Course And Ef­fect

Set­ting sail in Bossa's mind-blow­ing Worlds Adrift, a fully phys­i­cal and per­sis­tent world that's poised to change the MMOG's tack for­ever


How Bossa’s mind-blow­ing ad­ven­ture sand­box Worlds Adrift is set to change MMOGs for­ever

cour­ing its of­fi­cial YouTube chan­nel, it’s hard to grasp what Worlds Adrift is, never mind what it might rep­re­sent. Dur­ing our first hour at Bossa Stu­dios, our in­ter­est is aroused, but a fly­ing-cam­era tour around Adrift’s shat­tered, sky-bound is­lands fails to epit­o­mise the ti­tanic un­der­tak­ing, in con­junc­tion with tech startup Im­prob­a­ble, that Bossa is re­luc­tant to la­bel an MMOG. The in­no­va­tions aren’t sub­tle, but rather so enor­mous as to de­mand a shift in per­spec­tive be­fore their im­pli­ca­tions can be ap­pre­ci­ated. As co-founder Hen­rique

Oli­fiers and game de­signer Luke Wil­liams mess about in the clouds, ex­cite­ment starts to burn as if see­ing Minecraft for the first time, and sud­denly it seems a mon­strous er­ror of judge­ment that Bossa hasn’t shouted and screamed about th­ese fea­tures in its mar­ket­ing. But the team is wor­ried. Worlds

Adrift is such a dra­matic de­par­ture from Bossa’s pre­vi­ous games, in­clud­ing Sur­geon

Sim­u­la­tor and I Am Bread, that any­one not watch­ing live might rea­son­ably call it faked.

Worlds Adrift is a vast, per­sis­tent physics sim­u­la­tion of a like never seen in sin­gle­player games, let alone an on­line world, and the tech­nol­ogy it’s built on threat­ens to change not just how we play but how ev­ery on­line game is de­vel­oped.

Pro­ce­du­rally gen­er­ated is­lands lace a bil­low­ing cloud­scape, and from th­ese – in line with many games look­ing to cash in on the con­struc­tion and craft­ing boom – play­ers can gather re­sources and process them into items, build­ing air­ships to carry them on­wards to new lands and op­por­tu­ni­ties. To be­gin with, it’s won­der­fully serene.

Worlds Adrift can host any num­ber of play­ers, pulling new is­lands from a queue to main­tain a bal­ance of iso­la­tion, co­op­er­a­tion and con­flict. Oli­fiers hopes you’ll en­counter

other crews with the same fre­quency as paths cross in Jour­ney – a clear in­flu­ence along­side Skies Of Arcadia and The Leg­end Of Zelda: The Wind Waker.

“The set­ting it­self is com­mon enough, isn’t it?” Wil­liams says. “But the ar­gu­ment I use is Halo: peo­ple are al­ways like, ‘Yeah, you’re on this huge halo!’ Well, no. The halo is a sky­box and I’m just on ter­rain. A lot of th­ese games use float­ing is­lands as a set­ting – they don’t make full use of it.”

Bossa’s com­mit­ment to re­al­is­ing the bro­ken rem­nants of a planet plun­dered for a lev­i­tat­ing min­eral has called for tech­ni­cal in­ge­nu­ity. Clouds are not tex­tures but vol­u­met­ric shaders with strate­gic value, able to con­ceal flee­ing ships as they ma­noeu­vre through the fog. For this to work, each cloud must be iden­ti­cal for ev­ery player. Achiev­ing that syn­chro­ni­sa­tion meant two months of work, but Bossa has been ap­proached by two com­pa­nies hop­ing to li­cense the tech­nol­ogy for the likes of flight sim­u­la­tors.

Worlds Adrift is lit­tered with history of the peo­ple who used to live here (sev­eral nov­els’ worth, ap­par­ently) and by pho­tograph­ing and doc­u­ment­ing your trav­els, you gain knowl­edge of new schemat­ics to im­prove your en­gines, your can­nons or per­haps your hull, the bet­ter to pen­e­trate Adrift’s storm walls, which serve as pro­gres­sion gates. There’s no lev­el­ling, only the pur­suit of new tech, knowl­edge and the

EVE- like free­dom to de­cide your own ob­jec­tives based on what you know. Here, Worlds Adrift steps back in time to an age when insight into an on­line world was shared by play­ers and not walk­throughs.

“Th­ese storm walls change in thick­ness through­out,” Wil­liams tells us, “so if you


find where one’s par­tic­u­larly thin, that’s valu­able in­for­ma­tion that’s not eas­ily trans­ferred over a wiki or some­thing. You’re draw­ing your own maps, Phan­tom Hour­glass style, and your place in the world is com­pletely rel­a­tive to what you’ve drawn. The play­ers are gen­er­at­ing valu­able knowl­edge that’s ac­tu­ally very lo­cal.”

Each player’s unique per­spec­tive on the world is the first hint that Worlds Adrift is more than an air­ship-en­abled sur­vival sim. This knowl­edge, recorded on a ship’s map­ping ta­ble, is per­ma­nent, ex­ist­ing un­til mod­i­fied by a player or Bossa shuts down

Worlds Adrift. It can be shared, sold, stolen or de­faced, and the same is true of ev­ery last ob­ject: each tree, each branch you fell, each can­non­ball fired, and each wreck you aban­don will lie there – iden­ti­cal to ev­ery ob­server – un­til dis­turbed.

“While we have the back­story of the world and all the history the play­ers will be dis­cov­er­ing,” Wil­liams says, “the present­day [game]world doesn’t have any­thing un­less the play­ers put it in. So the sto­ries of the crashed ves­sels and the camps that you find – they’re all player placed. If this ship were to crash and all th­ese parts were to fly off or be part of a wreck­age, you’d be able to find that and go through it all and maybe find their map and all the notes of what they’ve done; you’ll find pic­tures of where they’ve been, and so there’s this strange sto­ry­telling that’s real history.”


Stranger still, the first thing you do in

Worlds Adrift is re­learn how to move – and how to move other things. WASD still works, of course, but walk­ing isn’t your go-to method of tra­ver­sal, for even Spi­der-Man would envy the free­dom of 3D mo­tion of­fered by your grap­pling hook. The rope, link­ing player to cho­sen pivot point, is a phys­i­cal ob­ject: swing past a tree and it will catch in the branches, chang­ing your course and mo­men­tum. Or, if you like, you could stand on the ground, weav­ing your rope around many trees to cre­ate a cat’s cra­dle. You don’t have to – there’s al­most cer­tainly no point in it, and Bossa didn’t code it so that you’d be able to do so. You just can, be­cause all of Worlds Adrift’s ob­jects are phys­i­cal en­ti­ties. If it ex­ists, you can in­ter­act with it as far as the laws of physics per­mit.

Wil­liams tries and fails to pull an en­gine up a hill, be­cause it’s heav­ier than he is. With pro­ducer Herb Liu’s as­sis­tance, it be­gins to inch for­ward. The ap­pli­ca­tion of two anti-grav giz­mos have it float­ing like a he­lium bal­loon. We then watch as Oli­fiers chops down a tree as he pleases, shear­ing a branch and then the lower trunk, top­pling it with a tug of his grap­ple.

“If you roll one of th­ese logs down a hill, you’re go­ing to kill the player,” he says. “In the same way, if you take one of th­ese logs and rope it to your ship, you can use it as a pendulum to hit an­other ship. You can do what­ever you want with th­ese things; they’re phys­i­cal ob­jects. They serve as har­vest­ing [re­sources], fine, but that’s not where it ends; that’s where it starts.” Liu oblig­ingly tries to crush him­self with the en­gine.

‘MMOG’ is a poor la­bel for Worlds Adrift, the con­cept of which is as dif­fer­ent from World Of War­craft as Su­per Mario 64 is to Zork. En­trenched ideas of lev­el­ling and quest­ing break down when you re­alise a dread­nought can be scup­pered by fly­ing above it and raining junk on the pi­lot (al­though the pru­dent may have built one or more backup helms be­low deck).

“It is lit­er­ally im­pos­si­ble to bal­ance be­cause of one small as­pect,” Oli­fiers states. “Be­cause it’s all phys­i­cal, there is no es­tab­lished strat­egy to do any­thing. There is no such thing as the best way of en­gag­ing in a bat­tle or do­ing any­thing. There’s no best way of chop­ping down a tree. It’s a self-bal­anc­ing sys­tem, be­cause some­one will find a way to do things and some­one else will find a way to top it in­fin­itely.”

Worlds Adrift is for the dis­cern­ing sky pi­rate: if you want to see the enemy smashed for no other rea­son than your own sat­is­fac­tion, you’ll want to dis­able their anti­grav cores; or maybe you want their ship for your own, in which case you’ll need to leave it un­harmed as you take out their respawn­ers. The capri­cious could en­ter from be­neath, leave a pho­to­graph as ev­i­dence of their pres­ence and van­ish into the night.

“It’s not tra­di­tional MMOG loot­ing – it’s ac­tu­ally loot­ing,” Wil­liams says as he cuts an en­gine free. “I have de­tached it from the ship, strapped some stuff to it so it wouldn’t fall, and I’ve now es­caped with it – I could take it back to camp or maybe use the ship’s har­poon and drag it with me.”

A life­long videogame player ac­cus­tomed to idio­syn­cratic re­al­ity-fudg­ing will find that this phys­i­cal world is not al­to­gether in­tu­itive. You have to learn how to prob­lem-solve again. For ex­am­ple, if you needed to punch through a hull with tough ar­mour plat­ing, the nor­mal so­lu­tion might be to ac­quire a

‘Su­pe­rior Can­non of Pierc­ing’. In Worlds Adrift, a heav­ier can­non­ball will serve just as well. It’s a shift from game logic to real logic.

In ad­di­tion, Worlds Adrift prom­ises an ecosys­tem of the sort that MMOGs have claimed to be im­ple­ment­ing for years, in which an­i­mals and plants boast com­plex, in­ter­de­pen­dent AI and life­cy­cles that go far be­yond spawn, fight, die.

“They don’t spawn,” Oli­fiers tell us. “If a crea­ture is in a mat­ing sea­son, it has to find a gen­der-spe­cific crea­ture com­pat­i­ble with it. In the same way, they oc­cupy an is­land, and if the player is just ar­riv­ing to that is­land for the first time, there will be a bal­ance there; but if you start to chop down all the trees and gather ev­ery­thing, the crea­tures will get hun­gry and be much more ag­gres­sive to­wards the player. The bot­tom line is: play­ers could wipe out life from an is­land. We have sys­tems to rein­tro­duce life later on, such as fly­ing crea­tures that mi­grate from is­land to is­land, or they take seeds with them or they poop some­thing and it grows back.”

This dy­namic par­adise is done jus­tice by its prodi­gious draw dis­tance. We’re far out into the sky and Wil­liams in­di­cates a mov­ing pixel in the dis­tance. Zoom­ing in with a cam­era (tele­scopes are in de­vel­op­ment) re­veals a fully ren­dered Liu jump­ing and swing­ing from an is­land that looks for all the world like it’s me­tres away.

We have an ap­point­ment with Her­man Narula, CEO of Im­prob­a­ble, whose tech­nol­ogy un­der­pins Worlds Adrift’s feats of physics and per­ma­nence. Oli­fiers, hav­ing spent the morn­ing in a height­ened state of ex­cite­ment, is now beam­ing at hav­ing con­vinced his au­di­ence that what they’re see­ing is real de­spite read­ing like science fic­tion. The man we’re due to meet makes him seem dour by com­par­i­son. Narula flits breath­lessly be­tween ado­ra­tion for the lat­est build (which he of­ten breaks off to try), an im­promptu lec­ture on the science be­hind it, and plan­ning new fea­tures with Oli­fiers. “I won­der if we can do wa­ter?” they mut­ter. “There’s an op­ti­mi­sa­tion or two we could do to ac­tu­ally get it to work.”

With drive like this, it’s lit­tle won­der that Im­prob­a­ble talked $20 mil­lion out of ven­ture cap­i­tal firm An­dreessen Horowitz. Early in de­vel­op­ing its tech, Im­prob­a­ble reached out to de­vel­op­ers brave enough to play guinea pig, and dis­cov­ered ev­ery­thing it hoped for in Bossa and Worlds Adrift.

“It’s the great­est metaphor for player free­dom,” Narula says. “It’s ex­plic­itly in its bones: soaring through the clouds on wings. It is the ab­so­lute op­po­site of the con­straints and bound­aries peo­ple nor­mally as­so­ciate with on­line games. The no­tion of a world where ev­ery ob­ject you en­counter will be in­ter­actable will be the best kind of state­ment for Im­prob­a­ble and our tech, which is about making ev­ery­thing come to life.”

“We are very dis­ap­pointed by the evo­lu­tion of on­line games in gen­eral,” Oli­fiers adds. “Where ev­ery sin­gle other area of game­play and game de­sign has been evolv­ing a lot in the past ten years, MMOGs are pretty much the same. Can you imag­ine play­ers who grew up play­ing

Minecraft, where they have the free­dom to do what­ever the hell they want… It’s im­pos­si­ble for a player like that to be put into an amuse­ment park.”


The so­lu­tion, as Narula de­scribes it, is a re­think from first prin­ci­ples about how an on­line world is ruled by its servers. In a tra­di­tional MMOG, any given re­gion of space is ruled by a server, han­dling ev­ery­thing from player po­si­tion to chat. Clients act and the server re­acts, ad­vanc­ing time by one frame and push­ing the re­sult back to ev­ery­one. Lag and in­sta­bil­ity de­velop where there are so many clients con­nected to the server that it can’t process all the in­puts be­fore ad­vanc­ing time. EVE copes by slow­ing time; other games en­sure what the server has to com­mu­ni­cate is in­cred­i­bly sim­ple, al­low­ing for mere scraps of logic un­re­lated to the play­ers.

“We don’t do this. This is bad. This isn’t the best way to build a mod­ern dis­trib­uted sys­tem,” Narula says. “Imag­ine if we pre­sented to the de­vel­oper a to­tally ab­stract world with no game loop, no time step – ev­ery­thing hap­pens si­mul­ta­ne­ously. All en­ti­ties really are is a lit­tle bit of state – like their health, their po­si­tion or their ro­ta­tion – and some be­hav­iour. At its core, Im­prob­a­ble’s sys­tem un­der­stands this ab­stract rep­re­sen­ta­tion of en­ti­ties and it cheats. Be­hind the scenes, we’re able to run this on lots of ma­chines be­cause th­ese en­ti­ties are fun­da­men­tally in­di­vid­ual pro­cesses.

“But ac­tu­ally this isn’t that great. It’s just this big ab­stract world. There’s no physics there – there’s no awesome game en­gine do­ing nav­i­ga­tion and other stuff. And those game en­gines, they need game loops, so what we do is we have a sep­a­rate layer and in­stead of one mam­moth, in­tel­li­gent server en­gine that un­der­stands the whole world, we treat en­gines like Unity as dumb lit­tle work­ers. We make them au­thor­i­ta­tive


over a sub­set of th­ese en­ti­ties in the world. So the en­tity says, ‘Hey! I got ro­ta­tion; I got po­si­tion – I want a physics en­gine to do my physics!’”

The largest de­ploy­ments tested by Im­prob­a­ble in­cor­po­rated 200 servers with 700 in­stances of Unity run­ning across them man­ag­ing a world 160 square kilo­me­tres – an area the size of Is­rael. Us­ing bots, Bossa launched a 300-ship bat­tle in­volv­ing 1,000 crew, and the game rum­bled on un­scathed. At this point, when server tech­nol­ogy is no longer a lim­it­ing fac­tor, de­sign con­sid­er­a­tions take over. Thou­sands of can­non­balls fly­ing from all di­rec­tions is un­playable chaos that would kill client-side hard­ware long be­fore the servers started sweat­ing. Now, de­sign­ers are tasked with cre­at­ing or­ganic worlds that bring peo­ple to­gether in a man­ner and vol­ume suited to the in­tended tone.

The re­sults of the di­vi­sion of labour are more pro­found than the abil­ity to in­clude true physics in an on­line game. Since the as­sign­ment of work is in flux, en­gines pop­ping up and go­ing down as needed, host­ing the world is cheaper and more stable than the sin­gle-server equiv­a­lent. The fail­ure of one en­gine can’t scup­per the game, which Narula demon­strates for us through a clean and leg­i­ble browser in­ter­face: he kills an en­gine re­spon­si­ble for a col­lec­tion of items scat­tered about a vis­ual rep­re­sen­ta­tion of Worlds Adrift’s is­lands, and its work is in­stantly dis­trib­uted be­tween those that re­main. At Bossa, the process is at­tended by one man in place of a dev ops team.

The ‘how’ of the dis­tri­bu­tion is hid­den from the de­signer – that’s Im­prob­a­ble’s job. The pro­gram­mer codes be­hav­iour in Unity, us­ing the Im­prob­a­ble API, and lets it fly – a new, live fea­ture that didn’t re­quire down­time. Worlds Adrift’s hang glider was im­ple­mented af­ter a sin­gle day’s game jam­ming. More sig­nif­i­cant still, Im­prob­a­ble does away with the need for the multi-year de­vel­op­ment of net­work­ing code for a game to sit on: any game built with Im­prob­a­ble is in­her­ently mul­ti­player, ex­pos­ing the costli­est genre to in­die de­vel­op­ers. Ad­di­tional en­gines will be in­te­grated with the Im­prob­a­ble tech as it de­vel­ops.

Im­prob­a­ble and Bossa, acutely aware of the value in their offering, are em­ploy­ing shock and awe tac­tics, which, if suc­cess­ful, could ce­ment their place at the fore­front of on­line games for the decade to come. As Im­prob­a­ble’s lab rat, Bossa has had to solve a great many prob­lems of its own, such as the global physics in Worlds Adrift, con­tribut­ing the re­sults back to Im­prob­a­ble’s code li­braries. Dean Hall’s mys­te­ri­ous workin-progress, Ion, also uses Im­prob­a­ble and al­ready in­cor­po­rates code writ­ten by Bossa. How­ever, the enor­mity of their joint am­bi­tions is made clear in the rev­e­la­tion that Worlds

Adrift will be fork­able in its en­tirety: any new de­vel­oper can press a but­ton and gen­er­ate a fresh copy of Worlds Adrift, run­ning on Im­prob­a­ble’s servers, to edit how­ever they please. Any­thing Bossa can do, you can do.

“To be clear here,” Narula says, “the mod­ders or the fork­ers of the game will be able to sup­port de­ploy­ments of big sizes, just like Bossa can. It’s a ser­vice that they’re also able to use. If their version of the game is more pop­u­lar, then peo­ple will be on that! And Bossa have a deal which means they’ll do well too. The cost will be pro­por­tional to the scale of the world,

which is hap­pen­ing or­gan­i­cally de­pend­ing on how suc­cess­ful their game is.”

Narula presents his ap­proach to on­line gam­ing with such el­e­gance as to make it seem im­pos­si­ble that this has never been at­tempted. It’s easy to forget that it has taken the com­bined brain­power of for­mer mem­bers of Google, Ama­zon, Gold­man Sachs and Cam­bridge Univer­sity to de­velop the tech­nol­ogy, a feat he didn’t see hap­pen­ing within the game in­dus­try it­self.

“The re­al­ity is that the in­fra­struc­ture we had to build didn’t ex­ist, and a lot of stuff we’ve had to pretty much in­vent our­selves,” Narula tells us. “The game in­dus­try wouldn’t have cre­ated an in­vest­ment hole that would have at­tracted the non-gam­ing tech­ni­cal tal­ent nec­es­sary to make it hap­pen, be­cause the prob­lem sits be­tween gam­ing and out­side of gam­ing.”

Im­prob­a­ble’s ap­proach to dis­trib­uted sys­tems is a so­lu­tion to a gen­eral prob­lem: many in­dus­tries, from health­care to util­i­ties, need re­al­time, in­ter­ac­tive sim­u­la­tions to run across mul­ti­ple net­works. Im­prob­a­ble splits work that was pre­vi­ously in­di­vis­i­ble. Given de­mand and ob­vi­ous in­vestor in­ter­est, it begs the ques­tion of why Im­prob­a­ble is work­ing with games at all. In col­lab­o­rat­ing with Bossa, Im­prob­a­ble hopes to show­case its in­ven­tion in set­tings that pro­voke won­der – tac­tile game worlds that con­vey the po­ten­tial of a new tech­nol­ogy in ways board­room Pow­er­Points can’t. Worlds Adrift, which al­lows you to soar with friends in a ship of your col­lec­tive de­sign, sift­ing wreck­age, and making history, looks set to do just that.

Sev­eral fea­tures are the prod­ucts of in­ter­nal one-day game jams, in­clud­ing the glider and a cam­era to take selifes that you can frame

Those dents aren’t likely to buff out. At­tach plat­ing to your ship and it con­forms to the cur­va­ture of the skele­ton, re­tain­ing its shape and any bat­tle scars should it pop off

De­signer Luke Wil­liams pre­vi­ously worked on Bossa’s IAmBread and Sur­geon Sim­u­la­tor

The ship­yard is an anti-grav­ity field used for con­struct­ing ships be­fore they get their own cores. It takes some of the has­sle out of phys­i­cal con­struc­tion, al­low­ing you to at­tach parts with ease

Is­lands are drawn from a queue and ar­ranged such that you’re never much far­ther than one tow­er­ing cloud­bank away from sight­ing land

Jour­ney is a pow­er­ful in­flu­ence. Though Bossa has stress-tested Worlds Adrift with le­gions of bots, it in­tends each meet­ing of crews to be sig­nif­i­cant and mem­o­rable

In the heart of Lon­don’s Shored­itch, de­sign­ers, artists and mar­keters work next to each other in Bossa’s open-plan of­fice

Her­man Narula, CEO of Im­prob­a­ble, is three years clear of Cam­bridge and, along with Rob White­head, has found him­self lead­ing an en­vi­ably tal­ented team

Newspapers in English

Newspapers from Australia

© PressReader. All rights reserved.