- intro / credits
^ session name ^ I've been you, me ^ how to get the most out of technical art
^ me: 20 years in games ^ almost 10 years with this title ^ still, i'm at a loss ^ TA is a new field - 2006 article - 2009 trip
^ Setting up the talk is hard ^ Because TA is so diverse
^ Will Wright hook
- I really wanted to do something similar
- because TA is hard thing to pin down
^ he'll give you a bunch on the soviet space program
^ or how mitochondria provides energy for your cells ^ but it all comes together in a beautiful whole
^ so, I did what I'm sure somebody in this room does and tried to google up some pictures I could use to illustrate the idea ^ This a tricky job because it's hard to describe what we do.
^ he's resourceful ^ we specialize in improvisation and making do with what we can.
^ TA's love solving problems — we are usually self-selected — for stubborness and curiosity!
^ MacG is a scientist, he likes to get to the bottom of things. so do we. ^ Its why we make good debuggers ^ segue: but macg is a primarily a technical person; that misses an important part of who we are. We're not just techie, we're also artists
- Leonardo is proof idea that you can be technical and creative
^ We all know he painted the mona lisa & the last supper
^ But he was also an inventor and tinkerer. And he was entirely self taught, which is true for most TAs - This image works because we're crazy half-and-half hybrids + we can talk to artists and to programmers; #hybrid + Sometimes, jack-of-all-trades-master-of-none
^ now this is not a leonardo butit's a renaissance classic; its about putting two worlds together. Plato contemplates the divine and aristotle the earthly. ^ We do that too: bringing different worlds together + because we speak both languages, we are often the negotiators or diplomats
^ segue: But what about teamwork?
^ as hybrids, you could say we're dual-classed
^ - RPG quad - Emphasizes teamwork #service - Engineering = Wizard. The glass cannon - Design = Theif. Leaves work up to everybody else - Art = Fighter. In the trenches - TA = Cleric. The support class.
^ what makes TA's happy is keeping everybody at max health; adding buffs, spamming heals, and raising the dead - or at least, recovering lost files. We're here to keep you in the fight
^ improvisation and border crossing define the way we work - but providing those buffs defines what we do
^ a Good metaphor! but service means more than spamming heals — nobody likes playing as just a buff buddy
^ This is an interesting metaphor - Service which knows its own domain very well. #logistics - gives deference, but expects to be recognized for expertise
^ There is a lot of special expertise on both sides that's needed to make things come off ^ lady cora knows who can't sit with who ^ but she doesn't know where the silverwear is, or how many pounds of potatoes you need to feed twenty guests ^ the folks downstairs are not as glamorous as the lords and ladies but they have essential skills and knowledge
^ Game production is of course even more complex and has more specialist knowledge ^ There are a lot of issues in production which are not art - you can't make art without resolving those issues ^ This is a critical part of what TA's do: the try to makes sure the party goes well for the artists
^ you could do worse as a metaphor than the relationship between upstairs and downstairs as a way of describing how art and techart relate.... ^ the best service uses the talent and creativity of the both sides
- An improviser
- A tech / art hybrid
- A support specialist
- A logistics wizard
^ Segue: I almost wanted to use that as the final image - partly because it's funny - But I realized, at long last, that I already had the right image - it's a T-Shirt you can order from tech-artists.org - And that sums it up: - TA = "making artists more productive" - The most precious resource in any production is creative #brain_cells - we make sure they don't get wasted on BS.
- An improviser
- A tech / art hybrid
- A support specialist
- A logistics wizard
- Fights for the users
^ So what do we do? This is all of it.
^ segue: But let me make that concrete
- we all know that parts of our jobs are garbage
- come in, 3 hours of emails, meetings, personnel conflicts...
- you're not in the mood to make great art
- finite reserve of decisions
- don't waste them
^ It's why obama, steve jobs, etc... ^ Creative thinking is a precious, finite resource ^ finite resources
^ But we give our artists tools that look like this. This is the opposite of distraction free! ^ The difference between a TA and and a working artist could be summarized like this: ^ Artist sees this and says, "ok, what do I have to do to meet my deadlines" ^ TA sees this and says "this is ridiculous, we have to fix it"
Creative decisions are a finite resource. Don't waste them on crap.
^ If you don't remember anything else I say today, remember this. ^ and remember that TA's are here to help save you from wasting creativity on crap.
^ TAs are there to fight that fight on your behalf - to make smart workflows - to keep people focused on creative tasks ^ That fight is the hear of evererything we do. ^ Segue: for the rest of the talk I'll walk through the dev cycle..
- Preproduction
- Buildout
- Production
- Wrap
^ Segue: for the rest of the talk I'll walk through the dev cycle.. - Preproduction - Buildout - Production - Ship - I'd like to walk you through the way that TA's deal with these cycle, and how our role changes. - It's important to see that what we we do changes a lot over the course of the production - understanding that is key to making good use of what we offer.
- Pre-production: Planning
- Buildout: Building tools
- Production: Help and training
- Wrapup: Crisis management
^I'm going to show you how the things the TAs do map onto these four big cycles. ^ In part this is to show you the breadth of what we do ^ In part to help you see that a good TA org has to move with the times
^ quick prepro sketch
^ lots of agendas going on ^ segue: this is the time of maximum uncertanty. But the best remedy for uncertainty is not argument in a vacuum, it's prototyping
^ Tas love prototypig - our inner macgyver - Naya FPS story
^ segue: Ta's love prototyping ^ ...but not all arguments are prototypable
^ A lot of the prepro process is about figuring out how the game will make be made ^ - preprois tough for many artists - bad pattern - engineer bullying - TA's stand up more - ta's = stronger negotiators - Ben W Story - Leonardo side
^ segue: once all the negotiations are done, we also have to design the tools that will implement all those workflow ideas ^ Here's another way that TAs come in handy ^ if you leave this stuff to pure tech people, you get things like this...
^ TAs are good at worfklow design - They have production know-how + eating your own dogfood is good - TAs do; tools programmers usually don't - They've usually seen the results of mistakes close up! + bad decisions often land in our laps + We know the hard truth: - no handshake agreement lasts - everything is iterative - yesterday's stretch goal is tomorrow's front line feature
- Prototyping
- Negotiating
- Workflow design
^ So this is what you are looking for in pre-production: ^ TA's are looking out for the way the game is going to be made ^ both features and ways of working ^ this is the where the combo of art and science becomes important - Dhunt story
^ If your artists - experiment and iterate, - not afraid to try new things - are able to work together... ^ then you've done your job
- Transparent
- Efficient
- Collaborative
- Iterative
^ As customers it may be hard for you to evaluate prepro at the time ^ it's hard to calculate the cost of disasters avoided ^ But you can see if the plans laid in this period are the source of the tools and processes you use later ^ you can judge them by their fruits
Graphic: The Dictator
- Titles don't matter much
- But TA needs a seat at the table
^ - Is TA a department or group? - Personally I don't much care - if you're in a situation where TOE is important, you're probably already in trouble - However, heres one useful principle to keep in mind - You want the TAD in discussions where process is important! - new features - new workflows - scalability - The key thing that matter is making the team pay attention to artists time and how it is spent. - It it takes a title to do that, then great ^ needs graphic? Repeat the ponies?
^ this section could all be construction themed
^ - moving from pre-pro to production - things get real - time to redeem the promises - take of the negotiator hat, put on the programmer hat
Graphic. Some kind of engineering fail. ideally social-realist
^ - this is stressful time for us - we have to deliver, people are waiting - we have to do it fast - we have to do it reliably
graphic: Spiderman? Great power, great responsibility ^ - Ta's have not always done this - when we got started 10 -15 years ago this was pretty haphazeard - nowadays it's a critical path for the whole compnay - with great power comes great responsibility - Haywood's text editor - moral: can't be too safe
- Good Implementation
- Smooth Releases
- Fast fixes
^ you may not be able to evaluate the programming of your TAs ^ but you're in a great position to evaluate the rollouts: - smooth delivery - fast fixes - tested, bulletproofed - trust - TA's aren't always good at that
- Planned
- Not disruptive
- Quick shakeout
^ Trust themes: what you should be looking for is ^ nondisruptive releases ^ well tested and basically functional ^ Glitches addressed quickly
GRAPHIC: trust
^ all of this boils down to trust ^ Do the TA's provide tools that artists trust ^ and maintain them ^ and improve them ^ trust is the foundation of buy-in
^ seque: building trust critical in this next phase of production
Graphic, relaxation
^ for us this is a more relaxing time, but also really important ^ This the good time, when the team is really running along ^ new stuff happening, getting in game ^ artists are perfecting new techniques
Graphic, service with a smile
^ this is when the service side of TA really comes into the fore ^ for example, automatic error reporting ^ fast turnarounds on fixes
Graphic: zombie error dialog
^ Service means attention to detail * like making sure that every error is being logged and reported * artists never report bugs voluntarily * but if every error generates an email - and a visit from the TA... that builds trust
graphic: TA backlog
^ seque: not every bug can be handled directly ^ not every suggestion can be acted on right away ^ releases that don't mess with schedules ^ new feature requests being handled; backlog
graphic: Relationship theme
^ Segue: but all this is technical. The real key to service is relationships. ^ prodution is the time when we should be building relationships. expect tas' to be out there with the artists ^ demoing, explaining ^ pushing feedback to engineers ^ this is self interest: - better educated artists are better for us - time spent with users makes us better at what we do
Graphic, learned helplessness
^ Anet story
^ segue: we need to build those relationships so we know wha'ts going on ^ this is where wee really build the relationships that let us know how to do our jobs right ^ we need to be communicating well so we can learn from the artists ^ and stand up for them in other phases of the project
- Approachable
- Responsive
- Embedded
^ Good service boils down to relationships: - be friendly - be knowledgeable - be helpful - be responsive - above all, be there - >> keep the joke about "if being a grouchy genius is needed, we have programmers for that"
- It's all working
^ big questions answered, big tools delivered
__
Graphic: some kind of not too serious disaster relief pictue
^ - now is when shit hits the fan - all the guesses made in the dark get tested - many will be wrong - "rebuild all the physics message" - "quantize all the vertex colors" - "cnahge all of these p's to q's"
Graphic: first responders
^ TA is the first line of defence when things go wrong - it comes with the territory -but there are good ways and bad ways ^ We have good skills for dealing with emergencies - debugging - more stubborn and more inclined to pick at clues - technical background - "we can spot a convex hull" - ad-hoc automation - Undead tree Story
Graphic sucks to be us
^ from a management and personal standpoint, it still sucks by definition, it's not planned - hard to improve, because every crisis is different - high time costs - unpredictable ^ Still, it comes with the territory ^ it's a key part of what we do ^ however... there are right ways and wrong ways to deal
Graphic: Elephant and broom
^ It's good to know that TAs are there to help out in emergencies ^ that doesn't mean you can embrace bad tools and bad worfklows ^ and trust us to pick up the pieces - value the heroics - don't take them granted - gruntwork saps morale
^ - wes g story - don't be a garbageman
Graphic "no garbagemen"
^ too many companies choose the bad way - human wave attacks - drudgery - keeping people tasked out 100% - if you waste artists time on crap, you're doing it wrong ^ - if you waste all TA time on crap, you're also doing it wrong! - sometimes you have to do it anyway - but it's a mistake
- automation
- open file formats
- asset tracking
- Defensive development
^ What can yoyu do to help? ^ Good preparation helps a lot ^ Key weapons are + good tools + open file formats + asset tracking + Cynicism - "We just wont do that" rarely works - when it does, it's not always a good thing + Don't overschedule
- Balance shifts over time
- Good department is balanced
- Tech
- Relationships
- Emergency services
^ get the balance right - don't overdo tech - no ivory tower - don't overdo service - ta's must be innovating - don't overdo human waves.
^ redo the graphic with the right heads
^ tools = leonardo ^ pipelines = mr carson ^ support = the medic ^ emergencies = macgyver
graphic: different aspects of TA peak at different times
^ tools = leonardo ^ pipelines = mr carson ^ support = the medic ^ emergencies = macgyver
- TA is complex discipline
- look for the key values:
- smart workflows
- good service
- solid relationships
- crisis management
- TAs are there to help
- Use them wisely!
- The rest is details
Graphic: stack of steve jobs t-shirts
^ Creative thinking is a precious, finite resource + don't waste it on BS stuff + Focus on story, moood, character + Not on file names, check boxes - All boil down to one big thing: + Making artists more productive
Creative decisions are a finite resource. Don't waste them on crap.
^ pause here...