March 13, 2024 • 54 mins
Enjoying the show? Hating the show? Want to let us know either way? Text us!
So much of Tricky Bits is an ongoing conversation between Rob and PJ about various tech topics. This particular episode revisits some old topics around game development, its history and costs, and what factors come into play that shape the market today.
As a heads up...this one is more conversational than some of the past episodes as we build up to a larger discussion for a future conversation around big and small media in various contexts (games, movies, television).
Finally...we announce our Discord server, where you can comment on individual episodes. You can find the link to the thread for this episode here: https://discord.com/channels/1146929485694902282/1217606185159491695
Come let us know what you think!
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:14):
Welcome back to Tricky Bits with Rob and PJ.
You know a lot of where thispodcast comes from is the
conversations that Rob and Ihave had through the years and
continue to have.
Often we're going out to have adrink, and this episode is
(00:34):
going to be a little bitdifferent than our prior
episodes.
Our prior episodes we tended tohave more of an outline, and
this is going to be a bit moreof a popery episode, but at the
same time it is in many ways aprequel to a topic we've been
(00:55):
discussing for a while, which isthis ongoing discussion about
big and small media as itpertains to games, movies,
television shows, really thewhole gamut, and we've been
trying to piece a lot of thisdifferent stuff together.
We haven't quite cracked it yet, but this is an episode that
(01:21):
gets you into our mindset aboutwhere a lot of these topics
intersect.
Whether at currently it's alittle bit of an experimental
episode for us, We'll see howwell you like it.
Don't worry, next week we'regoing to go back to our outlined
version, so you'll get to enjoythat again, but we're really
(01:41):
curious to see what you thinkabout this particular format.
With that, I'm going to drop you, as they say, in media race
Folks who don't speak Latin.
That means in the middle ofthings.
So one of the things I starteddoing research on yesterday,
after we talked about the bigmedia versus small media and I'm
(02:03):
only at the tip of the icebergright now is we talked about
this hollowing out of the middle, and the research I'm starting
to do is how has the cost of AAAgames changed over the years?
But I want to be able tonormalize it against both
inflation as well as the gamingpopulation.
(02:25):
What is actually the upsidesfor these things and really try
to relate that back to the cost.
Speaker 2 (02:32):
I also think it's the breakdown of the cost increase
of.
I don't think this comes backto the Unreal Engine and do you
have an engine team is fits inthis same thing it used to be.
It was all programs Back in,maybe one artist with lots of
programming time, and over timeit's shifted the other way.
Now AAA games are mostlypipeline and content.
(02:56):
If you can speed up thepipeline of how quick can an
artist see that result in thegame, you get better iteration
and you get more experimentation.
If you have to wait six hoursto see the result, you're only
going to do things you know work.
You're not going to experiment.
So I think a shorter iterationand a better pipeline gives you
(03:18):
better games for one.
And how that affects cost youcan dive into it but you can
just throw in the bullet pointsout there.
But today most of the costtoday is in content and if you
can speed the pipeline up thenyou save a crap ton of money
because most of the people onthe team are content related and
(03:40):
if you can speed all their jobsup, that makes things super
efficient.
So I think the big gains incost reduction are in the tool
side of tooling, the iterationtime, not necessarily in
programming.
I think engine teams have beenstable in size for a long time
and they cost a lot of moneybecause they all experience
programmers.
But in the grand scheme ofthings it's a tiny percentage.
(04:02):
These days, if you spend 10percent on programming on a game
, you're probably doing it wrong.
Speaker 1 (04:06):
So I'm going to challenge this one, rob.
Yesterday we did talk a lotabout how every Unreal game
looks the same right, and so ifwe basically say and I'm going
to ignore for a second yourpoint about spending the money
on a bespoke engine team toupgrade Unreal, Let me just hold
(04:26):
that one off to the side for asecond but if content alone was
the deciding factor and that wasthe only place for innovation,
then I didn't say it wasdeciding factor.
Speaker 2 (04:40):
I said it's much the cost.
Well, it's OK, that's fair.
Ok.
Even a triple A game takesSpider-Man.
Content was much of the cost,so it's not the deciding factor.
It's just that big, moderntriple A games have a lot of
content.
They're expected to be thislong.
They're expected to look thisgood.
That requires a lot of content.
You've got hard surface models,you've got material models,
(05:01):
you've got lighting.
There used to be one person.
Now they will separate.
Separate jobs yeah, Inprogramming you now have a game
programmer, tool programmer andmaybe a network programmer, but
some of this is libraries.
Middleware comes into it.
Middleware and art doesn'treally exist because it's the
theme.
You could use W-wise for audioand get rid of your audio
programmer, or maybe have oneaudio programmer to integrate it
and work with it.
(05:21):
Have it the same thing.
This collision could all be.
You integrate with it, when forart it's a lot more difficult
because whatever you do needsthe aesthetic of the end result
of what the art director laiddown.
So it's much harder tomiddleware art over then using a
third party and giving them thedesign doc.
Speaker 1 (05:41):
Sure, I mean, that's the content houses problem that
we've talked about.
Speaker 2 (05:45):
But bringing it back full circle.
Because of that and because ofUnreal, it is possible to make a
game that doesn't have barelyany programming stuff.
Unreal has all these things.
All these things that bespokeengines integrate with the
programmer audio, physics,collision, lighting, whatever it
may be, third party network andlibraries, ai, path tracing,
(06:05):
libraries, things like that canall be integrated by a program,
but they're all already inUnreal.
So you can do all these thingsthat all the teams are doing out
the box.
Everybody does it the same way.
You get what's in the box andthat's where it looks like an
Unreal game comes from.
Everyone has the same tools.
The best games are the ones youuse those tools best and the
(06:27):
ones that rain Unreal in,because it's very easy to hang
yourself.
There's plenty of rope there todo anything you want and you
can do in side by side.
In the store you'll get effectsthat are the same effect water
on a surface effect.
One of them is real time andone of them is for offline
rendering, and this one's crazyexpensive but looks really good
(06:48):
and you could pull it in yourgame and it will work.
But you're paying the cost andunless your game is entirely
based on wet surfaces and youcan afford the cost, then you're
going to screw yourself.
It's like building a house.
I got a million dollars tobuild a house.
It's like I'm not going tospend it all on the windows.
I need to do the rest of it.
I need a whole house.
Part of your job is knowing thelimit of what's reasonable and
(07:11):
realistic.
What's the cost of authoring,what's the cost of maintaining,
what's the actual performancecost in terms of GPU cycles and
memory and whatever?
And I think the teams that haveengine programmers are far more
aware of this, because theyhave engine programmers who are
looking at this stuff all thetime and looking at performance
all the time, and they'relooking at profiles and they're
(07:33):
looking at memory footprintgraphs and things like that and
they're like oh, what happenedhere?
Oh, that asset got added tothis.
Like, is that a viable thing toadd to the game for the cost?
Speaker 1 (07:42):
Rob, do you think that's something that's missing
in the general Developmentprocesses for everybody, Like
everyone, not just the engineprogrammers, but the gameplay
programmers, the artists, likeeveryone should actually have
that performance window up toreally understand, you know,
what these spikes look like.
Speaker 2 (08:00):
And much of what everyone should.
They're hard to read.
You can misread them and readthe wrong data from them.
I do think there should be aprocess that runs once a day and
goes oh, what happened today?
You have a buyer and you have ascript or code or a person that
pulls out the critical piecesof information in the way that
(08:22):
makes sense for the team.
How team-a does it and howteam-b does it may be different,
but the information that'simportant for the game being
made should be pulled out on adaily basis and graphed Over
time.
You can go, look, performanceis going down as assets get
added.
Someone added shadows oneverything and it's very easy to
(08:43):
add light to that cash shadowsand don't do anything, but you
still pay for the shadow.
So, working with what you have,pulling out that information
and, I think, graphing it overtime, even if you do once a week
okay, last week with 60 hertzan hour, 40, what happened?
If you do that once permilestone and you're like, oh
shit, it's really slow, it'shard to go back and undo what
happened then you miss themilestone.
I think even a week may be toolong.
Speaker 1 (09:04):
I was going to say probably a daily basis.
Speaker 2 (09:06):
I mean yesterday's build, today's build, this
happened and that can be allpart of the automated build
process to make sure it actuallybuilds.
So you have these like look,something happened here and it's
like we can undo yesterday'swork without busting the
milestone.
We can't undo, or we can atleast bubble it up to the top
and be like how important isthis Because it's really
(09:27):
crippling us.
Or can we reduce the quality?
Can we remove it completely?
Whatever it could be, it couldbe a new subsystem in code, it
could be a new asset, it couldbe a new lighting Many reasons
it can happen.
It's not all bad.
It's not all because of badartists.
Speaker 1 (09:39):
No, no, no, no, I get it, it's all about programming.
Speaker 2 (09:41):
There's a lot of reasons and games are so
complicated now that spotting itis straight away Is essential.
Because if something gets builton top of that problem, yeah,
now it's also a problem and itcan't be removed and it needs
what it's built on.
So spotting it straight away,fulton up to the top and
deciding then and there is thisneed, and sometimes the answer
(10:03):
is yes, because you all theseother systems are built on top
of that and that's the cost,deal with it, optimize Later on
in the game.
You don't know what to optimize.
You went for the profiler andthe problem with unreal is
nothing stands out.
You'll be basically, I call itdeath by a thousand cuts, sure,
and there's nothing that standsout in the profile of the times
right in front of you could seewhere it's all going, but
(10:24):
there's literally a thousandtiny things that shouldn't be
there.
Yeah all of them together add up, but any one of them that you
spend a day or two optimizingdon't make any difference.
Not getting those in the firstplace is the way to make unreal
sing in terms of, like the topmetrics that come to mind.
Speaker 1 (10:41):
I mean, is it really Frame rate and memory, or are
there more that are like hey,these are like the, the hot
spots you should really belooking at in that graph, or is
it?
Hey, it all depends upon whatplatforms you're targeting.
Speaker 2 (10:57):
Yeah, it depends on the game.
Ever like, oh look, this worksgreat on good, this new thing
works great on DX 12 and Superbad on DX 11.
It's like how important to take11?
Can we drop support for DX 11?
Does it have to work reallygood?
Can we optimize the game allthe ways?
Oh look, it works great on XboxX, but not an S.
You have to release an S.
(11:19):
What are you gonna do?
Can you take the asset out ofthis build?
Can you just make it simpler?
Same between PC and bothconsoles, like the PC is more
powerful than both the console?
Yep, a lot more memory to.
Should you be using that memory?
Speaker 1 (11:34):
You're you're operating in the triple-a space
and you're saying I was want tobe there.
Speaker 2 (11:39):
We added ray tracing now on the platforms that don't
have ray tracing, so it's notonly is it optimum in the
optimizing, it's like what's thealternative?
Does it have to be there andcan we design the game around
not having this?
And that's part of being a goodteam, and I think this is why
you pull the key rock stars outof a team and put them on a new
team, stadia style.
Then all you end up with is abasically a pit fight of a bunch
(12:03):
of egos, big dick measuringcontest, basically of who is
who's right, and it's like wesolved this problem this way and
we were right because our gamewas successful.
We solved it the exact oppositeway and we were also right
because our game was alsosuccessful.
Take a step back and look atwhat's needed for this game,
which is why teams that sticktogether tend to be better over
time.
(12:23):
The Notre Dame somnax of theworld.
Because of this, like everyoneknows what everyone does and how
they work right, kind of havean understanding of the limits.
I could do this, but it's notwaste of time because it's gonna
get pulled, because it's gonnabe too slow, or that's the ond
project.
I'm like we want this huge,animating, city-sized bad guy
and Engine is not really madefor it.
(12:44):
But let's do it anyway, and Doit the way we would have done it
all, the way art wants to do it, and Then go and look at what's
happening why is it breaking?
Let's increase command buffersizes, let's increase allocator
sizes so it works, and and thengo from there.
I'm like, yeah, naturally it's.
(13:04):
We could do it if we tweak this, this and this, or With the
lightings breaking because theassets so big or whatever it may
be.
It's like we can see theproblems and Address them case
by case if that's reallyrequired.
But that's on D.
Don't do that in productionright.
Speaker 1 (13:23):
But I think it goes to your point of if.
If I want my game to havefeature X and it's a hallmark of
the game, whether it's I wantto fight a giant city or
everything needs to be wetsurfaces, that's fine.
You're making a choice forwhere you want to put that
investment, both in terms of artand performance, right.
Speaker 2 (13:42):
It's like building the house.
Yeah, I want a swimming pool.
These people don't.
There's a cost associated withit.
If I you have a swimming pool,you can have an Olympic size one
and spend all the budget, andnow you've just got a swimming
pool, right, or I wanteverything clad in marble, and
that means I.
Everything else needs to be MDF, and I want a gold toilet right
why not, but and?
Speaker 1 (14:04):
It means that your pipes are made of lead.
Speaker 2 (14:05):
Hey you're, but that's where you put your budget
.
That's fine.
You still have to complete therest of it, right?
That's why you get games on agood tech demos and you get.
That's how games get funded andthat's ultimately what guts the
middle, the middle class ofgaming that's hollowed out right
now is because of that, and Ithink what real does play a big
role in In that.
I just don't know how.
Speaker 1 (14:27):
I do too.
This goes back to that dataquestion.
I found a great site where theylook at the cost through the
decades of, let's say that, thecost of a console and the cost
of its game, where, like thecost of a console in In 1977 of
the Atari 2600 was 200 bucks,which in 2020 dollars, which is
(14:50):
when they were doing thisarticle was just shy of a
thousand bucks, and the averagegame cost in 1977 is 40 bucks,
which translates to 170 in $20,and so there's a part of me that
really wants to say that Iactually think, as much as we
talk about, oh, game prices areincreasing, I Think there's
(15:13):
actually been a massivedeflationary effect in games and
I suspect what is actuallysustained it has been a larger
game population.
Speaker 2 (15:22):
But I don't know, and this is like the thing I want
to do.
The data on this a hundredmillion playstations.
Well, threes, fours, fives willbe there soon, are you sure?
Yeah 75% of them are activelyused.
Every every now and then.
That's a lot of.
That's more than right.
What every was 2600 and yes,games cost a lot more to make,
(15:45):
but there's a lot more peoplefunding them and I think it
balances out.
I mean, the big game companiesare not hurting.
Speaker 1 (15:51):
Oh, they're not.
I'm curious.
I said this again.
This is like the Marvel movieproblem that I'm saying.
Speaker 2 (15:56):
Well, I'm curious of the flip side of that.
I want I want to see statsNumbers from the games that
failed.
Sure, like, okay, this game gotcanned.
Why was it purely financial?
Was it never gonna technicallywork?
But the financial ones of like,how much did you spend on what
and where, and how did you nothandle the Monetary budget but
(16:17):
also the technical budget, whatpath was taken together today?
I think there's a lot to belearned.
That no one talks about it, noone does Public postmortems of
like this is where we screwed upand we went down this path and
we should have gone down thispath.
There's a lot of informationthere that you could really
learn we could use Duke Nukemforever as our case study.
(16:39):
Yeah and uh.
But even big games are like whydoes it?
Why did EA?
Okay, why did Sony just closethat studio down in London?
Obviously some thought wentinto it like kill the whole
studio, don't just let somepeople off, just get rid of the
whole game.
What state was it in?
And I think there's a lot tolearn there.
But you only learn this bybeing involved with your
particular game and Even, maybefor the team, don't have access
(17:00):
to this info.
So it's hard to learn frommistakes because they never
talked about.
Yeah, it's like came, came, go,shut down last a little bit
last year of it.
A leak of it may happen andyou'll look at it and I go is
actually look pretty cool, butthat's the only Insight that
you'll get be a few screenshots,maybe a leak of the of a
running game, of video.
You'll never get like a studioto sit down.
(17:21):
But like okay, numbers,technical numbers, hiring,
firing like that level of detailwould be a huge win for the
industry and I think it wouldshow that a lot of these smaller
indie, middle ground studiosare doing it wrong.
They're all on that path tofailure.
Some make it, some don't.
Speaker 1 (17:39):
The premise they're all on that path to failure
because the economics they'reworking with is is there,
approaching like a triple A gameor what's the?
What's the fundamental mistake?
Speaker 2 (17:48):
I think they're all on that path to failure and some
some happen to survive.
I think the path is the nextpaycheck is not guaranteed.
It's like miss a milestone.
What happens is like somepublishers are real bad about
miss a milestone.
We're not gonna pay you if youdon't pay you.
If I don't pay my staff, you'renot gonna get the next
milestone.
There's a lot of viciouscircles that come up when you
hit these, like hey, it's not upto performance, didn't quite
(18:11):
check all the boxes on themilestone checklist, you can
literally put the finger on okay, this missed milestone cause
this miss payment, which callsthis entire cascading effect.
It's just we don't have thatlevel of information and I wish
companies would just take a stepback and go okay, here you go.
He's the numbers.
Why are we so concerned abouthow much games cost?
The insomniac leak leaked abunch of money.
He's off 200 million for a game.
(18:31):
It's like it's just a number.
Yes, it's a big number.
If you're paying, it's a lot ofmoney, but it's also an
investment in the future.
Yeah, you're hoping to get areturn on that money.
So why?
Why is it all hidden?
Why do we care if insomniackeep their numbers secret or if
they publish them like whobenefits or Loses from those
numbers being published.
(18:52):
Maybe insomniac themselvesbecause competitors could be
like what we can do a half aprice, but can you?
On paper you can, yes yeah,that's the thing again.
Speaker 1 (18:59):
I will keep going back to my Marvel movie example
because, like, you'll dump 200million dollars on Infinity War
but if it makes a billion backfor you, great, that was a good
investment.
So if you're putting in 200million to insomniac, to make
spider-man or spider-man 2,guess what, you'll probably make
back that investment.
You know, if you have it's what50, 60 bucks a pop.
Speaker 2 (19:23):
I mean across 100 million machines the money I
mean in theory is that.
I mean, if you sell 10 millioncopies, yeah, that's a crap,
crap turn of money.
At 600 million dollars, yeah,that's retail, though which
Obviously there's all theBusiness tricks to that.
They play up like if I publisha pace for a game, they don't
(19:47):
pay off the development costswith retail money.
They pay off the developmentcosts with the royalty you would
have got.
So once you've paid it say youget, yeah, so the game cost 10
million and you're getting adollar per copy royalty.
Yeah, yeah, you've got.
(20:07):
Sell 10 million copies to payoff 10 million, right, oh right,
retail may just get to go tothe store.
It goes to Publisher, blah,blah, blah.
But what happens when they'reall the same person?
Like if I buy a game fromPlayStation Online, if, like,
there is no store, 50% right togo to the store.
Sony is getting that now, butit's not Sony development, it's
(20:29):
Sony store, which isn't goingtowards development costs.
So if there's a lot of fingersin the pie, which right, which
make the numbers a lot less rosythan you'll think when you look
at all, this took a billiondollars at retail in a week.
It's.
It's not that pretty for thebottom paper at the bottom,
which is always the developers.
The developers are always atthe bottom of the pile so.
Speaker 1 (20:51):
So this is, this is the Classic problem.
Speaker 2 (20:54):
Oh, yeah, it's exactly that.
Speaker 1 (20:56):
Yeah, this is Pay for that term before.
Speaker 2 (20:59):
It is the same people or this they hang out the same
thing and Same people.
In a lot of the times, ifeverybody was in it for the same
purpose, like retail money,would pay Everything off and
then would be divided up.
But it's not everyone wants tothese.
This group wants their cut fromthe first case copy sold, where
this team doesn't get a cut, tothe 10 millionth copy sold and
(21:22):
and yeah there's the problemthat is the Hollywood accountant
, but that is also part of theproblem.
Yeah, a lot of the money doesn'tfeed back to developers To do a
second version or to super upthe ante on the next version of
whatever it is.
We've got a lot of money comingin.
We can.
We can do it, and it also meansa lot of games would break even
before they actually do.
(21:43):
There'd be a lot moreblockbuster games.
But I mean, I guess right nowgames are like movies.
You'd get more triple a gamesbecause there'd be more money in
those triple a games if theywere paid at Retail value or
these development costs werepaid at retail and then the
final money was split.
You have to sell yes butobviously that money is coming
from someone else's pocket.
Who currently keep wants tokeep it, which is why the status
(22:03):
quo isn't gonna change.
It's not like there's moremoney to go around.
It would just be allocating itmore fairly.
Yes, and I think it's fair tosay no one makes any money until
all the expenses are paid, butthat literally isn't how it
works.
Speaker 1 (22:16):
You have the Sony store, you've got the Sony the
publisher, you've got Sonylicensing right, which is the
one who would have the rightsfor spider-man.
Speaker 2 (22:25):
And you also have the Sony license fee for the
PlayStation of like oh, everygame has a Built-in fee that you
have to pay even even ourfirst-party studios, even if
you're Sony here.
So yeah, it's just because Iguess they run is totally
autonomous divisions and it'slike, well, that's our
accountant, this is our income,that's our outcome and there's
(22:46):
the way it is.
But nice if it didn't happen.
You'd get a lot better games Ifthat wasn't the case.
But the financials are goingdown the movie path.
Games are following Hollywoodand the counting is what it is.
There's a lot of money in thegames Only if you can overcome
that hurdle, and a lot of gamesmiss that hurdle are great games
, but never make that hurdle andnever pay off the development
(23:07):
costs and they never get anotherone made the team this bands.
Blah, blah, blah.
Speaker 1 (23:10):
But this is why we have really this, this migration
up to the top of Only triple agames are making it.
Speaker 2 (23:18):
Yep, that's what happens, the guys who make the
development payback hurdleProgress to triple a, the ones
that don't either disappear,which hollows out the middle, or
they migrate to a lower end,down towards the mobile space
that basically draw from thecenter to both ends.
Is what's hollowing out the thethe middle there's a lot of
(23:38):
other reasons for it too.
But I mean, we just talkedabout unreal, about the
financial reasons, they.
They were all basicallyhappening at the same time,
which is a big problem for themiddle of gaming, did we?
Speaker 1 (23:50):
invent the lower end in the last couple of decades by
virtue of changing thedistribution model.
And what I really mean by thisis that before I had to have
physical media, you know, I hadto have a DVD disc or a blu-ray
disc to play Ration, clank orresistance, and that was a cost
(24:13):
that was there and it had to gointo a big box store.
Did we Invent the lower end byvirtue of changing distribution?
So we have an Apple App Store,we have steam.
Anyone can put their game outthere now and be distributed.
It may not be successful, butit can be distributed.
(24:34):
So did we invent this by virtueof how we change Just how we
even distribute these games out?
Speaker 2 (24:39):
I don't think one fed the other.
The app store was gonna happenregardless, independent of any
game thing, and I think itverified that digital
distribution was possible andthe big game still struggled
with digital distribution whenit's 100 a gig to a download.
We're definitely going downthat path.
(24:59):
But it does open up thatquestion that you mentioned of
like there's more money now Inthe pie because we're not paying
the retail store and we're notpaying this.
So there is more money acrossmore gamers, which is how the
games do get funded.
So a lot of that cost thatwe're not making DVDs and
(25:20):
there's millions of dollarsthere for millions of discs and
that cost is now up for grabs,for example, that does get split
up better across to the variousteams.
It's not just going insomeone's pocket, but there's
also a cost of the back end ofruling the online store, things
like that.
The physical media is whatcauses the yearly cycle.
In the PlayStation 1, which is2, you add Madden and NFL 1995
(25:46):
and it's no way to update itnext year when all the players
have moved around and all theteams have done Whatever they do
, and stadiums get built andknocked down and things like
that.
So you now need my 96, becauseNFL is now a different world.
There's no way to update that.
It basically a year by year.
It keeps up with what's goingon in the real world.
Speaker 1 (26:04):
Most of the same game engine, just Couple of art
changes.
Speaker 2 (26:08):
No, the engine does get better for sure.
I mean, you look at the diffacross them all.
Go back to the Sega Genesis andit's like there was huge steps.
When we switch platforms, thepoint is they have to do a year
release because there isn't away to get any more data on to
that PlayStation.
Right Today's world witheverything being online, I Think
they're taking the piss doing agame every year of like madden.
(26:31):
If, like, you could justdownload that database, it'll be
fine, but they won't.
We've already got this modeland I think that model feeds the
other big franchise model,which is why we get a call of
duty once a year.
It's why we get one of all.
The big blast dust is once ayear Because it goes back to
those sports games on thephysical media, on the machines
you couldn't update and for, forEA and for these games that
(26:54):
have a brand name.
Speaker 1 (26:56):
It is in the Economic interests of those companies to
keep making you pay 60 70 bucksa year To get an update.
Speaker 2 (27:06):
As you say, that could just be a download, yep
and it's Probably Charles six orseventy books for the download.
So you know the end customer isnot winning regardless.
Downloads Could accommodate alot of that, or you can play
this year's teams on last year'sengine, for example.
There is definitely like a Lotthat could be done to fix the
(27:28):
yearly updates, but there's nowhy people buy them regardless
and it's another billion dollarson the table.
Maybe we'll take care.
Speaker 1 (27:36):
So let's look at the flip side of this, though.
With the rise of digitaldistribution, did we create a
lower end, meaning like therewas perhaps a Lower bar you
needed to hit when you hadphysical media because of just
the cost of it I and with whatamounts to zero marginal cost to
(27:57):
distribute a game.
Did we craft this lower endwhere it know, where it couldn't
exist previously?
Speaker 2 (28:06):
I think just the normal distribution of
Developers crafted the bottomend.
Some point you have to say,like, if I started a game
company today, I couldn'tcompete with no a dog in
insomniac.
Yeah, no matter how good I amor matter how good a team I hire
, I can't compete with them.
Again, it's the, the legacythat they have, and the, the
(28:27):
ability to work together Doesn'tmean you can just create a team
and replicate it.
So when would I go if I startedthe game today?
I wouldn't be triple a.
I Don't think I would aim forthe triple a.
It'd be much better to go andlower end.
Go lower than where you belong,because then you'll stand out
(28:49):
has been really good and I thinkthat's why a lot of really good
Developers went to the iPhoneand things like that.
But don't kid yourself that thezero cost, the cost per
acquisition on an iPhone andgetting noticed in the store, is
incredibly expensive and it'snot much better in the
PlayStation store, the Xboxstore or steam.
(29:09):
So Although it is zero cost toget it in the store, to get it
out of the store onto people'sdevices is an expensive thing,
and Is it worse than marketingtraditionally?
I have no idea.
It's not what I do, but butthere is a cost there and but I
(29:30):
do think that also the lowerends always existed and and it
could exist on the same platform.
The Commodore 64 had greatgames and crap games.
Speaker 1 (29:42):
Some of the crap.
Speaker 2 (29:42):
Games were done by good developers and some of the
good games are done by unknowndevelopers.
And you add low-end, reallygood games and that's still the
same today.
And I think having the abilityto have a phone and different
types of devices Moves thedynamic of the high and low end.
You always say high ends, likeall PlayStation triple A's, like
(30:04):
that's the high end.
What is it?
You can do a lot of this on aniPhone and iPad or an Apple TV.
You'd be better off making iton a PC or an iPhone and then
using that to get onto thePlayStation, but by then your
games already out on the thephone platform.
So I think it gives an avenuefor development.
It gives a very nice avenue forexperimentation.
Speaker 1 (30:21):
Yeah, it's.
It's interesting you bring upthe Commodore 64 example,
because when I think about thebarrier of entry, I actually
think about it more in the Atari2600 NES era, where I needed a
fairly beefy amount of physicalmaterial between the cartridge
and the electronics to actuallyget my game out there, which
(30:46):
felt like it created this, thisbarrier.
Oh, it was contrast to theCommodore 64 where it's like,
hey, a floppy is pretty cheap,even back then, you know, as
long as I can get the data on toit, I can, you know, sell my,
you also have the free path thentoo.
Speaker 2 (31:02):
I mean, you could send your game to a or demo of a
game to a magazine, and they'dput it on the.
Sure shareware disc or classic,yeah, leave a shareware demo of
like oh, he's a game, it's ademo of the first level, yeah,
and buy it later.
Shareware to us.
How it how it all all started,sure, yes, the cost of entry on
the consoles Today is nothingcompared to what it used to be
(31:24):
today.
You got a dev kit which isreally just a modified version
of the base kit.
Back there, you add, the devkit was pretty much the hardware
.
There wasn't a differencereally between a Sega Genesis
that was used for development,one that wasn't because there
was no software in the Genesisper se, it was just a cartridge
slot.
But you would have a cartridgethat had ram in it, that
connected to a PC, and youdownload the game to that, and
(31:47):
every now and then you'd burnone to a ROM and stick it in to
make sure it actually worked.
And then, as we got to the, thedisk based systems, that option
became a lot fuzzier.
I was like how do you get thecontent onto the device?
when it's a disk based systemand Doesn't have a disc, the dev
(32:08):
kit might have a drive.
You're not gonna burn a discevery time you build a code.
So all the console, all thecartridge based consoles, pretty
much had RAM based, flash basedthey were.
They were slightly differentbecause they had debug support
and hardware level debugging buthalf of it would work on a
retail kit.
It weren't until we got to thePlayStation had they started to
(32:31):
have disk emulation.
So they had file systems thatwere on the PC and it would go
over a high speed high higherthan CD ROM speed connector to
the, the actual dev kit in thecustom operating system.
Because I had to read from thefile system and not from the
drive.
And Then that gets us to wherewe are today with them.
(32:53):
These things became signed andthey'd sign them differently for
retailing dev kits and Sure theonly difference today between a
retail and a dev kit is mostlythe signatures needed to get
stuff to run.
Speaker 1 (33:06):
Yeah, so I mean it's.
It's really a Encryption issueat that point in time and, like
we've talked about in the past,the hardware that is either for
the PC or the Xbox.
It's basically a PC and thisgoes back to my, my point of
like to have in these cartridgebase or even early disk based
systems where the dev kits wereexpensive or, you know, burning
(33:27):
a ROM was expensive, like youhad to have a fairly Large
upfront capital cost to getinvolved, so it didn't allow.
It doesn't seem like it wouldallow indie developers to get
into that field.
Speaker 2 (33:40):
There wasn't any indie developers on the on the
NES.
The NES in particular was brutalto get dev kits for.
Like.
Nintendo have opened up a lotrecently, but in the early days
they were Absolutely brutal toget anything on and but yeah,
you have to have.
They wanted to see financialsof like can you afford to make
the cartridges?
And they made the cartridges.
(34:04):
That's how they got the licensefee and that's where the
license we came from originallywas like paying more than you
should for having thesecartridges made.
And, obviously, the whole storyof EA, of where an Activision,
how they popped out, of Atari,because that's how the whole
third party world showed up.
There's a whole story therewhich been told many times
(34:25):
before.
But yeah, so you'd have to havethe upfront cash to buy, make
these things.
Indeed, it wasn't cheap.
It was many dollars per thing,especially with the box and
everything.
And If you're making a hundredthousand, you might need Three
quarters of a million dollars, amillion dollars, whatever it is
(34:46):
.
Just to make it a set tape or adisc was fairly easy, but still
a physical thing to manufacture.
Today there's none of that.
You can literally just pass thestore checks, the various
quality control checks for anyplatform.
Pc has steam, the Xbox,playstation have their
respective stores.
Yeah, apple, android have theirrespective stores.
(35:08):
You can basically make a gamewith zero.
You have the development cost ofthe game, but then after that
there's zero cost to get it intoas many people's hands as you
want.
There's no ongoing cost.
You could count the acquisitioncosts, but that's really
marketing.
Speaker 1 (35:23):
That's marketing and that was always the case, though
that's why I'm separating thatout that there's in terms of
distribution costs nowadays,digital effectively makes it
zero, and and this is also goback to your original question.
Speaker 2 (35:36):
I was like why are games in 2020 four dollars less
than they used to be?
Well, that's a factor thereright.
Speaker 1 (35:43):
Probably higher margins by a long shot for these
companies, since they don'thave to actually manufacture
units.
What if I'd fascinating is likewe go back in time, we get EA,
we get Activision and we getinsomniac naughty dog, all of
whom came out of either console,cartridge discs, times and now
(36:07):
occupy the space in the triple aspace.
I and I'm kind of wonderingwhere was the middle previously,
if not in the cartridge worldbut maybe in the console world,
like where was our middle?
Because I'm kind of wonderingif we always had this, like the
AAA problem was always therebecause of the cost of
(36:27):
development historically, and weonly have recently acquired
this notion of lower to middleend by virtue of the fact that
it's a lot cheaper to make thesethings.
Speaker 2 (36:38):
I think here, it's a regional thing.
In the US you always had the2600 and at the other side you
had the.
Apple II, for example.
The UK didn't get the 2600 andit didn't really buy into the
consoles until the much laterSega Genesis era is when the
(36:59):
consoles in Europe got popularand they still were different
because of the frame rates.
The UK always had the homecomputer scene.
It had the BBC Micro, it hadthe VIC-20, it had the a few
random the Auric and the Dragonand the Amstrad and the.
World.
That's way later.
I'm talking about the earlyspectrum, of course, and then
(37:22):
that became the Amiga and theArchimedes, amstrad, and
Spectrum kind of went away.
The 64 was in there too, ofcourse.
I forgot that one.
But all these home computers,and everyone was writing
software Consoles were not athing at all until much later,
in the early 90s.
So from a UK point of view,there was always this spread of
(37:44):
quality developers.
Speaker 1 (37:46):
Some were crap, some weren't.
Speaker 2 (37:48):
And we didn't have that overhead cost, we only had
the discount cassette costs.
The US was very different.
The US definitely went down theconsole path.
The bigger crash of video gamecrash of 83 was really the Atari
crash of 83 and only affectedthe US.
The rest of the world justcarried on as normal, just
within the game console.
Speaker 1 (38:08):
Interesting.
Speaker 2 (38:08):
So I think your view of that, my view of that, would
be very different.
Speaker 1 (38:12):
Well, I think it's interesting because, from your
perspective, the existence ofthat middle tier was perennial.
It was always there by virtueof people who were just
programming on these things,versus, as you say, my view,
which is that we had theCommodore 64 which had those
kind of qualities, sort of byAmiga, but then, effectively,
(38:35):
there's this giant step functionwhere you enter the console
world, where these capitalexpenditures are huge in order
to get involved.
So it actually eliminates theexistence of a middle tier for
the US, and I think the US hadmore money generally, so they'd
be like oh well, we'll do this.
Speaker 2 (38:54):
Money's easier to combine the US than it was in
the UK in the 80s and people inthe UK were happy with the
machines they had.
They were happy with the gamesthey ran.
There was the quirkiness of,like I said, NTSC, PAL update
rates.
So a lot of games made in theUS didn't actually work very
well in the UK, which is theprimary reason that the 2600
wasn't there at all, because youhad to program so close to
(39:15):
chasing the beam that PALliterally changed everything.
So they just didn't do it.
Playing American content wasn'tthe easiest thing, so they just
made their own and it was allon computers.
There's that whole spread fromthe ZX80 in black and white all
the way up to the Amigo and theArchimedes at the end of the
1632-bit days, and ultimately wealso switched the PC a lot
(39:36):
later.
I think the PC in the US wasalways a bigger factor than it
was in the rest of the world.
It wasn't until Windows 95 thatPCs in the UK really became
like oh, just get a PC.
Speaker 1 (39:47):
It was interesting because for us in the US in the
middle of the 80s you had threeat least in my mind, three
contenders.
You had kind of the Windowsslash PC world because IBM still
had its own IBM DOS for a longtime.
You had Mac and then you hadCommodore and we as a family and
(40:09):
Commodore was the smallest butI mean, well, I take it back
Commodore 64 sold really, reallywell.
So we had a Commodore 64.
We actually had a couple ofthem and then I later, in the
early 90s actually, as I wasentering high school, got an
Amiga 500.
And so it was still kind of incontention for a little bit.
But there was this like a curlthat was happening where,
(40:33):
especially like Windows 3.1coming out and then Mac started
like taking a lot, and thenCommodore made a few mistakes
where you basically only had abifurcation of choices a Mac or
a Windows box.
Speaker 2 (40:47):
And that's the same today for the most part.
But I also think there's a bigdifference in how the machines
were used In the US.
Commodore was very much aimingfor the business users.
They weren't any real greatAmiga like push the envelope,
Amiga games coming out of the US.
There's lots of businesssoftware coming out of the US
for video toasters and thingslike that.
(41:08):
Where the UK didn't do thebusiness side so much Like, the
Amiga in the UK was very much agames machine.
It wasn't really used for itsoperating system or it's the
other things they could do.
It was in some situations butthey were very much a games
machine.
Apple wasn't a thing in the UKin 80s and 90s.
(41:30):
We didn't have the Atari so wehad, like the Commodore machines
or the A-Cord machines, theSpectrums and the Amstrads and
things like that and gaming wason the forefront of all these
home computers in the UK whereit wasn't in the US, I would say
a lot of the gaming that cameout on the 64, on the Amiga from
the US was very bland.
Speaker 1 (41:46):
I'll admit that there were a lot of stuff that I did
with the Amiga that had nothingto do with games.
Speaker 2 (41:52):
Yeah, and the opposite in the UK.
So I just think how thesemarkets just merged together and
now it's I think it's where alot of good programmers came out
of the UK.
A lot of game programmers inthe US today are from the UK.
I think that's part of it.
Unfortunately the UK lost thatposition.
Back then they had some greatand this has nothing to do with
(42:14):
games today, but it had the.
The BBC obviously came out thewhole story of how it was
related to the computer programand they were going to make the
whole country computer literate.
And they did make the computerprogram and they did teach
people to program and all thatand I think it literally took a
whole generation and made themgreat programmers.
And later on that just kind offaded away.
(42:35):
And if you do computer sciencein the UK, if you do it in high
school, it's more or less how touse a computer Like how they
work is just not taught anymoreand back then it was very much
like this is how it works, don'ttinker with it, the whole.
And now it's obviously that's40 years ago when I was part of
that whole 1980s computerprogram mentality.
(42:56):
That mindset's gone and it'sunfortunate, I think I don't
think the UK's kicking out greatprogrammers these days, like it
was 40 years ago.
Speaker 1 (43:05):
Really unfortunate that we're not teaching the
lower level stuff.
Earlier on I saw a book in thelibrary where it was like, hey,
teach kids how to program andit's it's Python basically, but
that's not going to really getyou into the guts of the
electronics that way.
Speaker 2 (43:21):
It's not a bad way to go.
It's a different time, isn't it?
Back then, you had to get intothe guts of the electronics to
make anything happen.
The OS didn't get in your way.
Today, it's so difficult toeven draw a triangle on a screen
.
You need to understand DirectX.
You need to understand aWindows app.
You need to understand theconcept of brain buffers.
(43:42):
You need to understand visualstudios.
They're a very complicated tool.
You need to understand thebuild process.
There's a whole lot of things tolearn and a lot of boilerplate
code.
You have to write a shader.
You have to understand the 3Dmath just to get a triangle on
the screen.
Where there's a lot of that,that won't work first time.
If I wrote a single triangle onthe screen right now, I'd have
(44:04):
to run it a thousand times toget the damn thing to work.
What did you learn?
You learned to put a triangleon the screen.
You could do that Python too.
In Python, you can just go drawa triangle and use some library
and bang there it is.
It's very much like the olddays of instant gratification.
I think Web has some of theseproperties today too, where, in
basic, you could go draw arectangle.
(44:24):
Boom, there it is on the screenand tweak it, run it again.
It was just like instantaneousfeedback.
It wasn't this massiveknowledge base.
You needed just to draw onesquare on the screen.
The amount of code, the amountof work it takes to get
something to be in the placewhere you can experiment freely.
(44:44):
It's ridiculous today, which Ithink is why, even for things
like experimenting with 3D anddrawing graphs and things like
that, I think is why Python isused, because it's there and
it's easy and it is quitepowerful.
I mean, mackiddy wasexperimenting and it's all
Python and they use some librarywhich deals with all the
(45:05):
boilerplate and it just makes itinstantaneous.
You can say draw a triangle,draw a triangle, draw a triangle
, and they just appear exactlyas you said.
You could say plot circle,there's a circle.
It's not fast, but that instantgratification is how you learn.
It's like you'll get into theperformance.
And how would I make a gamewith this much, much later?
But today you kind of have tostart in the middle, you can't
(45:26):
start at the beginning.
Speaker 1 (45:27):
You can't, I mean.
So the good news bad news iscompared to when we were coming
up on this stuff.
I mean, at best we wereexperimenting on ourselves,
which could be great, andgetting it from magazines.
The good news bad news is thatall this information is out
there on the internet.
So if you had the same drivethat we did, all that info is
(45:51):
there to start playing aroundwith stuff.
Right, I mean, we're notlimited by, basically, lack of
information sources today.
Speaker 2 (45:57):
Yeah, but that all takes the entire conversation
full circle of why not just useUnreal it does.
It's like I could concentrateon the behind interest.
If I'm a gameplay programmerand I'm all by myself, yeah, and
I want to play with gameplaythings.
Do I have to write a 3D enginefirst so I can see what I'm
(46:19):
doing?
It's like some gameplay youcould do with, like text and A
star things like that you can dowithout any graphical support.
But generally it's like a lotof gameplay today needs the 3D
or the 2D or whateverenvironment of the game is
before you can do anything.
So that's what takes a lot ofindividuals to Unreal.
(46:42):
And I think it comes back fullcircle as to I'll just use this
because everything is alreadythere.
It only matters when you careabout what's there.
If I just need a 3D enginebecause I'm working on some
really cool fighting dynamics inanimation, I don't care how
fast the render is, I don't carewhat it's taking.
I got my dynamics, I can dowhat I need to do and I can add
(47:03):
my physics and I can do this andthat you only care about the
bit you're interested in, and Ithink Unreal is fantastic for
that.
It's got a lot of people intogame development that would
never have been there.
I think Unity is the same too.
There's so much boilerplate,there's so much that has to
happen to get to where you wantto be that you just use one of
(47:24):
the engines.
Speaker 1 (47:24):
And this goes back to our points that we made earlier
about where do you want toinvest, and we talked about it a
lot in terms of where do youwant to invest your development
dollars, but in this case we'retalking about where do you want
to invest your development timeas an individual, learning these
things?
Speaker 2 (47:38):
Yeah, An artist on the wheels, perfect.
Where else are you going to beable to do real time 3D modeling
and get an idea of what theperformance is?
So, yes, you can do it in Maya,but the performance is offline.
So again, I think, on one hand,unreal is awesome.
On the other hand, it's got alot to ask and so forth, what
(47:58):
it's done to the market.
Speaker 1 (48:00):
We've done a lot of tracing back in time.
I could make a proposal that welook at what you talked about
in terms of UK, where you had aspectrum of programmers that
were there and you were able tosend it to a magazine.
You were able to have somelevel of distribution.
I have this kind of conceptthat again I'm coming from this
(48:22):
perspective that the US marketwas really locked up in terms of
games and distribution for avery long time.
It was big box stores, it wasbig consoles.
There's a lot of capitalexpenditure.
I look at today and it's not tosay that this justifies anything
, but there is more opportunitytoday, it seems, whether or not
(48:45):
you use Unreal or not, where youhave the steam distribution.
You have Facebook and Instagramfor marketing.
You have these tools that havenever been available to anyone
like previously to actuallycraft a game and distribute it.
It doesn't necessarily meanyou're going to be successful.
(49:05):
Again, the marketing side ofthat is there for everybody.
The difficulty in making a goodgame is there for everybody,
but I'm wondering if actuallywhat we've come to today is
there is more opportunity in themarketplace, but it definitely
requires work on individual'sparts.
There's no trusting that somebig game publisher is going to
(49:27):
come in and swoop down and saveus.
We have to do it ourselves,yeah.
Speaker 2 (49:33):
That's a take.
I guess the market's mucheasier to get into today because
of the likes of Unreal andthings like that.
You can be a bunch of artistsand make a game, and it has
happened, I'm not even trying topraise Unreal and Unity at this
point in time.
Speaker 1 (49:47):
I look at them as tools, the idea that my iPhone
is a dev kit now for iPhone andmy PC.
Basically, I could bedistributing PC games regardless
whether or not I use Unity orUnreal.
Speaker 2 (50:02):
The distribution cost is gone.
There isn't any distributioncost.
There is because you have topay fees to these stalls and
things like that.
You can make it free and thendon't pay.
But that distribution cost nowbecomes server costs of like.
These stalls are not cheap tohost.
There's a lot of data beingmoved around.
Speaker 1 (50:21):
Yeah.
Speaker 2 (50:23):
I guess right, but to you it's zero cost.
If you make a free game in theApple Store, it's free to you to
pull up there.
It cost Apple to host it and itcost Apple to check it.
I think this bottom end of theiPhone and the whole how it
introduced free gaming and likepay to play type gaming, and
(50:45):
there's a whole story there too,which is a whole mess.
I think that was bad for gamingand I think that's trickled
into big games with the lootboxes and things like that and
obviously there's been a hugestink over there, lawsuits etc.
And it's not to say the bottomend.
Games are bad.
I mean, there's been lots offantastic games that have came
out of the iPhone that areactually just fun games and I
really appreciate that.
People experiment down there andif steam too, to some extent,
(51:10):
less so on the consoles.
For a while there was an indiescene on the consoles and it
seems to have just gone away.
I mean there used to be asection in the store.
Now they just in the same storeas everything else and you'll
never find them.
It's a healthy market in someways.
It's just there's no middleground.
It's gone.
You either make little games,experiment down there and hope
for a success and they can bevery successful.
(51:31):
Or you try to make a team andaim for AAA and ultimately you
might get your funding cut andyou never make it.
If you get over that fundinghurdle and get over that payback
royalty hurdle, welcome to theAAA space.
You were made it, but therearen't many going in and out of
(51:54):
that space.
It's fairly static and it takessuch a long time It'll be five
years before the scene evenchanges.
Speaker 1 (52:01):
I mean the players we're seeing there are the
players that have been there for25 years or 30 or more, and do
they just basically occupy thatspace in the same way that
studios occupy the space they dofor films?
Speaker 2 (52:16):
Yeah, I mean, it happens in movies too.
You get the Blair Witch project.
Every now and then.
Something like that shows upand takes the entire industry by
storm.
And it was cheap and it wasindie and everyone tried to copy
it.
No one could.
Speaker 1 (52:29):
And they even tried to make copies themselves with
sequels.
Speaker 2 (52:32):
Yeah, and it was just like it happens in movies and
there's lots of indie movies too, of like that would fill the
space of most smaller games, andthey just don't get the site
that the blockbusters get.
They don't get the marketing,they don't get the TV ads, and I
think the same for movies.
It's a big hollowed out in themiddle too.
(52:52):
There is no just.
It's just a movie.
It's either AAA blockbusterquality or it's this independent
movie scene, and I think gamesare just following the exact
same path and that is allfinancial.
Speaker 1 (53:09):
And that is our Popory episode, at least the
first one.
We hope you enjoyed it.
One thing that we have beendoing that you might not realize
is that every episode now has aDiscord thread on a Discord
channel that we have.
(53:29):
So if you go back and look atthe episode webpage which should
be available in most playersyou should be able to click on
that and get access to a Discordthread and channel that we've
created.
Give us some feedback.
What did you love about theepisodes?
(53:50):
What did you hate about them?
Every episode now has its ownthread so we can dive into
particular details and, yes, youwill be talking to Rob and PJ
directly.
We do not have an AI chatbottalking back to you.
Maybe that'll be fun one day,but right now you'll just get
(54:11):
pure Rob and PJ.
So come join us on the Discordchannel, let us know what you
think and we look forward totalking with you.
Welcome back to Tricky Bits with Rob and PJ.
You know a lot of where thispodcast comes from is the
conversations that Rob and Ihave had through the years and
continue to have.
Often we're going out to have adrink, and this episode is
(00:34):
going to be a little bitdifferent than our prior
episodes.
Our prior episodes we tended tohave more of an outline, and
this is going to be a bit moreof a popery episode, but at the
same time it is in many ways aprequel to a topic we've been
(00:55):
discussing for a while, which isthis ongoing discussion about
big and small media as itpertains to games, movies,
television shows, really thewhole gamut, and we've been
trying to piece a lot of thisdifferent stuff together.
We haven't quite cracked it yet, but this is an episode that
(01:21):
gets you into our mindset aboutwhere a lot of these topics
intersect.
Whether at currently it's alittle bit of an experimental
episode for us, We'll see howwell you like it.
Don't worry, next week we'regoing to go back to our outlined
version, so you'll get to enjoythat again, but we're really
(01:41):
curious to see what you thinkabout this particular format.
With that, I'm going to drop you, as they say, in media race
Folks who don't speak Latin.
That means in the middle ofthings.
So one of the things I starteddoing research on yesterday,
after we talked about the bigmedia versus small media and I'm
(02:03):
only at the tip of the icebergright now is we talked about
this hollowing out of the middle, and the research I'm starting
to do is how has the cost of AAAgames changed over the years?
But I want to be able tonormalize it against both
inflation as well as the gamingpopulation.
(02:25):
What is actually the upsidesfor these things and really try
to relate that back to the cost.
Speaker 2 (02:32):
I also think it's the breakdown of the cost increase
of.
I don't think this comes backto the Unreal Engine and do you
have an engine team is fits inthis same thing it used to be.
It was all programs Back in,maybe one artist with lots of
programming time, and over timeit's shifted the other way.
Now AAA games are mostlypipeline and content.
(02:56):
If you can speed up thepipeline of how quick can an
artist see that result in thegame, you get better iteration
and you get more experimentation.
If you have to wait six hoursto see the result, you're only
going to do things you know work.
You're not going to experiment.
So I think a shorter iterationand a better pipeline gives you
(03:18):
better games for one.
And how that affects cost youcan dive into it but you can
just throw in the bullet pointsout there.
But today most of the costtoday is in content and if you
can speed the pipeline up thenyou save a crap ton of money
because most of the people onthe team are content related and
(03:40):
if you can speed all their jobsup, that makes things super
efficient.
So I think the big gains incost reduction are in the tool
side of tooling, the iterationtime, not necessarily in
programming.
I think engine teams have beenstable in size for a long time
and they cost a lot of moneybecause they all experience
programmers.
But in the grand scheme ofthings it's a tiny percentage.
(04:02):
These days, if you spend 10percent on programming on a game
, you're probably doing it wrong.
Speaker 1 (04:06):
So I'm going to challenge this one, rob.
Yesterday we did talk a lotabout how every Unreal game
looks the same right, and so ifwe basically say and I'm going
to ignore for a second yourpoint about spending the money
on a bespoke engine team toupgrade Unreal, Let me just hold
(04:26):
that one off to the side for asecond but if content alone was
the deciding factor and that wasthe only place for innovation,
then I didn't say it wasdeciding factor.
Speaker 2 (04:40):
I said it's much the cost.
Well, it's OK, that's fair.
Ok.
Even a triple A game takesSpider-Man.
Content was much of the cost,so it's not the deciding factor.
It's just that big, moderntriple A games have a lot of
content.
They're expected to be thislong.
They're expected to look thisgood.
That requires a lot of content.
You've got hard surface models,you've got material models,
(05:01):
you've got lighting.
There used to be one person.
Now they will separate.
Separate jobs yeah, Inprogramming you now have a game
programmer, tool programmer andmaybe a network programmer, but
some of this is libraries.
Middleware comes into it.
Middleware and art doesn'treally exist because it's the
theme.
You could use W-wise for audioand get rid of your audio
programmer, or maybe have oneaudio programmer to integrate it
and work with it.
(05:21):
Have it the same thing.
This collision could all be.
You integrate with it, when forart it's a lot more difficult
because whatever you do needsthe aesthetic of the end result
of what the art director laiddown.
So it's much harder tomiddleware art over then using a
third party and giving them thedesign doc.
Speaker 1 (05:41):
Sure, I mean, that's the content houses problem that
we've talked about.
Speaker 2 (05:45):
But bringing it back full circle.
Because of that and because ofUnreal, it is possible to make a
game that doesn't have barelyany programming stuff.
Unreal has all these things.
All these things that bespokeengines integrate with the
programmer audio, physics,collision, lighting, whatever it
may be, third party network andlibraries, ai, path tracing,
(06:05):
libraries, things like that canall be integrated by a program,
but they're all already inUnreal.
So you can do all these thingsthat all the teams are doing out
the box.
Everybody does it the same way.
You get what's in the box andthat's where it looks like an
Unreal game comes from.
Everyone has the same tools.
The best games are the ones youuse those tools best and the
(06:27):
ones that rain Unreal in,because it's very easy to hang
yourself.
There's plenty of rope there todo anything you want and you
can do in side by side.
In the store you'll get effectsthat are the same effect water
on a surface effect.
One of them is real time andone of them is for offline
rendering, and this one's crazyexpensive but looks really good
(06:48):
and you could pull it in yourgame and it will work.
But you're paying the cost andunless your game is entirely
based on wet surfaces and youcan afford the cost, then you're
going to screw yourself.
It's like building a house.
I got a million dollars tobuild a house.
It's like I'm not going tospend it all on the windows.
I need to do the rest of it.
I need a whole house.
Part of your job is knowing thelimit of what's reasonable and
(07:11):
realistic.
What's the cost of authoring,what's the cost of maintaining,
what's the actual performancecost in terms of GPU cycles and
memory and whatever?
And I think the teams that haveengine programmers are far more
aware of this, because theyhave engine programmers who are
looking at this stuff all thetime and looking at performance
all the time, and they'relooking at profiles and they're
(07:33):
looking at memory footprintgraphs and things like that and
they're like oh, what happenedhere?
Oh, that asset got added tothis.
Like, is that a viable thing toadd to the game for the cost?
Speaker 1 (07:42):
Rob, do you think that's something that's missing
in the general Developmentprocesses for everybody, Like
everyone, not just the engineprogrammers, but the gameplay
programmers, the artists, likeeveryone should actually have
that performance window up toreally understand, you know,
what these spikes look like.
Speaker 2 (08:00):
And much of what everyone should.
They're hard to read.
You can misread them and readthe wrong data from them.
I do think there should be aprocess that runs once a day and
goes oh, what happened today?
You have a buyer and you have ascript or code or a person that
pulls out the critical piecesof information in the way that
(08:22):
makes sense for the team.
How team-a does it and howteam-b does it may be different,
but the information that'simportant for the game being
made should be pulled out on adaily basis and graphed Over
time.
You can go, look, performanceis going down as assets get
added.
Someone added shadows oneverything and it's very easy to
(08:43):
add light to that cash shadowsand don't do anything, but you
still pay for the shadow.
So, working with what you have,pulling out that information
and, I think, graphing it overtime, even if you do once a week
okay, last week with 60 hertzan hour, 40, what happened?
If you do that once permilestone and you're like, oh
shit, it's really slow, it'shard to go back and undo what
happened then you miss themilestone.
I think even a week may be toolong.
Speaker 1 (09:04):
I was going to say probably a daily basis.
Speaker 2 (09:06):
I mean yesterday's build, today's build, this
happened and that can be allpart of the automated build
process to make sure it actuallybuilds.
So you have these like look,something happened here and it's
like we can undo yesterday'swork without busting the
milestone.
We can't undo, or we can atleast bubble it up to the top
and be like how important isthis Because it's really
(09:27):
crippling us.
Or can we reduce the quality?
Can we remove it completely?
Whatever it could be, it couldbe a new subsystem in code, it
could be a new asset, it couldbe a new lighting Many reasons
it can happen.
It's not all bad.
It's not all because of badartists.
Speaker 1 (09:39):
No, no, no, no, I get it, it's all about programming.
Speaker 2 (09:41):
There's a lot of reasons and games are so
complicated now that spotting itis straight away Is essential.
Because if something gets builton top of that problem, yeah,
now it's also a problem and itcan't be removed and it needs
what it's built on.
So spotting it straight away,fulton up to the top and
deciding then and there is thisneed, and sometimes the answer
(10:03):
is yes, because you all theseother systems are built on top
of that and that's the cost,deal with it, optimize Later on
in the game.
You don't know what to optimize.
You went for the profiler andthe problem with unreal is
nothing stands out.
You'll be basically, I call itdeath by a thousand cuts, sure,
and there's nothing that standsout in the profile of the times
right in front of you could seewhere it's all going, but
(10:24):
there's literally a thousandtiny things that shouldn't be
there.
Yeah all of them together add up, but any one of them that you
spend a day or two optimizingdon't make any difference.
Not getting those in the firstplace is the way to make unreal
sing in terms of, like the topmetrics that come to mind.
Speaker 1 (10:41):
I mean, is it really Frame rate and memory, or are
there more that are like hey,these are like the, the hot
spots you should really belooking at in that graph, or is
it?
Hey, it all depends upon whatplatforms you're targeting.
Speaker 2 (10:57):
Yeah, it depends on the game.
Ever like, oh look, this worksgreat on good, this new thing
works great on DX 12 and Superbad on DX 11.
It's like how important to take11?
Can we drop support for DX 11?
Does it have to work reallygood?
Can we optimize the game allthe ways?
Oh look, it works great on XboxX, but not an S.
You have to release an S.
(11:19):
What are you gonna do?
Can you take the asset out ofthis build?
Can you just make it simpler?
Same between PC and bothconsoles, like the PC is more
powerful than both the console?
Yep, a lot more memory to.
Should you be using that memory?
Speaker 1 (11:34):
You're you're operating in the triple-a space
and you're saying I was want tobe there.
Speaker 2 (11:39):
We added ray tracing now on the platforms that don't
have ray tracing, so it's notonly is it optimum in the
optimizing, it's like what's thealternative?
Does it have to be there andcan we design the game around
not having this?
And that's part of being a goodteam, and I think this is why
you pull the key rock stars outof a team and put them on a new
team, stadia style.
Then all you end up with is abasically a pit fight of a bunch
(12:03):
of egos, big dick measuringcontest, basically of who is
who's right, and it's like wesolved this problem this way and
we were right because our gamewas successful.
We solved it the exact oppositeway and we were also right
because our game was alsosuccessful.
Take a step back and look atwhat's needed for this game,
which is why teams that sticktogether tend to be better over
time.
(12:23):
The Notre Dame somnax of theworld.
Because of this, like everyoneknows what everyone does and how
they work right, kind of havean understanding of the limits.
I could do this, but it's notwaste of time because it's gonna
get pulled, because it's gonnabe too slow, or that's the ond
project.
I'm like we want this huge,animating, city-sized bad guy
and Engine is not really madefor it.
(12:44):
But let's do it anyway, and Doit the way we would have done it
all, the way art wants to do it, and Then go and look at what's
happening why is it breaking?
Let's increase command buffersizes, let's increase allocator
sizes so it works, and and thengo from there.
I'm like, yeah, naturally it's.
(13:04):
We could do it if we tweak this, this and this, or With the
lightings breaking because theassets so big or whatever it may
be.
It's like we can see theproblems and Address them case
by case if that's reallyrequired.
But that's on D.
Don't do that in productionright.
Speaker 1 (13:23):
But I think it goes to your point of if.
If I want my game to havefeature X and it's a hallmark of
the game, whether it's I wantto fight a giant city or
everything needs to be wetsurfaces, that's fine.
You're making a choice forwhere you want to put that
investment, both in terms of artand performance, right.
Speaker 2 (13:42):
It's like building the house.
Yeah, I want a swimming pool.
These people don't.
There's a cost associated withit.
If I you have a swimming pool,you can have an Olympic size one
and spend all the budget, andnow you've just got a swimming
pool, right, or I wanteverything clad in marble, and
that means I.
Everything else needs to be MDF, and I want a gold toilet right
why not, but and?
Speaker 1 (14:04):
It means that your pipes are made of lead.
Speaker 2 (14:05):
Hey you're, but that's where you put your budget
.
That's fine.
You still have to complete therest of it, right?
That's why you get games on agood tech demos and you get.
That's how games get funded andthat's ultimately what guts the
middle, the middle class ofgaming that's hollowed out right
now is because of that, and Ithink what real does play a big
role in In that.
I just don't know how.
Speaker 1 (14:27):
I do too.
This goes back to that dataquestion.
I found a great site where theylook at the cost through the
decades of, let's say that, thecost of a console and the cost
of its game, where, like thecost of a console in In 1977 of
the Atari 2600 was 200 bucks,which in 2020 dollars, which is
(14:50):
when they were doing thisarticle was just shy of a
thousand bucks, and the averagegame cost in 1977 is 40 bucks,
which translates to 170 in $20,and so there's a part of me that
really wants to say that Iactually think, as much as we
talk about, oh, game prices areincreasing, I Think there's
(15:13):
actually been a massivedeflationary effect in games and
I suspect what is actuallysustained it has been a larger
game population.
Speaker 2 (15:22):
But I don't know, and this is like the thing I want
to do.
The data on this a hundredmillion playstations.
Well, threes, fours, fives willbe there soon, are you sure?
Yeah 75% of them are activelyused.
Every every now and then.
That's a lot of.
That's more than right.
What every was 2600 and yes,games cost a lot more to make,
(15:45):
but there's a lot more peoplefunding them and I think it
balances out.
I mean, the big game companiesare not hurting.
Speaker 1 (15:51):
Oh, they're not.
I'm curious.
I said this again.
This is like the Marvel movieproblem that I'm saying.
Speaker 2 (15:56):
Well, I'm curious of the flip side of that.
I want I want to see statsNumbers from the games that
failed.
Sure, like, okay, this game gotcanned.
Why was it purely financial?
Was it never gonna technicallywork?
But the financial ones of like,how much did you spend on what
and where, and how did you nothandle the Monetary budget but
(16:17):
also the technical budget, whatpath was taken together today?
I think there's a lot to belearned.
That no one talks about it, noone does Public postmortems of
like this is where we screwed upand we went down this path and
we should have gone down thispath.
There's a lot of informationthere that you could really
learn we could use Duke Nukemforever as our case study.
(16:39):
Yeah and uh.
But even big games are like whydoes it?
Why did EA?
Okay, why did Sony just closethat studio down in London?
Obviously some thought wentinto it like kill the whole
studio, don't just let somepeople off, just get rid of the
whole game.
What state was it in?
And I think there's a lot tolearn there.
But you only learn this bybeing involved with your
particular game and Even, maybefor the team, don't have access
(17:00):
to this info.
So it's hard to learn frommistakes because they never
talked about.
Yeah, it's like came, came, go,shut down last a little bit
last year of it.
A leak of it may happen andyou'll look at it and I go is
actually look pretty cool, butthat's the only Insight that
you'll get be a few screenshots,maybe a leak of the of a
running game, of video.
You'll never get like a studioto sit down.
(17:21):
But like okay, numbers,technical numbers, hiring,
firing like that level of detailwould be a huge win for the
industry and I think it wouldshow that a lot of these smaller
indie, middle ground studiosare doing it wrong.
They're all on that path tofailure.
Some make it, some don't.
Speaker 1 (17:39):
The premise they're all on that path to failure
because the economics they'reworking with is is there,
approaching like a triple A gameor what's the?
What's the fundamental mistake?
Speaker 2 (17:48):
I think they're all on that path to failure and some
some happen to survive.
I think the path is the nextpaycheck is not guaranteed.
It's like miss a milestone.
What happens is like somepublishers are real bad about
miss a milestone.
We're not gonna pay you if youdon't pay you.
If I don't pay my staff, you'renot gonna get the next
milestone.
There's a lot of viciouscircles that come up when you
hit these, like hey, it's not upto performance, didn't quite
(18:11):
check all the boxes on themilestone checklist, you can
literally put the finger on okay, this missed milestone cause
this miss payment, which callsthis entire cascading effect.
It's just we don't have thatlevel of information and I wish
companies would just take a stepback and go okay, here you go.
He's the numbers.
Why are we so concerned abouthow much games cost?
The insomniac leak leaked abunch of money.
He's off 200 million for a game.
(18:31):
It's like it's just a number.
Yes, it's a big number.
If you're paying, it's a lot ofmoney, but it's also an
investment in the future.
Yeah, you're hoping to get areturn on that money.
So why?
Why is it all hidden?
Why do we care if insomniackeep their numbers secret or if
they publish them like whobenefits or Loses from those
numbers being published.
(18:52):
Maybe insomniac themselvesbecause competitors could be
like what we can do a half aprice, but can you?
On paper you can, yes yeah,that's the thing again.
Speaker 1 (18:59):
I will keep going back to my Marvel movie example
because, like, you'll dump 200million dollars on Infinity War
but if it makes a billion backfor you, great, that was a good
investment.
So if you're putting in 200million to insomniac, to make
spider-man or spider-man 2,guess what, you'll probably make
back that investment.
You know, if you have it's what50, 60 bucks a pop.
Speaker 2 (19:23):
I mean across 100 million machines the money I
mean in theory is that.
I mean, if you sell 10 millioncopies, yeah, that's a crap,
crap turn of money.
At 600 million dollars, yeah,that's retail, though which
Obviously there's all theBusiness tricks to that.
They play up like if I publisha pace for a game, they don't
(19:47):
pay off the development costswith retail money.
They pay off the developmentcosts with the royalty you would
have got.
So once you've paid it say youget, yeah, so the game cost 10
million and you're getting adollar per copy royalty.
Yeah, yeah, you've got.
(20:07):
Sell 10 million copies to payoff 10 million, right, oh right,
retail may just get to go tothe store.
It goes to Publisher, blah,blah, blah.
But what happens when they'reall the same person?
Like if I buy a game fromPlayStation Online, if, like,
there is no store, 50% right togo to the store.
Sony is getting that now, butit's not Sony development, it's
(20:29):
Sony store, which isn't goingtowards development costs.
So if there's a lot of fingersin the pie, which right, which
make the numbers a lot less rosythan you'll think when you look
at all, this took a billiondollars at retail in a week.
It's.
It's not that pretty for thebottom paper at the bottom,
which is always the developers.
The developers are always atthe bottom of the pile so.
Speaker 1 (20:51):
So this is, this is the Classic problem.
Speaker 2 (20:54):
Oh, yeah, it's exactly that.
Speaker 1 (20:56):
Yeah, this is Pay for that term before.
Speaker 2 (20:59):
It is the same people or this they hang out the same
thing and Same people.
In a lot of the times, ifeverybody was in it for the same
purpose, like retail money,would pay Everything off and
then would be divided up.
But it's not everyone wants tothese.
This group wants their cut fromthe first case copy sold, where
this team doesn't get a cut, tothe 10 millionth copy sold and
(21:22):
and yeah there's the problemthat is the Hollywood accountant
, but that is also part of theproblem.
Yeah, a lot of the money doesn'tfeed back to developers To do a
second version or to super upthe ante on the next version of
whatever it is.
We've got a lot of money comingin.
We can.
We can do it, and it also meansa lot of games would break even
before they actually do.
(21:43):
There'd be a lot moreblockbuster games.
But I mean, I guess right nowgames are like movies.
You'd get more triple a gamesbecause there'd be more money in
those triple a games if theywere paid at Retail value or
these development costs werepaid at retail and then the
final money was split.
You have to sell yes butobviously that money is coming
from someone else's pocket.
Who currently keep wants tokeep it, which is why the status
(22:03):
quo isn't gonna change.
It's not like there's moremoney to go around.
It would just be allocating itmore fairly.
Yes, and I think it's fair tosay no one makes any money until
all the expenses are paid, butthat literally isn't how it
works.
Speaker 1 (22:16):
You have the Sony store, you've got the Sony the
publisher, you've got Sonylicensing right, which is the
one who would have the rightsfor spider-man.
Speaker 2 (22:25):
And you also have the Sony license fee for the
PlayStation of like oh, everygame has a Built-in fee that you
have to pay even even ourfirst-party studios, even if
you're Sony here.
So yeah, it's just because Iguess they run is totally
autonomous divisions and it'slike, well, that's our
accountant, this is our income,that's our outcome and there's
(22:46):
the way it is.
But nice if it didn't happen.
You'd get a lot better games Ifthat wasn't the case.
But the financials are goingdown the movie path.
Games are following Hollywoodand the counting is what it is.
There's a lot of money in thegames Only if you can overcome
that hurdle, and a lot of gamesmiss that hurdle are great games
, but never make that hurdle andnever pay off the development
(23:07):
costs and they never get anotherone made the team this bands.
Blah, blah, blah.
Speaker 1 (23:10):
But this is why we have really this, this migration
up to the top of Only triple agames are making it.
Speaker 2 (23:18):
Yep, that's what happens, the guys who make the
development payback hurdleProgress to triple a, the ones
that don't either disappear,which hollows out the middle, or
they migrate to a lower end,down towards the mobile space
that basically draw from thecenter to both ends.
Is what's hollowing out the thethe middle there's a lot of
(23:38):
other reasons for it too.
But I mean, we just talkedabout unreal, about the
financial reasons, they.
They were all basicallyhappening at the same time,
which is a big problem for themiddle of gaming, did we?
Speaker 1 (23:50):
invent the lower end in the last couple of decades by
virtue of changing thedistribution model.
And what I really mean by thisis that before I had to have
physical media, you know, I hadto have a DVD disc or a blu-ray
disc to play Ration, clank orresistance, and that was a cost
(24:13):
that was there and it had to gointo a big box store.
Did we Invent the lower end byvirtue of changing distribution?
So we have an Apple App Store,we have steam.
Anyone can put their game outthere now and be distributed.
It may not be successful, butit can be distributed.
(24:34):
So did we invent this by virtueof how we change Just how we
even distribute these games out?
Speaker 2 (24:39):
I don't think one fed the other.
The app store was gonna happenregardless, independent of any
game thing, and I think itverified that digital
distribution was possible andthe big game still struggled
with digital distribution whenit's 100 a gig to a download.
We're definitely going downthat path.
(24:59):
But it does open up thatquestion that you mentioned of
like there's more money now Inthe pie because we're not paying
the retail store and we're notpaying this.
So there is more money acrossmore gamers, which is how the
games do get funded.
So a lot of that cost thatwe're not making DVDs and
(25:20):
there's millions of dollarsthere for millions of discs and
that cost is now up for grabs,for example, that does get split
up better across to the variousteams.
It's not just going insomeone's pocket, but there's
also a cost of the back end ofruling the online store, things
like that.
The physical media is whatcauses the yearly cycle.
In the PlayStation 1, which is2, you add Madden and NFL 1995
(25:46):
and it's no way to update itnext year when all the players
have moved around and all theteams have done Whatever they do
, and stadiums get built andknocked down and things like
that.
So you now need my 96, becauseNFL is now a different world.
There's no way to update that.
It basically a year by year.
It keeps up with what's goingon in the real world.
Speaker 1 (26:04):
Most of the same game engine, just Couple of art
changes.
Speaker 2 (26:08):
No, the engine does get better for sure.
I mean, you look at the diffacross them all.
Go back to the Sega Genesis andit's like there was huge steps.
When we switch platforms, thepoint is they have to do a year
release because there isn't away to get any more data on to
that PlayStation.
Right Today's world witheverything being online, I Think
they're taking the piss doing agame every year of like madden.
(26:31):
If, like, you could justdownload that database, it'll be
fine, but they won't.
We've already got this modeland I think that model feeds the
other big franchise model,which is why we get a call of
duty once a year.
It's why we get one of all.
The big blast dust is once ayear Because it goes back to
those sports games on thephysical media, on the machines
you couldn't update and for, forEA and for these games that
(26:54):
have a brand name.
Speaker 1 (26:56):
It is in the Economic interests of those companies to
keep making you pay 60 70 bucksa year To get an update.
Speaker 2 (27:06):
As you say, that could just be a download, yep
and it's Probably Charles six orseventy books for the download.
So you know the end customer isnot winning regardless.
Downloads Could accommodate alot of that, or you can play
this year's teams on last year'sengine, for example.
There is definitely like a Lotthat could be done to fix the
(27:28):
yearly updates, but there's nowhy people buy them regardless
and it's another billion dollarson the table.
Maybe we'll take care.
Speaker 1 (27:36):
So let's look at the flip side of this, though.
With the rise of digitaldistribution, did we create a
lower end, meaning like therewas perhaps a Lower bar you
needed to hit when you hadphysical media because of just
the cost of it I and with whatamounts to zero marginal cost to
(27:57):
distribute a game.
Did we craft this lower endwhere it know, where it couldn't
exist previously?
Speaker 2 (28:06):
I think just the normal distribution of
Developers crafted the bottomend.
Some point you have to say,like, if I started a game
company today, I couldn'tcompete with no a dog in
insomniac.
Yeah, no matter how good I amor matter how good a team I hire
, I can't compete with them.
Again, it's the, the legacythat they have, and the, the
(28:27):
ability to work together Doesn'tmean you can just create a team
and replicate it.
So when would I go if I startedthe game today?
I wouldn't be triple a.
I Don't think I would aim forthe triple a.
It'd be much better to go andlower end.
Go lower than where you belong,because then you'll stand out
(28:49):
has been really good and I thinkthat's why a lot of really good
Developers went to the iPhoneand things like that.
But don't kid yourself that thezero cost, the cost per
acquisition on an iPhone andgetting noticed in the store, is
incredibly expensive and it'snot much better in the
PlayStation store, the Xboxstore or steam.
(29:09):
So Although it is zero cost toget it in the store, to get it
out of the store onto people'sdevices is an expensive thing,
and Is it worse than marketingtraditionally?
I have no idea.
It's not what I do, but butthere is a cost there and but I
(29:30):
do think that also the lowerends always existed and and it
could exist on the same platform.
The Commodore 64 had greatgames and crap games.
Speaker 1 (29:42):
Some of the crap.
Speaker 2 (29:42):
Games were done by good developers and some of the
good games are done by unknowndevelopers.
And you add low-end, reallygood games and that's still the
same today.
And I think having the abilityto have a phone and different
types of devices Moves thedynamic of the high and low end.
You always say high ends, likeall PlayStation triple A's, like
(30:04):
that's the high end.
What is it?
You can do a lot of this on aniPhone and iPad or an Apple TV.
You'd be better off making iton a PC or an iPhone and then
using that to get onto thePlayStation, but by then your
games already out on the thephone platform.
So I think it gives an avenuefor development.
It gives a very nice avenue forexperimentation.
Speaker 1 (30:21):
Yeah, it's.
It's interesting you bring upthe Commodore 64 example,
because when I think about thebarrier of entry, I actually
think about it more in the Atari2600 NES era, where I needed a
fairly beefy amount of physicalmaterial between the cartridge
and the electronics to actuallyget my game out there, which
(30:46):
felt like it created this, thisbarrier.
Oh, it was contrast to theCommodore 64 where it's like,
hey, a floppy is pretty cheap,even back then, you know, as
long as I can get the data on toit, I can, you know, sell my,
you also have the free path thentoo.
Speaker 2 (31:02):
I mean, you could send your game to a or demo of a
game to a magazine, and they'dput it on the.
Sure shareware disc or classic,yeah, leave a shareware demo of
like oh, he's a game, it's ademo of the first level, yeah,
and buy it later.
Shareware to us.
How it how it all all started,sure, yes, the cost of entry on
the consoles Today is nothingcompared to what it used to be
(31:24):
today.
You got a dev kit which isreally just a modified version
of the base kit.
Back there, you add, the devkit was pretty much the hardware
.
There wasn't a differencereally between a Sega Genesis
that was used for development,one that wasn't because there
was no software in the Genesisper se, it was just a cartridge
slot.
But you would have a cartridgethat had ram in it, that
connected to a PC, and youdownload the game to that, and
(31:47):
every now and then you'd burnone to a ROM and stick it in to
make sure it actually worked.
And then, as we got to the, thedisk based systems, that option
became a lot fuzzier.
I was like how do you get thecontent onto the device?
when it's a disk based systemand Doesn't have a disc, the dev
(32:08):
kit might have a drive.
You're not gonna burn a discevery time you build a code.
So all the console, all thecartridge based consoles, pretty
much had RAM based, flash basedthey were.
They were slightly differentbecause they had debug support
and hardware level debugging buthalf of it would work on a
retail kit.
It weren't until we got to thePlayStation had they started to
(32:31):
have disk emulation.
So they had file systems thatwere on the PC and it would go
over a high speed high higherthan CD ROM speed connector to
the, the actual dev kit in thecustom operating system.
Because I had to read from thefile system and not from the
drive.
And Then that gets us to wherewe are today with them.
(32:53):
These things became signed andthey'd sign them differently for
retailing dev kits and Sure theonly difference today between a
retail and a dev kit is mostlythe signatures needed to get
stuff to run.
Speaker 1 (33:06):
Yeah, so I mean it's.
It's really a Encryption issueat that point in time and, like
we've talked about in the past,the hardware that is either for
the PC or the Xbox.
It's basically a PC and thisgoes back to my, my point of
like to have in these cartridgebase or even early disk based
systems where the dev kits wereexpensive or, you know, burning
(33:27):
a ROM was expensive, like youhad to have a fairly Large
upfront capital cost to getinvolved, so it didn't allow.
It doesn't seem like it wouldallow indie developers to get
into that field.
Speaker 2 (33:40):
There wasn't any indie developers on the on the
NES.
The NES in particular was brutalto get dev kits for.
Like.
Nintendo have opened up a lotrecently, but in the early days
they were Absolutely brutal toget anything on and but yeah,
you have to have.
They wanted to see financialsof like can you afford to make
the cartridges?
And they made the cartridges.
(34:04):
That's how they got the licensefee and that's where the
license we came from originallywas like paying more than you
should for having thesecartridges made.
And, obviously, the whole storyof EA, of where an Activision,
how they popped out, of Atari,because that's how the whole
third party world showed up.
There's a whole story therewhich been told many times
(34:25):
before.
But yeah, so you'd have to havethe upfront cash to buy, make
these things.
Indeed, it wasn't cheap.
It was many dollars per thing,especially with the box and
everything.
And If you're making a hundredthousand, you might need Three
quarters of a million dollars, amillion dollars, whatever it is
(34:46):
.
Just to make it a set tape or adisc was fairly easy, but still
a physical thing to manufacture.
Today there's none of that.
You can literally just pass thestore checks, the various
quality control checks for anyplatform.
Pc has steam, the Xbox,playstation have their
respective stores.
Yeah, apple, android have theirrespective stores.
(35:08):
You can basically make a gamewith zero.
You have the development cost ofthe game, but then after that
there's zero cost to get it intoas many people's hands as you
want.
There's no ongoing cost.
You could count the acquisitioncosts, but that's really
marketing.
Speaker 1 (35:23):
That's marketing and that was always the case, though
that's why I'm separating thatout that there's in terms of
distribution costs nowadays,digital effectively makes it
zero, and and this is also goback to your original question.
Speaker 2 (35:36):
I was like why are games in 2020 four dollars less
than they used to be?
Well, that's a factor thereright.
Speaker 1 (35:43):
Probably higher margins by a long shot for these
companies, since they don'thave to actually manufacture
units.
What if I'd fascinating is likewe go back in time, we get EA,
we get Activision and we getinsomniac naughty dog, all of
whom came out of either console,cartridge discs, times and now
(36:07):
occupy the space in the triple aspace.
I and I'm kind of wonderingwhere was the middle previously,
if not in the cartridge worldbut maybe in the console world,
like where was our middle?
Because I'm kind of wonderingif we always had this, like the
AAA problem was always therebecause of the cost of
(36:27):
development historically, and weonly have recently acquired
this notion of lower to middleend by virtue of the fact that
it's a lot cheaper to make thesethings.
Speaker 2 (36:38):
I think here, it's a regional thing.
In the US you always had the2600 and at the other side you
had the.
Apple II, for example.
The UK didn't get the 2600 andit didn't really buy into the
consoles until the much laterSega Genesis era is when the
(36:59):
consoles in Europe got popularand they still were different
because of the frame rates.
The UK always had the homecomputer scene.
It had the BBC Micro, it hadthe VIC-20, it had the a few
random the Auric and the Dragonand the Amstrad and the.
World.
That's way later.
I'm talking about the earlyspectrum, of course, and then
(37:22):
that became the Amiga and theArchimedes, amstrad, and
Spectrum kind of went away.
The 64 was in there too, ofcourse.
I forgot that one.
But all these home computers,and everyone was writing
software Consoles were not athing at all until much later,
in the early 90s.
So from a UK point of view,there was always this spread of
(37:44):
quality developers.
Speaker 1 (37:46):
Some were crap, some weren't.
Speaker 2 (37:48):
And we didn't have that overhead cost, we only had
the discount cassette costs.
The US was very different.
The US definitely went down theconsole path.
The bigger crash of video gamecrash of 83 was really the Atari
crash of 83 and only affectedthe US.
The rest of the world justcarried on as normal, just
within the game console.
Speaker 1 (38:08):
Interesting.
Speaker 2 (38:08):
So I think your view of that, my view of that, would
be very different.
Speaker 1 (38:12):
Well, I think it's interesting because, from your
perspective, the existence ofthat middle tier was perennial.
It was always there by virtueof people who were just
programming on these things,versus, as you say, my view,
which is that we had theCommodore 64 which had those
kind of qualities, sort of byAmiga, but then, effectively,
(38:35):
there's this giant step functionwhere you enter the console
world, where these capitalexpenditures are huge in order
to get involved.
So it actually eliminates theexistence of a middle tier for
the US, and I think the US hadmore money generally, so they'd
be like oh well, we'll do this.
Speaker 2 (38:54):
Money's easier to combine the US than it was in
the UK in the 80s and people inthe UK were happy with the
machines they had.
They were happy with the gamesthey ran.
There was the quirkiness of,like I said, NTSC, PAL update
rates.
So a lot of games made in theUS didn't actually work very
well in the UK, which is theprimary reason that the 2600
wasn't there at all, because youhad to program so close to
(39:15):
chasing the beam that PALliterally changed everything.
So they just didn't do it.
Playing American content wasn'tthe easiest thing, so they just
made their own and it was allon computers.
There's that whole spread fromthe ZX80 in black and white all
the way up to the Amigo and theArchimedes at the end of the
1632-bit days, and ultimately wealso switched the PC a lot
(39:36):
later.
I think the PC in the US wasalways a bigger factor than it
was in the rest of the world.
It wasn't until Windows 95 thatPCs in the UK really became
like oh, just get a PC.
Speaker 1 (39:47):
It was interesting because for us in the US in the
middle of the 80s you had threeat least in my mind, three
contenders.
You had kind of the Windowsslash PC world because IBM still
had its own IBM DOS for a longtime.
You had Mac and then you hadCommodore and we as a family and
(40:09):
Commodore was the smallest butI mean, well, I take it back
Commodore 64 sold really, reallywell.
So we had a Commodore 64.
We actually had a couple ofthem and then I later, in the
early 90s actually, as I wasentering high school, got an
Amiga 500.
And so it was still kind of incontention for a little bit.
But there was this like a curlthat was happening where,
(40:33):
especially like Windows 3.1coming out and then Mac started
like taking a lot, and thenCommodore made a few mistakes
where you basically only had abifurcation of choices a Mac or
a Windows box.
Speaker 2 (40:47):
And that's the same today for the most part.
But I also think there's a bigdifference in how the machines
were used In the US.
Commodore was very much aimingfor the business users.
They weren't any real greatAmiga like push the envelope,
Amiga games coming out of the US.
There's lots of businesssoftware coming out of the US
for video toasters and thingslike that.
(41:08):
Where the UK didn't do thebusiness side so much Like, the
Amiga in the UK was very much agames machine.
It wasn't really used for itsoperating system or it's the
other things they could do.
It was in some situations butthey were very much a games
machine.
Apple wasn't a thing in the UKin 80s and 90s.
(41:30):
We didn't have the Atari so wehad, like the Commodore machines
or the A-Cord machines, theSpectrums and the Amstrads and
things like that and gaming wason the forefront of all these
home computers in the UK whereit wasn't in the US, I would say
a lot of the gaming that cameout on the 64, on the Amiga from
the US was very bland.
Speaker 1 (41:46):
I'll admit that there were a lot of stuff that I did
with the Amiga that had nothingto do with games.
Speaker 2 (41:52):
Yeah, and the opposite in the UK.
So I just think how thesemarkets just merged together and
now it's I think it's where alot of good programmers came out
of the UK.
A lot of game programmers inthe US today are from the UK.
I think that's part of it.
Unfortunately the UK lost thatposition.
Back then they had some greatand this has nothing to do with
(42:14):
games today, but it had the.
The BBC obviously came out thewhole story of how it was
related to the computer programand they were going to make the
whole country computer literate.
And they did make the computerprogram and they did teach
people to program and all thatand I think it literally took a
whole generation and made themgreat programmers.
And later on that just kind offaded away.
(42:35):
And if you do computer sciencein the UK, if you do it in high
school, it's more or less how touse a computer Like how they
work is just not taught anymoreand back then it was very much
like this is how it works, don'ttinker with it, the whole.
And now it's obviously that's40 years ago when I was part of
that whole 1980s computerprogram mentality.
(42:56):
That mindset's gone and it'sunfortunate, I think I don't
think the UK's kicking out greatprogrammers these days, like it
was 40 years ago.
Speaker 1 (43:05):
Really unfortunate that we're not teaching the
lower level stuff.
Earlier on I saw a book in thelibrary where it was like, hey,
teach kids how to program andit's it's Python basically, but
that's not going to really getyou into the guts of the
electronics that way.
Speaker 2 (43:21):
It's not a bad way to go.
It's a different time, isn't it?
Back then, you had to get intothe guts of the electronics to
make anything happen.
The OS didn't get in your way.
Today, it's so difficult toeven draw a triangle on a screen
.
You need to understand DirectX.
You need to understand aWindows app.
You need to understand theconcept of brain buffers.
(43:42):
You need to understand visualstudios.
They're a very complicated tool.
You need to understand thebuild process.
There's a whole lot of things tolearn and a lot of boilerplate
code.
You have to write a shader.
You have to understand the 3Dmath just to get a triangle on
the screen.
Where there's a lot of that,that won't work first time.
If I wrote a single triangle onthe screen right now, I'd have
(44:04):
to run it a thousand times toget the damn thing to work.
What did you learn?
You learned to put a triangleon the screen.
You could do that Python too.
In Python, you can just go drawa triangle and use some library
and bang there it is.
It's very much like the olddays of instant gratification.
I think Web has some of theseproperties today too, where, in
basic, you could go draw arectangle.
(44:24):
Boom, there it is on the screenand tweak it, run it again.
It was just like instantaneousfeedback.
It wasn't this massiveknowledge base.
You needed just to draw onesquare on the screen.
The amount of code, the amountof work it takes to get
something to be in the placewhere you can experiment freely.
(44:44):
It's ridiculous today, which Ithink is why, even for things
like experimenting with 3D anddrawing graphs and things like
that, I think is why Python isused, because it's there and
it's easy and it is quitepowerful.
I mean, mackiddy wasexperimenting and it's all
Python and they use some librarywhich deals with all the
(45:05):
boilerplate and it just makes itinstantaneous.
You can say draw a triangle,draw a triangle, draw a triangle
, and they just appear exactlyas you said.
You could say plot circle,there's a circle.
It's not fast, but that instantgratification is how you learn.
It's like you'll get into theperformance.
And how would I make a gamewith this much, much later?
But today you kind of have tostart in the middle, you can't
(45:26):
start at the beginning.
Speaker 1 (45:27):
You can't, I mean.
So the good news bad news iscompared to when we were coming
up on this stuff.
I mean, at best we wereexperimenting on ourselves,
which could be great, andgetting it from magazines.
The good news bad news is thatall this information is out
there on the internet.
So if you had the same drivethat we did, all that info is
(45:51):
there to start playing aroundwith stuff.
Right, I mean, we're notlimited by, basically, lack of
information sources today.
Speaker 2 (45:57):
Yeah, but that all takes the entire conversation
full circle of why not just useUnreal it does.
It's like I could concentrateon the behind interest.
If I'm a gameplay programmerand I'm all by myself, yeah, and
I want to play with gameplaythings.
Do I have to write a 3D enginefirst so I can see what I'm
(46:19):
doing?
It's like some gameplay youcould do with, like text and A
star things like that you can dowithout any graphical support.
But generally it's like a lotof gameplay today needs the 3D
or the 2D or whateverenvironment of the game is
before you can do anything.
So that's what takes a lot ofindividuals to Unreal.
(46:42):
And I think it comes back fullcircle as to I'll just use this
because everything is alreadythere.
It only matters when you careabout what's there.
If I just need a 3D enginebecause I'm working on some
really cool fighting dynamics inanimation, I don't care how
fast the render is, I don't carewhat it's taking.
I got my dynamics, I can dowhat I need to do and I can add
(47:03):
my physics and I can do this andthat you only care about the
bit you're interested in, and Ithink Unreal is fantastic for
that.
It's got a lot of people intogame development that would
never have been there.
I think Unity is the same too.
There's so much boilerplate,there's so much that has to
happen to get to where you wantto be that you just use one of
(47:24):
the engines.
Speaker 1 (47:24):
And this goes back to our points that we made earlier
about where do you want toinvest, and we talked about it a
lot in terms of where do youwant to invest your development
dollars, but in this case we'retalking about where do you want
to invest your development timeas an individual, learning these
things?
Speaker 2 (47:38):
Yeah, An artist on the wheels, perfect.
Where else are you going to beable to do real time 3D modeling
and get an idea of what theperformance is?
So, yes, you can do it in Maya,but the performance is offline.
So again, I think, on one hand,unreal is awesome.
On the other hand, it's got alot to ask and so forth, what
(47:58):
it's done to the market.
Speaker 1 (48:00):
We've done a lot of tracing back in time.
I could make a proposal that welook at what you talked about
in terms of UK, where you had aspectrum of programmers that
were there and you were able tosend it to a magazine.
You were able to have somelevel of distribution.
I have this kind of conceptthat again I'm coming from this
(48:22):
perspective that the US marketwas really locked up in terms of
games and distribution for avery long time.
It was big box stores, it wasbig consoles.
There's a lot of capitalexpenditure.
I look at today and it's not tosay that this justifies anything
, but there is more opportunitytoday, it seems, whether or not
(48:45):
you use Unreal or not, where youhave the steam distribution.
You have Facebook and Instagramfor marketing.
You have these tools that havenever been available to anyone
like previously to actuallycraft a game and distribute it.
It doesn't necessarily meanyou're going to be successful.
(49:05):
Again, the marketing side ofthat is there for everybody.
The difficulty in making a goodgame is there for everybody,
but I'm wondering if actuallywhat we've come to today is
there is more opportunity in themarketplace, but it definitely
requires work on individual'sparts.
There's no trusting that somebig game publisher is going to
(49:27):
come in and swoop down and saveus.
We have to do it ourselves,yeah.
Speaker 2 (49:33):
That's a take.
I guess the market's mucheasier to get into today because
of the likes of Unreal andthings like that.
You can be a bunch of artistsand make a game, and it has
happened, I'm not even trying topraise Unreal and Unity at this
point in time.
Speaker 1 (49:47):
I look at them as tools, the idea that my iPhone
is a dev kit now for iPhone andmy PC.
Basically, I could bedistributing PC games regardless
whether or not I use Unity orUnreal.
Speaker 2 (50:02):
The distribution cost is gone.
There isn't any distributioncost.
There is because you have topay fees to these stalls and
things like that.
You can make it free and thendon't pay.
But that distribution cost nowbecomes server costs of like.
These stalls are not cheap tohost.
There's a lot of data beingmoved around.
Speaker 1 (50:21):
Yeah.
Speaker 2 (50:23):
I guess right, but to you it's zero cost.
If you make a free game in theApple Store, it's free to you to
pull up there.
It cost Apple to host it and itcost Apple to check it.
I think this bottom end of theiPhone and the whole how it
introduced free gaming and likepay to play type gaming, and
(50:45):
there's a whole story there too,which is a whole mess.
I think that was bad for gamingand I think that's trickled
into big games with the lootboxes and things like that and
obviously there's been a hugestink over there, lawsuits etc.
And it's not to say the bottomend.
Games are bad.
I mean, there's been lots offantastic games that have came
out of the iPhone that areactually just fun games and I
really appreciate that.
People experiment down there andif steam too, to some extent,
(51:10):
less so on the consoles.
For a while there was an indiescene on the consoles and it
seems to have just gone away.
I mean there used to be asection in the store.
Now they just in the same storeas everything else and you'll
never find them.
It's a healthy market in someways.
It's just there's no middleground.
It's gone.
You either make little games,experiment down there and hope
for a success and they can bevery successful.
(51:31):
Or you try to make a team andaim for AAA and ultimately you
might get your funding cut andyou never make it.
If you get over that fundinghurdle and get over that payback
royalty hurdle, welcome to theAAA space.
You were made it, but therearen't many going in and out of
(51:54):
that space.
It's fairly static and it takessuch a long time It'll be five
years before the scene evenchanges.
Speaker 1 (52:01):
I mean the players we're seeing there are the
players that have been there for25 years or 30 or more, and do
they just basically occupy thatspace in the same way that
studios occupy the space they dofor films?
Speaker 2 (52:16):
Yeah, I mean, it happens in movies too.
You get the Blair Witch project.
Every now and then.
Something like that shows upand takes the entire industry by
storm.
And it was cheap and it wasindie and everyone tried to copy
it.
No one could.
Speaker 1 (52:29):
And they even tried to make copies themselves with
sequels.
Speaker 2 (52:32):
Yeah, and it was just like it happens in movies and
there's lots of indie movies too, of like that would fill the
space of most smaller games, andthey just don't get the site
that the blockbusters get.
They don't get the marketing,they don't get the TV ads, and I
think the same for movies.
It's a big hollowed out in themiddle too.
(52:52):
There is no just.
It's just a movie.
It's either AAA blockbusterquality or it's this independent
movie scene, and I think gamesare just following the exact
same path and that is allfinancial.
Speaker 1 (53:09):
And that is our Popory episode, at least the
first one.
We hope you enjoyed it.
One thing that we have beendoing that you might not realize
is that every episode now has aDiscord thread on a Discord
channel that we have.
(53:29):
So if you go back and look atthe episode webpage which should
be available in most playersyou should be able to click on
that and get access to a Discordthread and channel that we've
created.
Give us some feedback.
What did you love about theepisodes?
(53:50):
What did you hate about them?
Every episode now has its ownthread so we can dive into
particular details and, yes, youwill be talking to Rob and PJ
directly.
We do not have an AI chatbottalking back to you.
Maybe that'll be fun one day,but right now you'll just get
(54:11):
pure Rob and PJ.
So come join us on the Discordchannel, let us know what you
think and we look forward totalking with you.
Popular Podcasts
Las Culturistas with Matt Rogers and Bowen Yang
Ding dong! Join your culture consultants, Matt Rogers and Bowen Yang, on an unforgettable journey into the beating heart of CULTURE. Alongside sizzling special guests, they GET INTO the hottest pop-culture moments of the day and the formative cultural experiences that turned them into Culturistas. Produced by the Big Money Players Network and iHeartRadio.
On Purpose with Jay Shetty
I’m Jay Shetty host of On Purpose the worlds #1 Mental Health podcast and I’m so grateful you found us. I started this podcast 5 years ago to invite you into conversations and workshops that are designed to help make you happier, healthier and more healed. I believe that when you (yes you) feel seen, heard and understood you’re able to deal with relationship struggles, work challenges and life’s ups and downs with more ease and grace. I interview experts, celebrities, thought leaders and athletes so that we can grow our mindset, build better habits and uncover a side of them we’ve never seen before. New episodes every Monday and Friday. Your support means the world to me and I don’t take it for granted — click the follow button and leave a review to help us spread the love with On Purpose. I can’t wait for you to listen to your first or 500th episode!
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com