Greetings! so I gave the keynote at PAX DEV last month, and PAX DEV is pretty
secretive - they don't have video cameras. But I got permission to record the audio and post
the keynote speech on my youtube channel for anybody who wanted to watch the
keynote but wasn't able to attend pax dev. I talked about a bunch of different
topics and I hope some of them are inspiring or interesting to you so take
a take a look and let me know what you think! Enjoy. Greetings! Ok so my name is Ben Brode.
I worked for a Blizzard Entertainment for 15 years, the last ten of which I worked
on a game called hearthstone and the last two of which I was the game
director on Hearthstone. And four months ago I left Blizzard to start a new game
studio called Second Dinner. And since then we got enough funding to make our
first game, we hired a few super badass folks, and started working on some
exciting prototypes. And it's a really exciting time right now back at
the office so I figured I'd take a week off and come up to pax.
I'm really excited to be here today. My good friend and mentor Eric Dodds, the OG Hearthstone
game director, has told me every year how awesome pax dev is, so I'm very honored
to be a part of it. And I know part of the vibe of the event is to get nitty
gritty and share lots of behind the scenes, maybe not-for-public-consumption
data, but Second Dinner is not privy to any secret data right now, so feel free
to take pictures or notes or whatever. And I'm gonna record myself actually so
I could post this later maybe. So one of the hardest parts about giving a talk
for me is figuring out what is obvious and what is interesting and helpful.
I just have a hard time figuring out what other people know. And if I just tell you
things you already know, that is very low value to you. Although it might be fun to
listen to me try and get you excited about common knowledge.
For example, do you guys know if you take peanut butter and mix it with jelly, you
get an exciting new flavor?!
You already know that. Okay.
But I'm gonna do it anyway, so here's five things that are probably
obvious to you. Your games should be responsive. Protect low-skill players via
matchmaking. Make sure everyone shares a clear vision. These seem pretty obvious.
Keep players happy. Keep employees alive. okay
Well thank you very much. This has been a good talk. Okay my talk is called 'Little
things that make all the difference' and I'm gonna look at those five obvious
things and mention some little things that I noticed maybe some people aren't
thinking about or maybe mention some of the trade offs that may not be obvious
about those five things. So, obvious thing number one: your game should be
responsive. I think this is super obvious to everyone maybe except the people who
make gas station pumps. This is a picture of Rob Pardo talking about
Blizzard's core philosophies at GDC. I highly recommend this talk. One of their
core philosophies is "control is king" and I do think that Blizzard pays closer
attention to this than a lot of other companies. One of the things that Pardo
mentions in his talk is that this is a very difficult thing to get feedback on.
Players will definitely tell you if your controls are unresponsive, but they are
much less likely to notice if your controls are only slightly unresponsive.
So during the development of Warcraft 3 they discovered three frames of lag in
the mouse cursor and removed them. And that's a tenth of a second of lag and
that's just unlikely to be a big topic of feedback from players. And even though
it's almost imperceptible it actually makes a huge difference to how the game feels.
We have to specifically pay attention to controls and look for even tiny
optimizations. There's something I want to add to the
way that we talk about controls. I want us to make controls crispy. Crispy things
are awesome. They look delicious. They have a great crunch when you bite
into them, and they're fun to eat. But we use the word a little differently.
Here's how we define it: When controls, visual effects, and sound combine to
create a maximally satisfying interaction. This term was coined by our
principal engineer Mike Schweitzer. He's a huge fan of fighting games and that
genre is notorious for being super crispy.
Here's dragon ball fighterz: a game that has incredibly crispy controls. Hitting a
punch button gives you responsive feedback but also combines the visuals
and sound in a way that really enhances how it felt to press that punch button.
And this is the crux of crispiness: Using visuals and sound in
concert with controls to create a maximally satisfying experience. I think
the way that hearthstone handles the hand of cards is very crispy. Here's a
video. When you mouse over a card, the big view snaps instantly. I've seen some card
games that animate between that small size and that big view, and you might
think it's more crispy to add that animation, but actually I think it makes
it feel a little less responsive. This is the inherent tension between animation
and controls: they need to be instantly responsive and maximally satisfying.
For me, the thing that offsets that lack of animation between those two sizes is
that little slow animation after the card snaps to a big size where it drifts
up slowly. I think that makes it feel like it's more alive. One other thing
about crispy animations: don't let them get in the way of being able to play as
fast as your brain wants to move. Obviously Hearthstone lets you drag your
cards out as fast as possible regardless of which animations are playing - and if
you had to wait for the animations to complete before continuing to play the
next card, you can't have long awesome animations
because they would get in the way of whipping cards of your hand as fast as
possible. There's another metric of responsiveness which the time between
when a player decides they want to have fun and when they actually start having fun.
And if that length of time is too long they might opt for a lower
friction form of fun like browsing Twitter or watching Netflix.
It's basically like a Bing for the buck calculation. If I'm going to play for a
few hours then waiting a couple minutes isn't a big deal. Or if it's super super
fun then yeah I'll Drive over to Disneyland pay 200 bucks and wait in
line for two hours to ride Space Mountain. I do think this makes a
difference for everyone but if your game has been to be played in short bursts I
think it's critical. So, often the time to fun includes logging in, navigating the
in-game menus, finding a match, and learning the gameplay itself. But if the
beginning of the gameplay isn't immediately fun, it actually takes even
longer to get there. Or there could be additional steps before logging in, for
example when I would decide to play a VR game, I need to clear out the VR room,
boot my PC, plug in my headset and lighthouses, launch steam, then launch
steam VR, and then go through the rest of the flow. The friction is pretty high so
I only play if I know I'm gonna have a long play session for example an hour of
beat saber. Beat saber is amazing this is super fun. Like, it makes me feel like a
breakdancing Jedi. you have to try to beat Saber. OK sorry so I I want to talk
about some of the individual components of the chain to get in and start having fun.
the first is log-in speed and I just wanna call out some games that're just doing
an incredible job at this right now. I want to show you how fast clash royale
loads. This is from a cold start - it wasn't already running in the background, and
there it is it's playable. it's pretty amazing it's like 6 seconds. here's an
app my son likes called Metamorphabet. Here's how fast it loads. It's loaded
already. that's the actual gameplay, you can play it. It's obviously a whole
different beast because it doesn't connect to the internet you don't have to login but I
think it's still incredible again from a cold start. One thing that's a pet
peeve of mine is when it feels like developers are intentionally making me
wait longer to have fun, and startup logos sometimes feel like that to me.
I think probably some people are using these to mask some of the loading time
but not everyone is doing that. Does your game need menu animations? So, menu
animations add time to the critical time-to-fun countdown, but they can
help make a game feel AAA. Like, the way that these chains move and the
sounds that they make I think really contributed to how awesome this game
felt when it came out. They're not too long so I think it'd be a tough call whether
or not you'd optimize for getting into the game faster or making the animations
cooler. This particular game has pretty long play sessions so I think
that's part of the big consideration you'd make when
making a decision like this. Does your game need a story? Telling a good story and
having a great narrative is important. A lot of people do an incredible job with
narrative gameplay. But not every game needs a story. Like, you download a chess
game and before you even play the tutorial there's a cutscene about how
the enemy King has issued you a challenge and the fate of your kingdom
is at stake and the mysterious rooks have returned after years and what about
the prophecy?? If the narrative is not the reason that players are playing your
game it's possible it's just getting in the way of them having fun. So I went
over a couple things that I like to think about for shaving things off of
this flow but there might be other creative ways to get it as short as
possible and I think this is really a team wide effort. Your UI designers
need to minimize clicks and screens in the flow to play. Your client engineers
need to work some heroics to minimize load time and your animators need to
find ways to make the game feel awesome without necessarily animations
slowing down the flow. Or you could just make your game as fun as Beat Saber and
I will put up with any amount of bullshit to play it. Ok, obvious thing number 2:
Protect low skill players via matchmaking. So in the time before
matchmaking, you'd see some matches like this. And the big sumo guy here is a
player who is hyper skilled and the cute little sumo kid represents a player who
is very low skilled. And imagine being the little guy. You have no chance here
and if you're in the bottom 10% of players almost every game will feel like
this to you. And that sucks. You probably won't be having much fun. You'd probably
quit. You know what'd be better? If the little guy won more games. So most modern
games use some kind of skill based matchmaking to find fairer matches for
players. They assign a value to a player's skill (we usually call that matchmaking
rating or MMR) and match them against players with a similar skill value. And
maybe this won't surprise you, but players like winning. And when they win
they have more fun. And when they have more fun they quit less often.
But where does the trend of high winrate stop being correlated with lower churn
is it at 50% winrate or does it keep going? And I think winning is
winning. Players who win 90% of their games are having more fun than players
winning 50% of their games. So I said we wanted to increase the win rate of the
little guy here right but in this example it's a zero-sum game and those
extra wins means we have to give them those wins from somebody else -
giving extra losses to somebody else. Which means we have to lower the win
rate of the rest of the players. So when you pull the little guy's win rate up, you
have to push the win rate of everyone else down. Now I don't know what the
shape of this line actually looks like - I'm sure it depends on the game and how
impactful winning and losing is in your particular game - but as you move your
population more towards the middle of the red line from the edges, the people
on the left who were winning very few games are now quitting less
and the people on the right who were winning a lot of games are now quitting
more. So depending on the shape of this red line, this could still be overall
good for the churn rate, or it could be neutral, or even negative. So if the shape of
the line looks like this, so that maybe winning zero percent of your games means
you're 100% likely to quit but hey if you're winning ten percent of your games
maybe really fifty percent likely to quit, but if you're winning 90 or 100
percent of your games you're very unlikely to quit, because you're still riding
high after winning all those games, then normalizing win rates towards the middle
actually reduces the overall number of people quitting, so matchmaking seems
super beneficial. And this shape makes intuitive sense, right? If you're
winning a lot of games you're very unlikely to quit whether it's 70 percent
or 90 percent. But be careful about assuming this kind of thing. Figuring out
the correlation between win rate and players quitting might be very helpful
for your game as you consider how much normalization of win rates to do.
And this is a complicated issue. Obviously, new player attention is very
important, so you may not want to just throw players in to a system with no
matchmaking. You might want to have some other system that lets players take the time
they need to become skilled enough that they can have a better experience once
exposed to the broader player base. But there's some tricky things you
can do to cheat the system. And I experienced it firsthand pretty
recently. I downloaded PUBG on my phone and I played three games, and I won my
second and third game. And I was shocked, because I beat 99 other players as a
total scrub. So I posted about my victories on Facebook, which is already
pretty risky because that's kind of braggy, and a few of my friends pointed
out that in my first few games my opponents were entirely BOTS who were designed to
lose to me. So I ended up looking really dumb, and I deleted my post, but if not
for those friends I would still feel really good about beating those BOTS.
What they were essentially doing is making it not a zero-sum game. actually
battle royale games are I guess kind of like negative-sum games. They
introduced AI opponents that suck, to inflate the win rates
of the worst or newest players, without harming the win rates of the better
players. Another way to avoid the zero-sum problem is to try and match
players together who have different goals. so here's the sumo scene
reimagined like that. And one thing I've seen a lot of games do that is similar to
this is give out rewards for playing but not necessarily winning. Another cool
trick I've seen is to reduce the sting of a low win rate. So this game Duelyst
has a hidden achievement that you earn for losing three games in a row, so maybe
you get to convince yourself that you meant to do that.
Skill matching makes it hard to feel power gains. So another thing to keep in
mind about matching based on skill is that it makes it hard to feel when
you've gotten better at the game. Whether that's because you actually improved
your skill or because you've leveled up your character or maybe purchase some
new cards or items or whatever. Often when that stuff happens you get a
short-term boost and really feel good for a few games, but depending on how you
tune your matchmaking algorithm it'll actually very quickly bump your MMR up
and find you fair opponents again. So no matter how much better you get at the
game your win rate always stays around 50%. And for me at least it saps me of
some of my will to improve, because doing better just means I'll have to fight
harder opponents, and that kind of stresses me out. This is my friend Josh
Menke, who did a talk on matchmaking at GDC a few years ago, and this is a good
graph to show how fast modern algorithms can figure out a player's skill. And you
can see the comparison to ELO (the red line there at the bottom) and what this
means is that modern games have the capability to get you even matchups
after just about three to five games from no data, and almost instantly adapt
to changes in power level. So this is this is pretty obvious, but
the tighter your tolerance for skill based matching, the slower your
matchmaking speed, which increases the time to fun. And this could be an
important consideration based on how many players are playing your game,
how many modes you have splitting your playerbase, and what you think the
experience should be like in off-peak hours. So I talked a lot about the
downsides of skill-based matchmaking, but I'm actually not sure that its overall
bad. They are real upsides to using a system like this but it's important to know
what the trade-offs are. For example Hearthstone's Arena mode matches players
based on the number of wins in the current run, not on a persistant skill
rating and I think that makes it both a worse mode for new players and a more
fun mode for the most skilled players because skill directly
influences your win rates rather than it all being normalized to 50%. So we use
skill-based matchmaking because we want things to be more fair, but in a way
skill-matching could feel unfair, because working hard and improving doesn't mean
you get any better win rate for that work. Matchmaking on player skill is
optimizing towards 50% win rate. Is that the best thing to optimize for? Sometimes
50% win rates will be the most fun, and sometimes it won't be, and you have to
consider the overall effect on fun from all different player perspectives as you
tweak your matchmaking systems. Alright, obvious thing number three:
Make sure everyone shares a clear vision. Making games is super hard, and one of
the hardest things is making sure your team understands what your game is, what
your game is not, and why it's going to be awesome. I think everyone knows this
is important, but I think it's critically important, and this is one of the big
reasons I've seen game teams fail. If you don't agree on what you're building you
end up with a hodgepodge of stuff that doesn't fit together into a cohesive
product. Okay so we're going to do a little exercise together. Imagine you're an artist. And if
you're an artist, you don't have to imagine. And you're working on a new
zombie movie. And you're the artist responsible for concepting what the
zombies are going to look like in this movie. Try to imagine, you're the artist.
Start thinking what are the zombies gonna look like in this zombie movie we're making.
There are a lot of us working on the movie. Someone else is doing set design,
someone else is writing all the dialog, and we all need to be on the same page
to make this movie feel believable. But zombies are zombies, right? They're
zombies. How likely do you think it is that we're all thinking about the same
visuals for our zombie? So, clearly there are different interpretations.
So let's align on this. Let's say we're gonna make a gruesome, realistic
zombie. So now we should all be sharing the same vision. But wait! All we
know is that this is a zombie movie. And I have a pretty good idea of what a
zombie movie is like there's still a lot of room for interpretation in the
details. Is this a movie about the confusion and horror that happens at the
beginning of the outbreak? Or is it about how humans have adapted in a
post-apocalyptic world? The movie comes out really differently in both of these
visions. So let's make a movie about surviving in a gruesome, realistic,
post-apocalyptic zombie world. So, good. Now we should have a clear vision, I
think. Except that that's probably still too broad, even to start concepting
what zombies would look like. Is this about what life is like when you're
all alone and exploring the humanity of zombies? Or it's about what happens to
the political climate and the different nations of Earth? or maybe it's about how
daily life changes, or doesn't change, when death lurks around every corner?
Vision is multifaceted and even when you get very specific, choosing the
exact vantage point to explore your theme can be very nuanced. I've seen
meetings where everyone looks at each other and says what they think the
vision is, and then nods, and then we all go back to our desks and execute on
slightly different nuanced versions of that vision. So my first tip is to use
imagery. This is just a Google search for 'spooky forest' and spooky forest is a
wide range but when you download 50 images from the internet sort them into
'yes this is the right vibe' and 'this is this is wrong and here's why' you can
start to create a map so that everyone knows what is exactly okay and what isn't.
In my experience, people in the industry often use movies as their
reference points for vision. And I felt super funny
when I started in the industry because I hadn't seen movies like Indiana Jones or
Heat or Conan the Barbarian or Predator or Terminator or Superman or Batman or
Die Hard or Weird Science or Blade Runner or like a hundred other movies that
everyone references. I've seen all those movies now so don't be too worried. But
images or references to games or movies we've experienced in the past can help
provide a shortcuts to clear vision.
Discussing things from a high level is helpful, but using concrete examples
helps you figure out exactly where people have different views. So I've
often found that you need a check-in right after a team starts working,
actually executing on the vision, to make sure that there wasn't something that
you were not on the same page about. Start small. I think that a vision is
harder when there are more cooks in the kitchen. If you're trying to come up with
a top level vision for a new game, I think spinning off a small team to
incubate a project with the minimum number of people can help end up with a
higher chance of achieving a clearer vision that everyone is aligned on, and
then you can slowly add people once your imagery or prototype is clear enough.
Iterate on your vision pitch. Eventually you will need to impart your vision to
people who will consider spending money or time on your game, and your vision
needs to be clear and exciting by then. You need to play test your pitch on new
people just like you play test your game. As you explain the vision to people, pay
attention carefully to what excites them and what confuses them. And if people
don't understand, or aren't excited, try a different approach with the next person
you talk to. And this will pay off because eventually you'll be telling
people who are farther away from the dev team, like publishing groups, or regional
offices, and you'll really need the vision to be clear. And then when you
announce your project to the public, the more you understand what will excite
people the better your announcement will be. So obvious thing number four: keep
players happy. Wouldn't this be nice? I do think it is super challenging. It's
also probably most relevant if you're running a live service game. There are
too many different types of players, and you can't please all the people all the
time, but I have some tips I think could help. Become a teacher.
So this is important for two reasons. One, because the more knowledgeable your
player base the more likely you are to get high-quality feedback. And two,
because when players don't have all of the information, they are more subject to
the dunning-kruger effect so here's David Dunning,
and this is the biggest photo I could find on the Internet of Justin
Kruger. I went on his Facebook page and everything... he's not very public. And in
1995 these two guys heard the story of a guy named McArthur Wheeler AKA "The Lemon
Juice Bandit". And MacArthur believed that because lemon juice is used as a
component in invisible ink, that if he put lemon juice on his face, he would
appear invisible to security cameras. So he did that, and he went and robbed two
banks, and of course his face was not invisible, and so after the local TV
stations broadcast the security camera footage he was immediately identified
and apprehended. And when he found that he had appeared on the tapes he kept
muttering, "but I wore the juice...." Four years later, Dunning and Kruger published
this paper of detailing the relationship between confidence and experience.
So, here's a graph. Confidence is on the left, and Experience is on the bottom. And this
red line shows the ideal relationship between confidence and experience. As you
get better at something you should become more confident, right? So Dunning and
Kruger ran a scientific study to figure out the actual relationship between
confidence and experience. They gave people a test and they asked them how
they thought they would score on the test before they began, and then compared
that with the actual score. And here's what the results looked like. What's crazy
here is the folks with literally no experience were the most confident.
And this totally happened to me. I'd been working at Blizzard for five years
when I changed careers from producer in creative development
to associate game designer on the hearthstone team. And I was pretty upset
because I had worked my way up from associate producer to producer over
several years, and I felt like I was moving backwards. I complained about it a lot,
but resolved to prove how good of a designer I was. And it took one week working
under Eric Dodds to realize I was definitely associate level. And I stopped
complaining. So you get even a little experience you actually assume you're
worse than you are because you now realize how much you don't know. And this
effect can a really decrease sentiment because people with no experience in
your field often assume they know better than you and then get angry about you
not doing things the "obviously correct" way. So here's an example about how this
this played out for me. Players started asking this question: why weren't we
buffing bad cards? Wouldn't the game be more balanced and more interesting if
the bad cards were a little better? And this is this is a very good question, and
has the opportunity to spawn an interesting discussion. But because of
the dunning-kruger effect, some of the players that don't know all the
reasoning presume that they do. So they would imagine these as the only possible
answers. So imagine for a second those are the real reasons why we didn't want
bad cards. How angry would you be? I think rightfully angry. So here's where
becoming a teacher can help. You can expand the amount of information these
players have access to. And now faced with other credible possibilities
players have increased their total understanding of the problem and because
of the dunning-kruger effect this increase in knowledge actually causes a
decrease in confidence, which means even if they still believe you're greedy and
lazy they're less likely to be confident about those beliefs. I want to emphasize
this has nothing to do with how smart or stupid your community is or any
individual is. In any size of community there are people with a range of
experiences and some players already have a depth of
understanding about the issues, but everybody starts somewhere. I also want
to emphasize this doesn't mean that the developer is right or that the community
is wrong. I've often felt that after teaching people the complexities of an
issue that the community has made very compelling arguments and I've changed my
stance. For example, Blizzard changed their stance on things like additional deck
slots. But it's hard to have a well-informed
healthy discussion if people
aren't exposed to the full picture.
If you're a developer, engage directly.
People are more inclined to trust individuals than companies. This is a
tweet by Greg Street who's I think excellent of doing this thing.
I felt that creating a direct relationship with the community helped a lot, and when
an individual who has the authority to change the game says "we promise we'll
make this better" it has more teeth because there's personal responsibility
behind it. Developers can also speak better to some
of the nuance involved in the decision making. Make yourself visible. When you're
planning an announcement one thing I think helps us to immediately make your
presence felt in the comments. So here's us announcing Second Dinner. And there
I am in the comments making my presence felt. And often when you show up in a
discussion it feels like people are on better behavior. I noticed that in
threads where I engaged in the comments the sentiment felt much higher to me, and
I made sure that when there was a controversial subject I was there being
visible. Respond fast. The internet moves really fast if you don't get your
response into a reddit thread in the first 30 minutes you might as well not
post at all. Other platforms are a little less punishing to delay, but a lot of our
community was active on Reddit. One example was this bug that got reported.
We had just released hero skins and there were
players who felt the value wasn't high enough for the cost. And it was a really
hot topic in the community. And one player made a post saying that there was
a bug with the new hero skins. There was a character who called
another character the wrong name and it seemed careless.
And this played into the narrative that we didn't really work very hard on these,
we just kind of jammed 'em in there. And this post immediately shot up to the top
of the subreddit and got thousands of uploads. And many hours later, we realized
that it was fabricated. There was no bug. And when we came in and posted the facts
it was too late. The posts got almost no visibility. People are much less
likely to go back into a thread and reread the comments once enough time
passes. The damage had already been done. Misinformation is pretty common
obviously and without replying to clear things up it just keeps spreading.
I actually wanted to create, like, a hotline where people could text me and recommend
threads to reply to, so I could get in there really fast, because I think I made a
huge difference. Ask for feedback. So another thing to
keep in mind is the balance of power between players and developers is all on
the side of the developers. If a player wants you to change something, they have
to convince you to make that change for them. They can't do it themselves.
And it's a frustrating relationship to be in. Players have
incredible passion for the game and it reflects on them because they play it a
lot, so when when you're announcing a change also
make it clear to the community that you're open to feedback about it. It makes it go
down easier because it'll feel less like a mandate from God and more like a
friend floating a cool idea. And if it's not something you're open to feedback on
you can instead ask for feedback on the on the way of the announcement was
delivered. I felt like reaching out was always very beneficial and I wanted to
do it a lot, so I focused on reducing the friction between an idea that I had and
actually shipping a video or blog post. And one of the things I did was to buy
myself a microphone and a webcam, and I started doing videos from my house, so I
could iterate without wasting anyone else's time, or waiting for a slot to open
with the video team. Here's a pic of the setup I used for one of our videos.
That's my bed in the foreground, and my dog over there. I also spent a lot of
time discussing social media strategy with our PR and community teams to
really learn those crafts and eventually I grew in my confidence
about how to interact the community and could sense when there was a type of
communication that I would need back up on and which I could just pull the trigger
on right away. And that meant that there was less friction and a lot of times
when I wanted to post. Okay, obvious thing number five: Keep employees
alive. I guess you probably don't have a lot of control over this, but I do have one
thing that I think we should work on. Stop working so much fucking overtime.
So this is an article from last year "76 percent of game developers still labor
under crunch conditions". Here's another article talking about "the horrible world
of video game crunch". It doesn't have to be this way. Obviously things come up.
Sometimes this comes from a level higher than we can control in the organization
and we don't get a say. Sometimes your team only has enough money to last
through the end of the month and if you can't finish your next milestone
the game will get canceled. Obviously if there's a service outage it's all hands
on deck. There are some legit reasons to do OT, but as an industry we crunch
a lot, and some of that is preventable. Control the scope. A friend of mine from
the industry told me a story about a time when he was crunching on a game and
some of the design leadership had a meeting to discuss post-launch DLC.
It was a one-hour meeting and instead of generating new ideas for DLC, they
decided the ideas they came up with were too exciting and they wanted to jam them
in before ship. And this is really tough, because none of us got into the game
industry to make shitty games. We want to make freaking awesome games and we have
ideas that we know players will love. We have to make the tough call to cut some
of that exciting stuff to avoid overtime. And it's a hard call to make, and your
team will be sad because you cut an awesome thing. There's only a few
variables you can adjust and if you're unwilling to adjust the ship date or the
scope over time is the only answer. But you can adjust scope and the ship date.
Hearthstone shipped way after it's announced ship date without
single-player content, without spectator mode, and without tons of features that
obviously the team added over time.
Don't let people set micro-cultures of OT.
So, sometimes it's not coming from above but there's a culture that makes you feel
bad for leaving after eight hours. Like, you get up to go and people are like,
"oh hey are you leaving already? What time is it?" And one thing you could do is just
tell people that it's not okay to work OT. Have HR give you a weekly report on
who's working OT hours and then go to each person on your team who is working too
much and tell them how important it is to not burn themselves out.
if you must crunch consider focus bursts. So if you need to crunch,
make sure you have very clear goals.
Time-box specific windows. Don't let crunch
extend beyond certain time frames. Identify which specific people will be
working overtime and specifically which goals they are attempting to complete in that time.
This isn't just about quality of life. Like, this *is* about quality
of life, but it's also bad for business.
So if you work people too much they can churn out and leave the
company. Often people working a lot of overtime end up taking their time back
and they start messing around on the clock and playing games or watching
videos, and that could be super damaging to morale. And obviously as
people are more tired they aren't as focused and they do lower quality
work and create more bugs. There are studies about overtime and
they show that bursts of overtime longer than two or three weeks result in so
much reduced productivity that it would have been more productive to have not
crunched at all. I just want to say that I know this isn't easy. I know this is a
sensitive topic. But I do think it is a fight worth fighting. So those are my
five topics: Responsiveness. Matchmaking. Vision. Player sentiment. And Crunch.
I had a lot of fun thinking about these from a different perspective. And all this
is just a huge learning experience for me. I've just embarked on my new
journey at our new company, and I'm learning a ton right now so I'd love to hear your
thoughts on all this. And just to wrap up I want to say how awesome it is to be
here in a room with all of you talented game developers. I just recently got back
into making games after being in leadership and management for two years,
and it is super fun. Like, making games is more fun than playing games. PAX DEV is
a special opportunity. Not just to listen to awesome stories and talks by your
peers, but also to meet the person sitting next to you and make some
contacts in the industry. So we've all got unique perspectives. We have a lot to
learn from each other, so introduce yourselves throughout PAX DEV and ask
each other questions, share what you're struggling with, and help each other make
better games. Also feel free to come hit me up if you have questions about Second
Dinner, or if you have any good restaurant recommendations in Seattle, or
if you want to challenge me to a game of Beat Saber. Thank you very much, and enjoy
PAX DEV!
For more infomation >> Game H - Duration: 0:46. 
For more infomation >> #2 GUESS THE KPOP IDOL BY THEIR BABY PIC | KPOP GAME - Duration: 7:19. 
For more infomation >> Fun Baby Care Kids Game | Sweet Baby Girl Summer Fun 2 | kidstv 05 - Duration: 21:06. 
For more infomation >> Cận cảnh hách game bầu cua bịp,tài xỉu bịp,xóc đĩa bịp mới nhất - Duration: 8:27. 
For more infomation >> Shillong Teer 29-09-2018 | Shillong Teer Result | Teer Game - Duration: 4:19. 
For more infomation >> Game of the Week: Trinity vs. St. X - Duration: 2:37. 
For more infomation >> Achromatic | Chilled-Out Game Reviews - Duration: 2:49. 
For more infomation >> Police: Shooting at game was self-defense - Duration: 2:34. 
For more infomation >> Hills of steel PHOENIX Tank vs LASERJAW - Kids Tanks | Games Bii - Duration: 10:26. 
For more infomation >> SEGMENT 1: Georgia Games - Duration: 10:07. 
For more infomation >> Silsbee High School gears up for homecoming game against Lumberton - Duration: 2:52. 
Không có nhận xét nào:
Đăng nhận xét