For general feedback about the game.
Steam SupportVisit the support site for any issues you may be having with the game or Steam.
If you wish to watch this video, you will need to Download the Flash plugin.
Since the release of Meet the Sandvich in August, several people have asked us how something we'd boasted would be "our magnum opus," "over four hours long" and "make Citizen Kane look like something dumb a complete idiot would make," ended up being one un-dramatic minute spent inside a refrigerator.
Now that more than a month has passed and the wounds have healed on Meet the Sandvich, we can finally give you an honest account of this aborted 240-minute epic, and by that we mean point fingers: It was all the bean counters' fault.
The suits took issue with every brave, authority-questioning page of our Meet the Sandvich script-specifically that there were supposed "similarities" between it and the 1987 action film Predator, and more specifically that it was word for word the 1987 action film Predator.
"Could you explain to us how your script is in any way different from Predator?" asked one of the suits.
"Predator takes place in Guatemala," Erik Wolpaw explained, using his I'm-explaining-something-idiotic-to-a-child voice. "Meet the Sandvich takes place inside a refrigerator in Guatemala."
"Where you can hear Predator happening outside of the refrigerator," clarified Jay Pinkerton.
"Also, Predator is two hours long," Erik continued. "Ours is four hours long."
"Right," said a suit, "but only because the script includes the first half of Predator..."
We nodded.
"... then all of Road House, and then the rest of Predator."
More nods. For a bean counter, this guy had really done his homework. Much like Dutch, when Dillon explains to Dutch how the Predator uses the jungle to its advantage, these suits really "got" it. We let our defenses down.
"Our script is pretty much exactly like Predator," Jay agreed.
"We took the script for Predator and put our names on it," Erik added, in case this hadn't been clear.
"Also, we had to paste the script for Road House there in the middle," Jay said when it suddenly occurred to him that business types who don't understand the creative art of writing might think merely writing your name on the Predator script isn't a full time job for two people. With everyone apparently now on the same page, Jay began working the room and shaking hands.
As it turns out, we'd been led, like the Predator by Dutch, into a trap. Instead of sharpened bamboo and weighted jungle pulleys, though, this was a trap of words, and instead of the Predator, it was us.
The suits demanded we give them the script so they could shred it. Just like in Predator, Jay doused the script pages in mud to conceal their heat signature from the suits. Unlike in Predator, this did nothing. The script was confiscated and destroyed. This created a small problem for all the Team Fortress actors we had standing around in the recording booth, watching us through the window with increasingly visible agitation.
"Hello?" barked Rick May, the voice of The Soldier. "Are we doing this or not?"
"Oh man, we are going to get fired now," Jay said. "We'll have to get jobs! We'll have to..." Jay tried to think, failed, then gave up in frustration. "What do people who aren't writers do?"
"I think they... sell things," said Erik, clutching at fuzzy images of people he'd walked past on the street on the way to nice restaurants. "Shoes and watches. From behind tables." He remembered his butler, who wasn't a writer. "Oh! Or they serve things."
"What are you two idiots talking about in there?" asked Gary Schwartz, the voice of the Heavy.
Jay grabbed the intercom mic. "Shut up, please," he observed.
"Anyone else notice this script is basically Predator?" Nathan Vetterlein, the voice of the Scout, asked. "It's not just me, right? This is word for word Predator."
"You think you can do any better?" spat Erik, a cloud of sandwich chunks hitting the sneeze guard they put between actors and writers in professional recording studios.
"You think what we do is easy?" screamed Jay, sending a further cascade of sandwich onto the sneeze guard. "Where would you have put the script for Road House into the script for Predator, smart guy? At the beginning? Because then it's just two movies," he said, not even believing how stupid someone could be to do that. He put his sandwich down to better illustrate the point with his hands, using his open palms to signify two different scripts, then bapping them together to further illustrate how stupid that would be.
Just as Jay's sandwich hit the table, a light went on in the refrigerator in Erik's head. "Sandwich... Sandwich! Say..." He punched the intercom. "Hey, guys? This is gonna sound crazy, but I have a new direction I want to try. Who's seen the movie Predator?"
"I have!" Jay exclaimed, thrusting his hand into the air.
Erik began to pace. "We open on a chopper. Inside sits Dutch, muscles bulging thr—"
"Look," Rick May interrupted, "how about we just improvise something?"
"Improvise..." Erik said, stretching the "s" out thoughtfully while staring at the ceiling.
After a minute of Erik saying "sss" followed by a gasping gulp of air followed by a few minutes of silence, Rick May tapped on the glass and asked us if we wanted him to tell us what "improvise" means.
It turns out "improvise" is another word for "the writers just watch," which sounded good to us. Better yet, the improvising was a big success, and led to the dialog you heard in Meet the Sandvich: rude, raw and spontaneous, like freestyle rap (or, as Valve writer/amateur freestyle rap impresario Marc Laidlaw refers to it, the poetry of the streets.) Here's a sampling below of some of the stuff we didn't end up using:
Heavy:
"What's that, Sandvich?"
Scout:
"Gimme back my legbone!"
"He's like a big shaved bear!"
"Pain!"
"I regret everything!"
Soldier:
"We got him by the short and friskies!"
"Don't do it!"
"You're writing checks your butt will find uncashable!"
"You only get one!"
"You call that killing me?"
"I will pay you all of my money!"
"Half the blood in your body!"
After an hour of high quality stuff like the dialog above, the writers had gained enough confidence to improvise some of our own. Here's a line we improvised ourselves from the movie Predator:
"I don't have time to bleed!"
Another improvisational gem, courtesy of Valve's award-winning writing staff, unilaterally tag-team writing alongside the powerhouse writers of Road House:
"Pain doesn't hurt!"
Here's a line we improvised that almost did make it into Meet the Sandvich, and was only removed after a non-writer dropped by our office and pointed out that we hadn't actually written it per se, and in fact had stolen it outright from The Simpsons:
"He's already dead!"
The point for all you young writers is just this: if the constraints of your current project don't specifically forbid it, your best writing will always result from writing your name on the Predator script. If, for whatever stupid reason, this isn't an option, improvise. Improvised dialog feels more real because it's lived. More importantly, it requires no actual writing.
Work on the next class pack is gearing up, a little slower than we'd like. In the meantime, I thought I'd post some answers to the common email questions we receive.
How do respawn waves work? Is my respawn time affected by my performance? Why do they exist at all?
Respawn waves occur on regular intervals, based on the map settings. Most of our maps use a 10 second respawn wave time. That 10 seconds is then modified by the map state, generally reduced for the team that controls the most capture points. Each team's respawn wave time is then scaled down if the team has less than 8 players in it, to a minimum of 5 seconds if team has 3 or less players in it. When you die, you are assigned to the respawn wave after the next one. So if the respawn wave time is currently 10 seconds for your team, you'll respawn somewhere between 10 and 20 seconds from your death. Your individual performance doesn't affect your respawn time in any way.
We found respawn waves were a good solution to several problems.
Well, it's been a couple of weeks since we updated the blog, and we thought it was time to let you know why we've been so quiet. The last couple of weeks we've actually been moonlighting on Left 4 Dead. While the L4D team has been focusing on finishing up the core cooperative gameplay, we've been helping them by working on their Versus mode. Versus is the recently-announced competitive mode where one team plays the Survivors and one team plays the Infected. We've learned a lot while working on TF2 in the last year, and we were able to apply some of those lessons to Versus mode. As a result, we've just finished adding critical hits, respawn waves, flawless auto-team balancing, and facestabs to L4D, while removing anything that looked remotely like a grenade.
I kid, I kid.
Now that we're wrapping up that work, we'll be returning to TF2. We've got a bunch of exciting changes coming down the pipe for the next class update. Feels good to be home again!
The path to arriving at a final design for a character is sometimes a long and winding one. This was especially the case for the Demoman and we thought it would be interesting to step through that process and shed a little light on how we arrive at our final designs. Early on in the process, soon after we decided that we were going to start a new, more stylized direction for TF2 it opened up a whole range of possibilities in terms of what that meant visually. One of the first (albeit short-lived) ideas that was being thrown around was to try to recreate a look of claymation for our game. The idea was to make the world look like a miniature set built for clay figures which animated and gibbed into chunks of clay.
A quick set of concepts were made to test the idea but we quickly realized that though the novelty of such a direction would be interesting, it didn't fit some of our higher level goals in terms of what kind of product we were trying to deliver. However, we did want to hold onto this concept of simulating real world artistic materials in the game and this led to the more painterly look which eventually shipped with the product.
Now that we were back to the drawing board an effort was made to first think of these classes as full characters with a history, distinct personalities and even nationalities. Short biographies were drafted which helped the artists concept characters which they imagined would match these biographies. For the Demoman, although the angry Scotsman was already an archetype common in media we decided to embrace that. The short bio was:
With this bio to springboard from and constant meetings between the artists and feedback from the team, a number of concepts were created:
As we were designing the characters, a stronger emphasis was simultaneously being placed on readability of the classes when playing the game, and so as you can see in the examples above, the silhouette of the Demoman was becoming more and more distinct. The first few concepts didn't really separate him out from the rest of the classes but as he was refined the more substantial upper body shape helped make him recognizable immediately.
At this point, the 3D model for the Demoman was being built and we were preparing to move on. However, there were still some who didn't think he was quite there. A nagging feeling that perhaps the embracing of the archetype was a little to generic. Perhaps a little too much like a groundskeeper in Springfield. We needed a twist to make him more interesting. We didn't want to lose his personality but felt like visually he needed something that broke the cliche. A suggestion was made to have him of African descent and we ran with it.
With a few modifications to his overall shape we soon arrived at the final concept and what you all see and play in the game today:
We decided to create a new TF environment after sensing some desert-fatigue, both internally and externally. The new theme had to be a departure from the current look without being so different that the characters felt out of place in it. We also wanted to leverage some content from our existing environments, since crafting every asset from scratch would mean we'd be moving the release beyond Valve time and into the realm of geologic time. Given these goals and constraints, we settled on the Alpine theme for two reasons: Its green foliage and blue-gray rocks let us introduce a new, cooler color palette to the game, and the lumber mill / mining camp buildings fit our existing set of structures. For Arena Lumberyard, the goal was to establish a clear difference between the two sides of the map. One side climbs up toward a large mountain peak, and the other dips down into a vast valley. We're excited to have a new visual theme to play with, and we'll definitely be using it in future maps.
