Graphics engine choice (RE: [MUD-Dev] size)

Derek Licciardi kressilac at
Sun Oct 5 00:41:38 New Zealand Daylight Time 2003

From: Sheela Caur'Lir
> Bruce Mitchener wrote:
>> Bo Zimmerman wrote:
>>> From: Amanda Walker

> Mostly because technology have moved so fast recently, that you
> may not attract players a year from now, with an engine of
> yesterday.


> And I think it will happen more and more, that games, even the big
> ones, will be based on a licenced engine, so the developers can
> concentrate no game content to a larger degree.

I'd have to agree here.  Assuming we're after a commercial MMO.
Assuming we're evaluating the roll your own engine or buy one off
the shelf...  I tend to lean on the buy one off the shelf for the
following reasons.

  1) Stability. Period.  How many MMOs have we played that simply
  cannot get it right in the engine department.  Bad launches cause
  fan disappointment, and lost revenues.  Buggy clients are a HUGE
  turnoff to the initial impression and initial impression can often
  mean the difference between a subscriber and only a sold box.
  Anything that minimizes the risk of a bad launch has to be
  seriously considered.

  2) A new engine is not needed.  UO is evidence of this and I think
  it applies to newer games.  RPGers time and time again show with
  their dollars that they want story, depth, and game play before
  graphics.  You don't need to be able to hang with the FPSes.

  3) The engine has to last 5+ years.  Now you can staff up like
  Criterion or NDL to keep your engine glowing through each DirectX
  release and each new shader language or you can let someone else
  do it.  I'd prefer the latter because MMOs are a hard enough sell
  anyway and the Graphics Engine resale market is saturated at this
  point.  Why build your own technology if you're never going to
  resell it in some way.  An asset that is never used might as well
  be leased.  Remember MMOs are a 24x7 service.  Developing some
  home grown wizbang graphics engine that crashes every time the
  toilet flushes simply is not the definition of providing
  service. Admit you don't have the QA department for a graphical
  engine.  Admit you don't have the talent and admit that your
  engine will crash on every platform you do not test it on.  At
  that point the buy decision becomes much more appealing.

Lastly, You're already behind on the content race.  Why waste a huge
segment of development time waiting for the engine to be stable and
developed to the point where your entire art pipeline is complete?
The 12 months of idle art time could have been put to far better
use.  It doesn't make sense.  Buy the engine and implement 100 new
quests and I am sure your players will thank you for it with their

That being said, there is only one case in my mind where it makes
sense to build your own.  That's if you are an SOE or an NCsoft and
you plan to spread the cost of development across many projects.  In
this case you can afford to staff up an internal NDL sized
development team just for your engine, though even that is risky
because there is no one you can turn to when something goes wrong.
In the end you're still left with an asset that is not worth
anything to the outside world because of the conditions of the
Graphics Engine marketplace.  Again, an MMO is a service.

MUD-Dev mailing list
MUD-Dev at

More information about the MUD-Dev mailing list