinitely deep (many games are Turning-complete for example). Any game can become interesting if you're allowed to go nuts talking about "what gets utilized in the speedruns, or at the highest level of PvP? What are the player-built levels like? What happens if there's a buffer overflow?" And to be sure, when people write about that it is very cool! I encourage them to continue doing it, especially because there are a few kinds of game that can only be talked about this way (minecraft, Noby Noby Boy). But that isn't what I do.

I talk about level design, and I hope to show how important level design is. So you have some interesting *possibility* within your game engine - great, that is a good reflection on your engine. But if the player has to spend hours doing dull things to find it, that sucks. People's lives are short! There are many better things they could be doing than carrying out dull tasks in a video game so that they can make a minor discovery! And game developers need to respect that. If they don't, their games will not be played by adults who value their time. It is perfectly reasonable for someone to evaluate depth based on "what was sum total of the stuff that I saw in my playthrough?".

Problem is, making sure that players pick up on and fully comprehend a possibly-very-delicate phenomenon in your game engine is extremely hard. It is why I am not impressed when I hear someone say "we're still discovering interesting things in [game] today!". That doesn't mean it's much deeper than other games. It just means that it was bad at expressing itself in a concise way, but it has a dedicated fanbase.

I fully agree with paragraph 3 - the midair roll is very thoughtful. I'll even tell you this: Tomb Raider 3 was the only entry I played in my childhood, and I would never have done this article if it hadn't been for the interesting opening to All Hallow's (hopefully you know what I'm talking about?). I do admire that part, but it wasn't interesting enough for this article.

"The sequels generally make more complex and interesting use of space and the moves provided" - I agree there are some interesting spaces (and they are made possible by the moves that bore me), I said as much. But "interesting use of moves"? I don't think you can substantiate that - though I would LOVE to be proven wrong. As I say I played a lot of TR3 , and researched TR2 for this article, but after many hours I had come up with very little (jumping-while-shooting-the-bell I will give you). Even Stella didn't recommend playing 4 and 5.

All this said, I would be quite interested by player-made levels. I came up with lots of ideas myself while writing the article, but I think if I were to make them nobody would be very interested ;) Since you've mentioned they are hard-yet-forgiving, I can see that you have understood my point about communication.

Shaun Friend
profile image
I admit I wasn't expecting a reply, which is why I kept my post pretty impersonal. I'm glad to be able to follow up though. Guess I'll go through your responses in order.

First I'll note that my laments about modern platforming are based around 3D games; I don't think 2D games have fell into these issues with inconsistent design, and are still mostly based around giving consistent limits to the player character. I agree that 3D platformers do a much better job avoiding the inconsistent design that action adventure games seems to have fell in to, but even they feel thin on the ground lately (the Mario Galaxy games are the only ones that immediately come to mind, unless you count Mirror's Edge under this genre).

Your link to the Shadow of the Colossus article makes me realise you're the one who wrote things like the article on Medusa heads and others that have been some of my favourite recent articles on game design, which is cool. I'm not sure on what I could add to that end part of the SOTC article, but I guess I agree about variety vs consistency, and would say that modern action/adventure platforming seems to be suffering from too much bias towards the former.

Your second statement seems to have cut off so I can't really answer that without the full thing, but I'd agree with your first statement. Many of Tomb Raider's stronger aspects being born of technical limitations and perhaps unintentional design is pretty clear in several ways, right from the base design of it having to be optimised around a D-pad rather than analogue sticks. This is slightly off-topic from platforming and I'm not sure if you know of it (here's the video if you don't, but I think the fact that Core Design's cancelled version of Anniversary shows the T-Rex encounter as a cutscene just like Crystal Dynamic's released version is a big hint at this. I'd say the T-rex encounter in the original is such an iconic moment in major part because of how it doesn't disrupt the base gameplay at all, and simply expects a natural reaction to something horrible coming around the corner at you and runs with the consequences of the choice you make in that moment of terror. There's a lot of belief (that I fully agree with) that Anniversary butchered the sequence, and I think that's the fundamental issue; making it a cutscene and removing all player interaction from the reveal inherently destroys what made that part work in the original. The fact that Core Design's own version of Anniversary would have ruined the encounter in the same way makes it clear that they would have probably did a cutscene there if they had the technology in the original game too; which I think is a pretty big hint of technical limitations driving many choices in it. this tangent is admittedly a bit of a leap, but it's the best way to sum this up I can think of.

I don't agree with the idea that any game is "infinitely deep" though, and that's there's no difference except player interest; many games clearly have lower skill ceilings regardless of how many take interest in them. I can relate this to the series itself (I admit this is a little technical as it based around the speedrunning community); despite having a pretty equivalent fanbase and heavy initial interest speedrunning of Crystal Dynamic's Tomb Raiders have mostly died simply because there's not too much you can do with them. Beyond certain glitches, past a point it just becomes about throwing yourself at stuff and praying it works because there's almost no room for player skill or style to come in, which isn't very fun either, and, despite them being easier and more accessible than the older games, far more runs of the newer games have been ended in ragequits because even basic things are just luck past a certain point. Whereas the earlier games have many subtle (and not so subtle) ways to interact with things that allow for constant refinement, and players have different ways of doing things without them necessarily being "worse" than each other, along with that final element of consistency meaning it's possible to keep getting better and better at them with little except your own skill level holding you back from doing so. regardless of how much of it is intentional, the level design plays a vital part in all this too. Unless I've completely misinterpreted your point about depth or something.

"The sequels generally make more complex and interesting use of space and the moves provided" I admit I kept this pretty vague because I wasn't expecting much of a response, but this is somewhat of a gut thing based on my experiences with things such as speedrunning; between certain new moves and the design itself the levels in the sequels generally have many more ways to optimise your movement through the spaces, as opposed to the original where things remain relatively basic. Maybe I'll think about this more in terms of posting something more concrete.

One thing that quickly comes to mind that would fit this article (in the "parody" category) is a room from "Living Quarters" in Tomb Raider 2 (around 6:44 in this video You trigger two timed trapdoors over a spike pit and have to precariously shimmy across them, the timer is extremely forgiving and it'll only really catch you out if you don't realise they are timed, but the mere knowledge that they are combined with the slow shimmy animation makes every moment agony as you think it could give away at any moment, it's a small acknowledgement of the slow speed of the shimmy, while also making use of that to create high tension despite relatively safety.

As for Tomb Raider four and five, I think the first half of four is very strong, but I think the game falls apart the moment you reach Cairo (and for me that marks the moment the original series falls apart for me too), and I think five is very weak beyond one or two vaguely interesting levels.

Hamish Todd
profile image
Ah, how embarrassing that I forgot that paragraph! Don't worry about it, I was going to delete it since I had moved its thought somewhere else.

I agree Re: 2D-3D split. It's annoying because it's not that hard to get a pretty challenging 3D platformer without too much effort, like Kororimpa or Igneous.

I went a little nuts saying "all games could be infinitely deep". I kind of meant that if you have a level editor you can treat any game like Minecraft and play with the physics engine.

Allow me to modify my words to make it more relevant: I want to say that it is not very hard to make a game with the depth Tomb Raider had. I could take any of the by-the-numbers platformers on the nintendo 64, most any sports game, most any First Person Shooter. They will all have possible sequence-breaks, AI exploits, path-drawing questions, and physics bugs for an enthusiastic player with savestates and random-number-generator-manipulators to get their teeth into. Do you know anything about the Pokemon Yellow speedrunning community? It. Is. Bananas.

The reason the crystal dynamics TR games come out looking quite so ridiculous is that action adventures have measures in place that seem to *prevent* depth.

I watched a couple of TR 1-3 speedruns, and while I'm not acquainted with the community, I'll bet that nobody likes those block puzzles, huh? A lot of time spent sitting there watching the same four-second canned animation. Modern action-adventures, including crystal dynamics' tomb raiders, their problem is the same as what is happening with those block puzzles: the desire to show off animations takes priority over giving players autonomy.

This conversation makes me want The Last Guardian to come out o_0 will it keep SotC's potential-filled physics engine, or will animations take over?

Shaun Friend
profile image
Maybe you already found it but if you're checking out TR speedruns then is the best place to start, as it includes the fastest known times for each level and various categories of run, along with a breakdown of all the tricks used in runs. You're right about the block puzzles, although the majority nowadays are either skippable or can be heavily cut down.

You are right that many games are filled with tons of glitches and tricks that can be used in higher-level play, but I don't agree they are all potentially equal because of that; in many cases the tricks just simplify things or add a bunch of luck manipulation; it's pretty sporadic whether they actually add to the overall skill ceiling or not. While there are major exceptions, when I think about it (in terms of 3D games anyway) this seems to happen most often among relatively early 3D games like Super Mario 64, Tomb Raider and Quake (and games sharing that engine at the time), which makes it seem like an inadvertent feature of the relatively undeveloped engines of that time period. It's shame more modern games don't try intentionally experimenting with that kind of "strangeness" though.