Hero shooters are the latest gaming trend, team-based competitive multiplayer shooters featuring objective-driven game modes and large casts of colorfully designed characters sporting fantastical weapons and abilities. That s a serviceable but still somewhat shortsighted definition. The hero shooters we have today already cover a range of styles and goals. They re all shooters, so we see a lot of overlap in how they actually play, but each one approaches things differently and has something distinct at its core.

Basically, hero shooters are like the Powerpuff Girls, so let s talk Chemical X.

We ll start with Blizzard s Overwatch, seeing as how it has become the poster child for the genre. Overwatch is popular for many reasons, but primarily because of its heroes and their diverse playstyles and personalities. Game director Jeff Kaplan said it best when we spoke with him at BlizzCon 2015: the heroes are the content of the game.

More than any other hero shooter, Overwatch puts individual heroes first, not overarching classes. This wasn t always the case, though. In an April interview with GameSpot, Kaplan said the Overwatch team started with a class-first perspective. It was lead hero designer Geoff Goodman who eventually proposed making as many classes as we could come up with and simply turning them into highly specialized heroes.

Overwatch heroes are roughly grouped into four classes, but hero-specific strengths are stressed more than in, say, Team Fortress 2. In Overwatch, you pick a hero based on who your team needs and who the situation calls for, whereas in TF2 you have more room to choose a class based solely on how you like to play, illustrating the subtle difference between team-based and class-based gameplay. A skilled Heavy can fill several roles and always be useful, but you can only do so many things with Roadhog or Reinhardt, which brings us to hero switching.

Instead of making individual heroes more flexible, Overwatch made all hero switching available on the fly. You can choose from several suitable heroes in any situation, but you can t take the same approach with all of them, at least not effectively. Reaper and Tracer, for example, are both offense heroes who can make short work of enemies, but they play very differently and have different counters.

This creates the type of rock-paper-scissors meta we see in MOBAs like Heroes of the Storm, which is a noticeable influence on the way Overwatch heroes interact, both as allies and as enemies. There s also a bit of fighting game spirit to Overwatch. The way they use huge casts to segment fundamentally similar gameplay forms an interesting parallel with games like Street Fighter V and Tekken 7, which also live and die by their rosters.

Heroes are obviously central to all hero shooters, but careful character design is of unmatched importance to Overwatch because each of its heroes must not only serve a unique purpose but also mesh well with tightly clustered teams. Everything starts there, and anything that encroaches upon the fun hero gameplay, as Kaplan puts it, is swiftly changed. Furthermore, Kaplan told Kotaku in July, maps and modes are quite deliberately designed to promote hero interaction by bringing teams together. This lines up with what assistant game director Aaron Keller told us in September: "if and when we release a new game mode in the future, it will be about teams pushing on a single objective.

At the opposite end of the spectrum and it is a spectrum, this genre-hopping blur of a genre we have Gearbox Software s Battleborn, which doubles down on MOBA elements while turning Overwatch s character design on its head. Overwatch is about selecting and re-selecting the most suitable hero, but Battleborn is about building a hero that does what you want.

In Battleborn, hero abilities aren t static. Heroes bring the same skills into every round, but through the game s helix system of A/B upgrades, these can be tweaked to fit the play style you prefer. Characters fall into designated classes like healer and skirmisher, but they can still wield a mix of offensive, defensive or supportive abilities, just not all at the same time. So, rather than which hero to play, the question becomes how to play your hero.

You can further customize heroes through gear which augments stats like cooldown reduction and maximum health. Gear functions like the items in Dota and League of Legends: earn currency and buy pieces as the match progresses, with the added kick and progression of collecting gear through loot boxes and, from your collection, building character-specific loadouts. The influence of MOBAs is even more obvious in Battleborn s waves of minions and its base-destroying incursion game mode.

This should also sound familiar to Borderlands fans. Tailoring skill trees to suit different play styles? Boosting your most relevant stats with equipment? Hoarding gear? That s a day in the life of a Vault Hunter if I ve ever heard it. And come to think of it, Alani could pass as a sister siren to Borderlands 2 s Maya. There s plenty of Pandora in Battleborn, that s for sure, just as Overwatch s focus on individual hero characters can be traced back to Warcraft.

Beyond prominent MOBA elements, Battleborn brings some promising ideas to the still-developing hero shooter genre. For starters, it has a story mode. It isn t a particularly spellbinding tale (though the stellar opening cinematic helps) but it does provide welcome context for the game s world and the motivations of its heroes, not to mention a more in-depth tutorial and testing ground. Meltdown, Battleborn s take on payloads, is also noteworthy. In it, teams defend their minions as they march toward the goal and try stop enemy minions from doing the same. This means players have two objectives and can choose to defend or attack in every round, which ties into the game s split hero development.

Overwatch and Battleborn alone illustrate some of the balancing acts hero shooters have to manage. Through the strict limitations it places on game modes and character abilities, Overwatch gains the freedom to create characters like Mercy and Symmetra, who aren t at all suited for offense roles. Battleborn is all about customization, so it s possible for players to consistently play the hero they re most attached to rather than the hero their team needs right now. There are pros and cons to each approach: I can t see incursion working in Overwatch, but Battleborn heroes tend to run together a bit, and there s no doubt as to who s winning the sales race. Somewhere between these two extremes lies Paladins, Hi-Rez Studios free-to-play contender. Lead designer Rory Drybear Newbrough described Paladins as half shooter, half MOBA when we spoke with him last December, and the current state of the game backs him up.

Although still in beta, Paladins foundation is well established, including its much-vaunted deck system. For each hero, you can build equippable decks of five cards which, once acquired mid-match, buff various abilities and help fine-tune your play style. It isn t as open-ended as Battleborn s helix system, but that s not necessarily a bad thing. The goal of decks, Newbrough said in a recent interview with PaladinsWorld, is to give players multiple build options without warping characters to the extent that they are difficult to identify or play around. This preserves the concept of instantly recognizable characters a cornerstone of hero shooters since Team Fortress Classic while creating greater room for player choice.

Paladins also pursues new combinations through its heroes, which, according to Hi-Rez COO Todd Harris, are predominantly rooted in the studio s own Global Agenda despite similarities to Overwatch. Some heroes overlap more than others, but even familiar abilities can be interesting when rearranged. For instance, Androxus wields a defensive deflection ability, generous jumps, a six-shooter revolver and a rocket-powered ultimate. Similar abilities are seen in Overwatch but the combination is novel, and Paladins is not devoid of originality. As Blizzard grapples with Symmetra s place in the meta, Ying offers a good example of how to combine turrets and teleportation in a way that s powerful and satisfying.

Even so, and even for a beta, Paladins has yet to hit its stride. Everything checks out on paper: it takes inspiration from the most successful hero shooter yet, the biggest genre in esports and lessons learned from Global Agenda. It s also free, which is a good thing to be in an increasingly competitive market. The trouble is that it doesn t yet have an identity of its own. Overwatch is a dynamic team-based FPS, and Battleborn, while a bit of a commercial flop, is firmly a first-person MOBA with added objectives. Paladins isn t quite there yet. It seems to know what it s made of but not what it wants to be, and in this it echoes the state of the hero shooter genre as a whole.

The good news is that the concern is not what hero shooters can t do, but what they haven t done so far. Creating new and interesting heroes is great, but they stand the best chance of finding an audience if they do something we haven t seen in hero shooters before. The only guiding principle seems to be to build game modes that intuitively support your heroes abilities, something Overwatch and Battleborn both managed despite being wildly different. There s no limit on what those game modes can be, only on what type of heroes suit them.

Overwatch makes a good argument for hero-driven design, and Battleborn shows that unconventional elements can be folded in without upsetting the core FPS gameplay loop. More importantly, Overwatch s 22 heroes can t do everything, nor can Battleborn s 25. Meanwhile, Paladins is still trying to figure out what its heroes are best at, something its forthcoming new mode, which Newbrough describes as a co-op experience against challenging AI, may answer.

This is perhaps the most valuable lesson for budding hero shooters like Gigantic, Lawbreakers, Battlecry, Dirty Bomb and who knows how many games to come. I don t want to see this genre become a race to dethrone Overwatch by way of character design. That s a loser s market, and hero shooters have the potential to revive and improve so many game types. It s all a matter of finding a niche and creating a good-sized roster of heroes who really synergize with it.

For one, objective-based multiplayer goes way beyond 'defend the thing and push the thing', or in Paladins case, 'defend the thing then push the thing'. Where s the fast-paced capture the flag hero shooter that promotes mobility and map awareness using heroes couched in creative movement? Lawbreakers definitely has a shot at filling that slot, but it often feels like more of an arena shooter that happens to have five classes which is fine, by the way. The point is, CTF is another classic game mode that could benefit immensely from a broad cast explicitly engineered with it in mind.

Then there s the big, white elephant that seemingly nobody is talking about: where is the hero shooter that shamelessly apes the original Star Wars: Battlefront? With droidekas, shock troops, engineers and plenty more unique classes, it s already half a hero shooter. Just imagine a modern spiritual successor, divorced from the tapestries of Star Wars but informed by the years of MOBA refinement on display in Battleborn. Sprawling, ever-changing battlegrounds peppered with command posts and crawling with NPC squadrons led by heroes I have a dream!

Future hero shooters should also consider options outside PvP. Players can work as a team without fighting other players, and PvE-focused games like Left 4 Dead are crying out for hero shooter cousins. Overwatch proved this in its recent dalliance with horde mode, the Junkenstein s Revenge brawl. As our own Evan Lahti said, that brawl didn t really work because Overwatch doesn t have the maps or heroes to support it. Where is the hero shooter that does?

For once, the bewildering vagueness of the term hero shooter can be called a positive thing. This cockamamy genre could become a nearly boundless formula for reinvigoration, a way to explore and experience familiar game modes from a refreshing new perspective via dozens of fun characters.

Team Fortress? More like Scream Fortr oh, I see Valve already did that, as it's the name of TF2's annual Halloween event. Joe brought the slightly disappointing news that there would be no all-new giant spookathon this year; instead, the game is getting a few new titbits, while all the old Scream Fortresses will be returning, as with last year.

The TF blog announces that Scream Fortress VIII has just kicked off, bringing with it "three new Halloween maps! All classic Scream Fortress Maps! All-new contracts! New taunts! And a Creepy Crawly Case with the chance to get our newest unholy Halloween creation: a taunt Unusualifier!" Yeah, that's right, an Unusualifier look it up in your Lexinomicon.

What's more, for the duration of Scream Fortress VIII, "all unusual cosmetics that you unbox from any case or crate will have a Halloween 2016 unusual effect, and the chances of receiving an unusual cosmetic across all cases, even old ones, will be DOUBLED," says Valve. Double-things are almost always good news whoever heard of Single Fine, for example?

After six years of consecutive harrowing Team Fortress 2 Halloween events, Valve took last year off to work on its then new Invasion update. It seems the same is true this year, as the developer is instead "working on something even better."

Similar to last year, every prior Scream Fortress event will briefly return including 2015's Merasmissions while a new Pyro Pack, improvements to Comp Mode, a new update, a new campaign, community maps, and a new taunt are but a few of the things TF2 is introducing to ensure All Hallows' Eve doesn't pass us by.

In typical Team Fortress fashion, here's some weird words taken from the game's blog:

"And it wouldn t be a Halloween pre-announcement blog post if we didn t scare most of you TO DEATH. For instance: What if just now you heard the rattling moon-lit sound of a SKELETON? No, nothing? Everybody still alive? Okay, but what if that sound was coming from INSIDE YOUR OWN BODY? Because there s a SKELETON in there RIGHT NOW? Oh, you already knew that, did you? Well, what if we were to tell you it s STEVE BUSCEMI S SKELETON?

"How did it get INSIDE you? What did he do with YOUR skeleton? Why don t you ask him, because he s RIGHT BEHIND YOU! No, look down! Farther! Because he s just a puddle of flesh with two BULGING EYEBALLS staring up at you! And a SNAGGLE-TOOTH! That s right, you just BODY-SHAMED STEVE BUSCEMI! Because it was YOU writing this blog post the WHOLE TIME! AWOOOOO!"

Make of that what you will. No exact date just yet, however you'll be able to revisit your favourite TF2 Halloween events at some stage "next week."

The Church in the Darkness plays like top down Hitman-lite in a 1970s religious cult. Your character decides to check in on a relative and infiltrates the sect to figure out how they're doing and what's really going on at church camp.

You navigate buildings and avoid the scrutiny of the cultists by staying out of sight, but unlike most stealth games, they won't get suspicious if you're in their line of sight, only if you get too close. By investigating the buildings around camp and searching for supplies, you'll find tools to help out with the denser parts of camp. For instance, I found a worker outfit that decreased the 'suspicion' cone around each camper so I could navigate more freely. I got in without alerting anyone, but you're free to go guns blazing if you like. I have concerns about the stealth feeling a bit too simple, but until I know what kind of challenges and tools the final game has in store, I'll keep checking in.

All the while, the church leader spouts their doctrine over the loudspeakers, but it's never quite the same every time. Voiced by videogame VO power couple Ellen McLain (GLaDOS) and John Patrick Lowrie (TF2's Sniper), the two leaders' teachings change with each playthrough. During one, they might actually be a fairly peaceful, if peculiar, religious group. During the next, they might be getting ready to take the world down with them. It's a creative form of narrative direction that I hope will influence how players choose to go about infiltrating camp. If they're a nice crew, the moral impetus might be to get in and out without harming a fly. If they're bloodthirsty zealots, well, a few flies won't matter.

The Church in the Darkness arrives some time next year.

Every once in awhile you find a crossover between two different videogames that on paper makes no sense until you see it in action. That's how I feel about Steam user Pancake's "MegaMan 6" collection of TF2 maps. Divided up into eight separate arenas themed after a different boss, you can now wax nostalgic about how much harder videogames used to be while cracking headshots as Sniper.

"It was really a combination of boredom and seeing how people create maps based on other old videogames," Pancakes tells me. "I saw that people loved playing Mario-themed maps and I wanted to test the limitations of how closely a map could resemble the original game it borrows from."

In that pursuit, I'd say Pancakes has done a damn fine job. The project has taken him over five years to complete, and it hasn t been easy. Serious limitations in the Source engine made running a map with eight separate stages quite a challenge, Pancakes tells me, and that's not to mention all the effort it takes to rip the sprites and artwork from Mega Man 6. Each arena is also accompanied by the respective music from that stage, which does wonders for driving home that Mega Man nostalgia.

While this is technically just one map, it's actually divided into randomized arenas that you'll be placed on each round. There's a standard king of the hill version but also a 'VS Saxon Hale' mode to use if you have the required mod. So far, the only caveat is that, unlike other custom maps, this Mega Man 6 bundle doesn't play nice with bots. You're going to need to spend quite a bit of time setting up proper navigation meshes instead of just letting the console do that all for you. There's a few guides to help you get started if you're interested, but you're better off forgetting the bots and wrangling up a group of friends to play with.

What amazes me is that this isn't even the first time someone's had the idea of doing a crossover between Mega Man and Team Fortress 2. DeviantArt user AgentMidnight made some excellent artwork depicting what TF2 would've looked like as a Mega Man game instead of the other way around.

If Mega Man was never really your jam, there's no shortage of Team Fortress 2 custom maps inspired by the games of yesteryear. Steam user Litronom has taken this obsession to a whole new level by creating several dozen inspired by old Nintendo games, including everything from Bomberman 64, Zelda, Banjo Kazooie, and more. Unlike Pancakes' maps, which strike a nice balance between being a homage and also being fun to play on, Litronom's maps are almost exact recreations of their source material not exactly the place to have a competitive match.

Either way, I love that mods continue to be a venue for people to express their passion for other games. They might just be a passing curiosity, but there's something so inherently satisfying about reassembling videogames inside of other videogames. So if you have any particular fondness for that golden age of consoles and are keen on reliving those memories, be sure to download the whole lot from the Steam Workshop.

Joe Wintergreen is an Australian indie game developer who is currently working on mostly a stealth FPS under the Impromptu Games banner. He also recently delivered unto the internet a series of code snippets from the great FPS Half-Life 2 by way of his Twitter account. There's not much to see of the code, but that's not why we're here. What's really relevant are the code comments, and the light they shine on how the magic is made.

The first tweet in Wintergreen's thread sets the tone for what's to come:

Good to know Valve and I pretty much have the same comments in our AI pic.twitter.com/TrWO8KUAadSeptember 1, 2016

Striders will intentionally shoot things up even if you're not in their line of sight, just because it looks cool:

Strider might still shoot if it knows it won't hit ya if there's something there it'd be cool to hit pic.twitter.com/fHwWO7Cw6CSeptember 1, 2016

Charlie don't surf, and the Combine don't dance:

A fix to make combine soldiers not dance. pic.twitter.com/aXDrhAUk5gSeptember 1, 2016

Zombies had to be toughened up after the shotgun's power was increased, to keep things in balance:

Cool little comment about rebalancing zombies after some changes to the shotgun happened. pic.twitter.com/XxM3ntFm4vSeptember 1, 2016

This is a stupid fix but it works, so whatever:

Might as well continue this thread since I found another good HL2 comment. Feel validated, everyone! pic.twitter.com/gYBCZJJvaRSeptember 2, 2016

There's only one fuck in the entire codebase, according to Wintergreen, and this is it:

Respect to Valve for having only one "fuck" in the entire HL2 codebase. pic.twitter.com/zKNF4ygjIvSeptember 2, 2016

Alyx Vance: Nice girl, handy in a fight, terrible eyesight:

Ah, good bit of HL2 Ep1 commenting (kind of thing they'd put in the dev commentary). On Alyx/barnacles in the dark pic.twitter.com/DhYyooJOyGSeptember 2, 2016

Some of comments really give a sense of what goes into making a game and keeping it intact. One in particular is actually a multi-paragraph telling of how the Strider minigun was accidentally nerfed in the Orange Box because of a bug in the original Half-Life 2 that nobody noticed. There's also a bit about the low violence mode, and references to the Combine Advisors, large sluglike aliens with creepy psychic powers who ended up not being used in the game. [Correction, sort of: I've been reminded that they did appear in Half-Life 2: Episode 2, but you didn't actually fight them.]

It's a fun bit of videogame history, and there are quite a few more than just what's embedded here. If you've got a few minutes, you can catch the entire thread on Twitter.

It's no secret that Overwatch takes quite a bit of inspiration from Team Fortress 2, but thanks to the efforts of Serbian modder Srpski eki (which translates to "Serbian Hammer") Team Fortress 2 is now taking a page from Overwatch's playbook. Srpski eki has recreated a version of Overwatch's escort map, Watchpoint: Gibraltar, that can be played in Team Fortress 2 through Steam Workshop and while it's not a perfect imitation, it comes pretty damn close.

As explained by Srpski eki , his version of Watchpoint: Gibraltar is in an alpha stage and is in dire need of some actual art as most of the level is comprised of featureless geometry. Despite its early state, eki 's map is still a fully functioning recreation and is surprisingly good at imitating Overwatch. Just about every mechanic of the escort game mode is carried over thanks to how similar Overwatch's escort mode is to TF2's payload mode. Even though the recreation isn't perfect, you'll hardly notice once the fight starts.

Because I have no friends, I went ahead and loaded a server with bots to play alongside me. They're far from a perfect substitute for humans but still do a great job of illustrating just how similar the DNA between Overwatch and TF2 really is. On offense, we were able to quickly push through to the final objective without much resistance because the enemy bots were too busy getting lost running around the level which was a bit disappointing. On defense, however, the round played surprisingly like Overwatch to the point of even having the same stalemates in certain parts of the map. Even despite the fact that TF2 differs from Overwatch in some fundamental ways, like needing ammo for your weapons, it's pretty amusing to see how the two are similarly structured. I captured a short video of a round against bots so you can judge for yourself. Also check out the gallery below for some close side-by-side comparisons between the Overwatch and TF2 version of the map.

The only real frustration would be the fact that, ultimately, Watchpoint: Gibraltar was built with Overwatch's heroes in mind. As similar as many of them are, I doubt that the map will be as balanced for Team Fortress 2's classes. Also, getting shanked by an enemy spy has made me wonder if there might be room for deceitful shenanigans with a new hero in Overwatch.

Right now there doesn't appear to be anyone hosting the map on the community servers, so if you want to give it a spin you'll need to round up some friends to play with or load up bots like I did. If you're unfamiliar with setting up bots in TF2, you can read this short guide on how to add them in your game. Because this is a custom map, you'll also need to generate a navigation mesh so the bots can move around, which isn't nearly as intimidating as it sounds. After you get the bots loaded, pop open the developer console (default is the '~' key) and type in "sv_cheats 1" and then type in "nav_generate". This will freeze the game for a bit while it does some calculations and then reload the map, allowing the bots to move around just be warned it's not perfect.

Now that I've played Srpski eki 's Watchpoint: Gibraltar, I'm honestly a bit surprised to see that modding Overwatch features into Team Fortress 2 hasn't been more of a thing, what with the two being so similar. The only other Overwatch mod I could find was this highly questionable haircut for TF2's Scout. Who said that people from Brooklyn couldn't be posh?

Following the rollout of Team Fortress 2's major Meet Your Maker update earlier this month, Valve has issued a patch to address some of the biggest problems with the new matchmaking functionality. The company acknowledged last week a lot of the most pressing concerns, and many of them are now fixed thanks to the new patch.

First of all, match leaving in casual mode will no longer incur a penalty, but to balance that out, Valve will increase the penalty in competitive mode in a forthcoming update. "The current system increases matchmaking ban times based on the number of abandons over a period of time," the notes read. "We are making a change to more quickly move serial abandoners into really long ban times. We will also subtract the maximum number of rank points possible, per abandon. The amount lost will be far higher than what could normally be lost in a completed match."

As for changes that will come into effect with the new patch, queue times should now take less than 90 seconds across the board, and empty player slots in in-progress games will now be filled up more frequently. Vote-kicking functionality has been added, and players can now select their preferred maps (though if they're added to an in-progress game, that won't apply until the next match).

Valve also outlined further plans for future updates, including ways to address griefing and high ping. The full update notes are over here. In the meantime, read Josh Wilkinson's impressions of the new matchmaking update here.

Matchmaking is a completely new experience for TF2. It combines the unfettered and wacky world of public play with some intense competition and small team sizes. From casual pubbers to competitive superstars, everybody will need to be aware of the differences and adapt to reach the top. Climbing the ranks from Fresh Meat to Death Merchant isn t easy, so here are some do s and don ts for both new players and experienced pros trying to excel in matchmaking.

Comp responsibly

The key to any good team is finding a composition that works. You need a balance of healing, damage, and mobility to win in matchmaking as in all 6v6. While the classic team composition in competitive TF2 has been one Medic, one Demoman, two Soldiers, and two Scouts, this has been turned on its head in matchmaking. Without any class limits or weapon bans, teams are free to run multiple Medics, while classes like Heavy can run amok with items like the Gloves of Running Urgently. The prevalent strategy at the moment in matchmaking is to run two Medics, two heavier classes such as Soldier, Demoman, or Heavy, and two Scouts. This gives your team a great balance, but teams can still succeed with a huge variety of compositions.

The most important thing to remember is that you need a Medic if nobody else is stepping up, take initiative.

Study the maps

If you re a diehard casual player dipping your toes into competitive TF2 for the first time, you ll need to learn the maps. Matchmaking omits maps suited for large teams (like Goldrush, 2fort, or Badwater) even though they are excellent in public play. Instead you ll find yourself on capture point maps such as the newly official Sunshine and Metalworks, along with Gullywash, Snakewater, Foundry, Granary, and more. Attack/Defend and Payload maps such as Gorge and Swiftwater are also featured, so it s important to have a good grasp on the maps before playing competitively.

Learning the names for areas as well as discovering all of the flanking routes is a necessity you should also use your new knowledge to flank your opponents and attack from alternate routes to gain an advantage in games.

Practice new skills

For any newer players, this is the perfect time to learn some advanced techniques. TF2 is a game with a lofty skill ceiling and many techniques that the best players use aren t obvious when playing casually. Matchmaking has also drawn in a broad spectrum of players with varied experience; if you want to get to the top then you ll need to know how to rocket jump, airstrafe, dodge, and aim like a pro. This will help you move around faster than your opponents, avoiding their shots and continuing to deal damage. While the game itself doesn t have tutorials for these skills yet, there are a wealth of guides online over eight years of them! With practice these techniques can become simple and the advantage they give you is staggering, especially in matchmaking where mobility is so important.

Don’t lose track of ubers

Due to its small team sizes, matchmaking revolves around the use of Ubercharges. Don t lose track of them, as your positioning and decisions to attack or defend should be based on which team has uber. Medics are the most important class in the game for both their healing and their Ubercharge, which give a huge advantage to the team. They can allow you to push through choke points, attack sentry nests, pull off a clutch defense, or destroy the whole enemy team. Be aware though that the same can happen to you, so it s important to roughly keep track of the enemy Medic s percentage as well as your own. Every 40 seconds, a Medigun uber can be built keep that time in your head and have a healthy respect for the German doctor and his patients.

Tweak your loadouts

All of the weapons that we in the TF2 competitive scene decided to ban are legal in TF2 s new matchmaking mode. If you believe a weapon is unbalanced and needs a nerf, there s no better way to demonstrate that than to use and abuse it. Using incredible individual weapons such as the Crit-a-Cola can be devastating, but also keep the synergistic weapons in mind such as the Disciplinary Action. Some classes practically require unlocks to be effective: the Reserve Shooter and the GRU spring to mind. Utilizing the right combination of these powerful unlocks can increase your team s abilities even without altering the composition.

Switch it up

Don t stick to a single class in matchmaking. While you may have a main that you love and adore, matchmaking is the perfect opportunity to test out your skills on a variety of classes. Only a few classes are useful all the time in matchmaking, and one of the core concepts in the game is switching up your classes to keep a good team composition for the situation. This doesn t mean you should play a different class every life, but like Overwatch, be prepared to switch it up to fit your team. The team needs mobility and damage at a midfight, a tanky defense when on last, and a lot of balance in between. Picking a class that fits those roles is a great way to start thinking about how you can best help your team to win.

Josh Sideshow Wilkinson has more than 7,800 hours played in TF2. In the last six years, he's climbed to the top of the competitive scene, placing 2nd in Europe last season with his team Perilous Gaming. Sideshow is also a writer, caster for teamfortress.tv, analyst, and tournament organizer.

Every game is ambitious. It s not easy to turn a beautiful idea into a finished, playable game as developers have said time and again, sometimes it feels almost impossible. As miraculous as finishing any game might be, not all games are created equal. Some stretch the boundaries of technology to their breaking point. Others take a leap into the unknown with new design schools, often so effectively that years later, it s hard to remember them ever having to be invented.

Think, for example, of Monkey Island s Three Trials structure, as used by almost every adventure afterwards. Or its sequel s Four Map Pieces , as later picked up by BioWare. And sometimes, both art and science combine to push the envelope and we get something truly, impossibly special. Here are our picks for the top 20 ignoring the very early games that had to prove computers could handle gaming at all.

For more on some of the most monumental games ever to grace the PC, check out our feature on the most important PC games.

King's Quest (1983)

For the longest time, adventure games were where people looked to see the latest innovations. King s Quest set that bar early on, jumping from simple text and pictures to 3D environments, huge worlds, and a fairytale land of mystery to both wander and wonder at. Admittedly, the last part was helped by some dreadful puzzles. King s Quest was originally commissioned by IBM as the showpiece for its long-forgotten PCJr system, but the series would go on to demonstrate just about every major technological advancement for the mainstream: ADLIB sound, VGA graphics, full speech, and high resolution. 3D didn t work out so well, but until that point, King s Quest was where many players went to get their glimpse of the ever-advancing future.

Commander Keen (1990)

If you want to experience pure hell, try the average 80s PC platform game. Long before making Doom, the team that would be id Software wanted to prove that the PC could handle experiences that played as smoothly as dedicated consoles. Commander Keen wasn t just a fluid experience by the standards of the time, but a fast one, with pogo-jumping, shooting and big levels to explore. Looking back, it s hard to appreciate what a development it was, but we re talking an era where games like the original Duke Nukem (or Nukum either way, the one who wore a pink suit and watched Oprah) were constantly being held up as the PC s answer to Mario. Commander Keen didn t qualify either, but it paved the way for many sequels and the formation of id itself.

A bit of bonus ambition: before making Keen, id tried to convince Nintendo to let it port Super Mario Bros. 3 to the PC by building a working demo (in their off hours in a single week, no less). Nintendo said no thanks, but you can see footage of the demo here.

Maniac Mansion (1987)

If you made a game like Maniac Mansion right now, people would still rightly call it ambitious. A choice of seven characters, each with their own skills. A non-linear adventure with five different endings depending on choices and characters. Real time elements, like ringing the doorbell and having a character come downstairs to check on it. Puzzles involving multiple characters in different rooms of the house or simply the option to do things like put a kid in an empty swimming pool and then fill it back up. And on top of all of this, Maniac Mansion brought the world the SCUMM system (Script Creation Utility For Maniac Mansion) that would define about half the adventure game market for the next decade. All of this, in 1987. Few adventures have ever done so much.

Ultima Underworld (1993)

Like most of the games on this list, Ultima Underworld is a fusion between ambitious technology and ambitious design the design side specifically being to take one single dungeon and try to breathe life into it. To add nuance to its different races, there to be talked to instead of just beaten up. The Stygian Abyss wasn t just a battlefield. It was a fallen community. A place to live in. The experience of being thrown into a dungeon and just expected to survive.

What really sold it though, if your PC could run it, was the technology. Before even Wolfenstein 3D, Ultima Underworld offered a full 3D environment complete with slopes, lighting effects and more, in a bit of technology that could only have been more impressive if well, the viewing window had been a bit bigger. Underworld 2 greatly increased the scope of the game, visiting other worlds and making it a bit easier to see, but what the first one managed remains a technological victory worthy of any heroic age.

Ultima VII: The Black Gate (1993)

Get used to seeing the word Ultima. Ultima VII came out in 1993, and still games like Divinity: Original Sin measure themselves against its success. Its biggest success was creating a living world, where peasants went home at night, weather blasted the world, your companions had to be fed, and, yes, where you could get some flour and water, mix it into dough, stick it in an oven, and get your own deliciously crispy bread. On top of this was an incredibly mature story that continued the series love of more advanced storytelling than most games of the era (previous ones having tackled racism, the perversion of good, and the quest for a hero worth being called one) with a complex tale of good intentions subverted by an otherworldly being of pure, but incredibly smug malevolence.

Ultima Online (1997)

Last time! Where Ultima VII brought a living world to single-player RPGs, Ultima Online brought it to multiplayer. It wasn t the first MUD or MMO, but most of them followed the Diku model popularised by Everquest: go forth, slay. Ultima Online wanted to create an actual world, where players would gather resources, craft houses, become shopkeepers and more, with hero just one of the many careers available. It wasn t without its problems, the first of them being the discovery that given a world to explore and exploit, players will typically turn it into a survival of the fittest Hell. But, its scope, its potential, and the joy of it when it worked created an epic experience that s still running today, and stories like the assassination of Lord British that will never cease to amuse.

The Elder Scrolls II: Daggerfall (1996)

The second of the Elder Scrolls games asked one hell of a question: could you make a world with over 750,000 characters and a map the size of Britain actually feel like a world? We re putting this one here instead of Elite, partly to ring the changes, but mostly because few procedural games have pulled it off so well enough political relationships, guilds, interesting stuff to discover, and cool mechanics like being able to get turned into a werewolf or vampire.

It s not that difficult to create raw space. Daggerfall s own predecessor Arena offered even more. Its sequel, Morrowind, did what most games tend to, and hand-crafted a far smaller area in intricate detail. But for a moment with Daggerfall, we had a game that showed you could be epic, procedural and interesting, without simplifying everything down to the ASCII style of Rogue or putting all the impetus on the player to pretend that there was more going on behind the surface than was ever going to meet the eye.

Duke Nukem 3D (1996)

While another case of a game that s not aged all that well, Duke Nukem 3D was the game that took FPS action out of military bases and sewers and relocated it to city streets, cinemas, and other more realistic locations. That plus a complicated scripting system to blow them all up, clever tricks to fake a 3D engine (even though it was only 2.5, much like Doom) and endless imagination took Duke from being a moderate shareware star to the highest tiers of game characters. No wonder the world was willing to wait so long for Duke Nukem Forever. Even if it wasn t worth it, in the end.

Outcast (1999)

The PC has never really had its own Legend of Zelda. Action. Exploration. A whole new world to explore. Outcast is arguably the closest its come.

A graphical powerhouse of a game that immediately impressed with its freedom, with the AI of its characters, with the glorious effects in everything from jumping into water, to your personal scanner rippling gridmarks across the scenery. There was only one problem. It was all done with voxels at a time when 3D cards were finally allowing for decent polygonal worlds, putting all the work on CPUs that couldn t handle it. If you could play it, Outcast was an unforgettable experience. Too bad for most people it was one that had to wait until the GOG version that finally made it run, long after its prime.

Grand Theft Auto 3 (2001)

It s easy to dismiss the sheer effort that goes into creating a city. After all, we ve walked, run, driven and carjacked around so many. GTA 3 wasn t even the first, with racing games in particular having set the pace. But could you get out of the racing cars and ramble? Enjoy a pumping gangster soundtrack? Run around with automatic weapons and go on missions with a huge cast of crazy characters? Just sit back and listen to an hour of talk radio? Nope. GTA III was magic, and so many sequels on, it s still raising the bar for what virtual cities can and should be.

Give or take a few terrible cock jokes, anyway.

EVE Online (2003)

Ultima Online intended to let players call the shots. It didn t quite work. With EVE Online however, CCP had the courage to actually let it happen, creating one of the most talked about online games of the last few years. Tales of empires at war, of con artistry on a scale that would make Count Lustig blink, the epic sagas of backstabbings and betrayals that no other game can match. CCP likes to describe EVE using the phrase EVE is Real , and while there may not be any starships flying distant galaxies under your favourite forum s command, they still have a point.

Civilization (1991)

All of human history in a single game? There s not much more to be said, really. As achievements go, the only bigger one would be making it one of the greatest games of all time. Not to cast aspersions on the likes of Elite for creating a universe in slightly fewer bytes than the average person would make in a toothpaste and peperami footlong, but the thing about space is that it is mostly empty. Just saying. The world however, in as many ways as you can imagine? That s ambition, even if using it educationally does mostly teach people never, ever to mess with Gandhi.

Deus Ex (2000)

Real world. Real conspiracies. Where do we even begin? Deus Ex not only set out to create some of the most realistic real-world locations we d ever seen (not a tautology the games before hadn t exactly done a great job most of the time), but also turn them into nothing short of a psychopath s toolbox. Multiple paths and solutions. Characters who reacted to your decisions. Tiny decisions determining who lives and who dies. All wrapped in some of the best writing and wrapping the PC had known up to this point. There s a reason why so many years on, it s the original Deus Ex that still stands out as both one of the greatest games ever, and the template of a dream for future immersive simulators to study at the feet of as they try to surpass it.

Dwarf Fortress (2006)

Simulations don t get any deeper than this. Literally, or figuratively. Dwarf Fortress or to give it its full title, SLAVES TO ARMOK: GOOD OF BLOOD, CHAPTER II: DWARF FORTRESS is an ASCII gem best summed up by its creator saying in 2011 that we shouldn t expect version 1.0 for at least twenty years. That s what you get in a game so crazily detailed that a cat can go into a tavern, pick up spilled alcohol on its paws, wash itself off, and get drunk. This was never intended behaviour, just the sum of smaller sub-routines coming together and making their own reality. In retrospect, that twenty years to complete doesn t sound so much at all.

Half-Life 2 (2004)

In a way, Half-Life 2 s most ambitious part isn t even in the game. Valve had an idea for a new store, called Steam . You might have heard of it. Half-Life 2 was, if not its Trojan horse, then its vanguard. You wanted to play the best FPS ever made at the time? Then you got it through Steam. And that worked out pretty well.

Even if you ignore Steam, Half-Life 2 reinvented the shooter with its focus on physics, with every chapter introducing new mechanics and new exciting concepts like the gravity gun or playing point-defense with turrets. It also created a continuous world like no other, putting the final nail into the coffin of games that didn t prize a sense of presence as well as place in their shooter campaigns. Much copied, but still rarely bettered, Half-Life 2 set out to be both the best shooter around, and its next great leap forwards.

Planetside (2003)

Some games just shouldn t be possible. Even knowing the technology that powers them, the epic battles of the Planetside series have always had a degree of magic to them. For the handful of players lucky enough to have a system and connection that could handle it, heading out into one of Planetside s huge battles is a defining moment in games. For the rest, it says a lot that it still felt just as impressive when Planetside 2 rolled along only a couple of years ago. 5v5? 12v12? That s all well and good. But an explosive, expanding, all-access battlefield where the war never stops? That s military action with a little sorcery mixed into the formula, even today.

Spore (2008)

It failed. Yes, we know. It failed. But this is ambitious games we re talking about, and few games shot higher than Spore. Leading a tiny organism through every stage of life. Constructing it using the surprisingly powerful and fun editor. Sending it out to meet other players aliens in a great throbbing galaxy full of freshly created life. That may have been the point where the charm ran out, but the open-ended action and procedural generation and early focus on user generated content that led up to that point still stands out as a technological, if not gaming success.

Crysis (2007)

"But can it run Crysis?" was a relevant joke among PC gamers for at least three years for good reason: well after 2007, Crytek's shooter could still bring CPUs and graphics cards to their knees. Crysis took Half-Life 2's early use of physics and applied it to a dense, free-roaming world. Being able to shoot a tree, watch it fall over, and then shoot the trunk into smaller pieces was revelatory players gladly gave up framerate in favor of insane graphics and physics processing. Cutting edge AI and the systems-driven sandbox gave Crysis the depth to match its insane graphics, and no shooter since has managed quite the same combination of wow and substance.

DayZ (2012)

From its beginnings as a popular mod, DayZ spawned one of the most popular genres in gaming today. The framework for this multiplayer zombie survival game was Arma 2, up until that point one of the most ambitious simulation games and a bastion for fidelity and scale on PC. DayZ built upon Arma 2 s ambition, borrowing and later adapting its 225 km2 terrain, Chernarus, which was created from satellite-modeled slices of the Czech Republic.

The month that DayZ caught on, creator Dean Hall was already laying out incredible plans about features he wanted to add, as he told us in an interview. Underground structures. Dog companions. Realistic disease systems. A couple months later we were hearing about destructible terrain and player cities. Part of Hall s stated approach was to experiment with big, bad ideas, but the reality of implementing them quickly in Arma s Real Virtuality engine for DayZ proved to be a massive challenge.

Outside of these early technical roadblocks, as a multiplayer game DayZ was uniquely trusting. The systems that DayZ inherited from Arma granted it some depth, and being dropped into a massive, hostile environment with no instruction empowered players to tell their own stories, often through surprising, weird interactions with other survivors.

Our 2012 mod of the year remains in Early Access after arriving on Steam as a standalone game on December 2013.

The Witcher 3: Wild Hunt (2015)

It s amazing to think that in just three games, CD Projekt Red has gone from unknown studio to absolute top-tier RPG developer. The Witcher 3 is their masterpiece, from the hand-crafted world to the sheer number of characters and plots. It s a game that excels on every level, from scripting subtle enough for a character to break off combat when they hear your name, to the global nature of some of the most amazing graphics and scenery in any PC game ever, and the sheer artistry of just about every major quest or aside. You never know what s coming next, from the teary humanity of the Bloody Baron s agonising storyline, to a gaggle of Witchers drinking too much, dressing up in drag, and drunk-dialling wizards across the whole continent.

Star Citizen (20XX?)

No, it s not out yet. It doesn t matter. Chris Roberts play to create the ultimate space game already qualifies. Elite style action combined with a dedicated, AAA Wing Commander-style campaign starring Mark Hamill. First person action aboard ships. Deep space exploration. A persistent universe allowing for company, or the solitude of the stars. A crowd-funded budget of $117,259,371 and counting, with players happily putting down real money for in-game ships and unlocking features like pets and modular ship designs and new AI characters to scatter around on planetside environments. If it s not the greatest game ever, expect literal, physical riots.