[NTLUG:Discuss] Wine presentation

[Steve Baker]

Steve wrote:

> You sorta make it sound like Carmack's team, full of artists and
> programmers rather than the evil money men, took a gamble and were
> rewarded from it.

No - far from it.  They did it "for love" and were *not* rewarded.
They actually sold less Linux copies than the low estimate they made
up-front.

> I think what happened was that the only reward they
> got was doing something for Linux.

Yes.

Carmack clearly does love Linux - you can see that in all the efforts
he put in to the Mesa drivers for the G200 and G400 cards.  Those systems
had lack-lustre drivers - partly because the implementation wasn't great
and partly because Matrox had failed to provide documentation for a critical
'fast mode' that the card uses under Windoze.  Carmack not only polished the
drivers to the point where they worked well - he actually got them to go
faster than the Windoze drivers *without* putting them into the mysterious
'fast mode'.

That was a direct contribution to Linux from a *very* busy man...so there
must be some love there!

> There wasn't an economic motive
> behind it. If there is any sort of opportunity cost for their efforts,
> the company lost money. I think Carmack knew this going in. Wish I could
> find the interview where he talks about this (also to be sure that I'm
> accurate on the views...).

Yep.
 
>  > Games programming is generally a ferocious business (which is why I don't
>  > do it for a living - preferring to write games for fun and give them
>  > away - taking a more relaxing day job instead).
> 
> No reason why a simpler game than Quake, something like Myst or Tetris,
> something that relied more on concept than just sheer brute-force pixel
> pushing, can't do well on Linux financially. But today's games more and
> more resemble the movie industry for better or worse.

There are over 50 copies of Tetris for Linux.  It's a popular game for
people to write because:

  * The well known screwup over the original ownership means you
    don't get sued for cloning it.
  * It doesn't take *any* fancy artwork.
  * You can write it in a couple of evenings.
  * Tetris is *FUN*.

Myst is the antipathy of a do-able Linux game - and there are no clones
of it (that I know of)...because...

  * It needs *MASSES* of beautiful artwork.
  * It needs cunningly thought-out puzzles.
  * It needs a well-written plot/back-story.
  * It needs 'hauntingly beautiful' music.
  ...although the programming isn't all that challenging...

Programmers are *generally* poor at 'artistic' stuff and there are
really virtually zero "OpenSource Artists".

My experiences:

For my first OpenSource game (Tux the Penguin - A Quest for Herring) I made the
mistake of designing a game that needed clever puzzles and good artwork.  I
naively presumed that if I wrote the code and threw together some convincing
demo levels, there would be freeware artistic types rushing in to provide
some cool levels and interesting puzzles.  It never happened.  I didn't get
a *SINGLE* contribution of that kind.  Tens of thousands of people have
emailed me to say that it has *potential* to be a good game - over 200,000 people
have downloaded it...it's been released on numerous Linux distro's...but still,
not *ONE* 3D model, cool texture or music track has ever been contributed (although
I've had a reasonable number of software patch contributions).

For my second game, I kinda learned my lesson.  TuxKart was designed to have
much simpler artistic needs - and I also built it with very clear, relatively
easy, instructions for building more race-tracks.  Also, it's possible to
make relatively repetitive tracks that are still fun to race.  I did get an
amateur musician who contributed some REALLY good music for the game - he has
a job in the games music business now - so I don't hear from him anymore.

So, Tuxkart has two graphically repetitive levels and two rather cheesy levels that
I'm not proud of.  Still, it's definitely a fun game - I still enjoy playing it.
(Especially since I got it to work with a kludged-together pair of stereo LCD
shutter glasses last weekend!)

Then I did a small commercial game (actually just a playable demo to show off
some new 3D hardware).  That time, I had professional artists paid to do the
modelling...it looks a million times better than my own efforts.

I've started in on my next OpenSource game (The Chronicles of the Evil Overlord)
with the aim to have MUCH less artwork to build up-front.  Much of the scenery is
algorithmically generated - what remains *should* be within my skills range...but
as I get deeper into the implementation, it looks more and more like I'll never
get it to the "playable" state because the artwork needs have grown and I'm out
of my depth again.

Through all of this, the programming part has been quite do-able and great fun.

So - I'm hovering on the edge of giving up on Evil Overlord and trying something
less ambitious.  It's all rather depressing frankly.

>  >  a) Their games look expensive compared to the same game under Windoze
>  >     (which being six months old is now discounted below production
> costs).
>  >
>  >  b) They miss the big advertising campaign that promoted the original
> game.
>  >     The magazines have long stopped printing reviews and cheats and other
>  >     things that keep the game 'alive'.
>  >
>  >  c) All of the other 16 to 20 year olds have already played the Windoze
>  >     version to death and are looking forward to the next big thing.
> 
> The business model isn't fundamentally broken for the reasons that you
> mention. These same 3 problems are problems that MacOS users usually
> encounter. However, you can still make some money doing MacOS ports
> because the market is there. The returns kinda suck, but they're there.
> It is not there yet for Linux for games for these types of games.

So why do you think there are Mac games but (virtually) no Linux games?

IIRC, there are more Linux users than Mac users out there in the world.
 
> (BTW, those Mac porters are also doing it for the love of the system.
> They could be making more money working elsewhere, but they want to see
> games come to the Mac. I don't hear much about Windows developers doing
> something for the love of the system and its users.)

Hmmm - that's an option I hadn't thought of.

Get together a credible team of Loki-like porters who'll work for free
and give the results of the port back to the original company for free
so they can sell the game back into the Linux community (maybe for some
agreed lower price).

----------------------------- Steve Baker -------------------------------
Mail : <sjbaker1 at airmail.net>   WorkMail: <sjbaker at link.com>
URLs : http://www.sjbaker.org
       http://plib.sf.net http://tuxaqfh.sf.net http://tuxkart.sf.net
       http://prettypoly.sf.net http://freeglut.sf.net
       http://toobular.sf.net   http://lodestone.sf.net