Leo L.Schwab Talks About Monster Manor

[Lost Dragon]

Generalized 3D rendering is a tough problem, and even tougher to do

quickly. Solving for the general case is usually far too slow, so you make observations about the world you intend to present to the user, and try to find ways to eliminate costly steps, like sorting.

 

Total Eclipse eliminates sorting by fixing the viewing angle

straight ahead; rotation about the viewing axis doesn't alter the sort

order. In this way, they can just set up a nearly static array with a

sliding offset and changing coordinates. Then they perform a rudimentary clip for polygons not within the view, project the thing, link all the cels

together and squirt 'em at the hardware.

 

They may not even eliminate

backfaces, choosing to let the hardware do that, or rely on painter's algorithm.

 

Monster Manor also eliminates the sorting step by observing that the

entire game is bricks on a grid, and scanning that grid in such a way thatneverything gets extracted in pre-sorted order (poof!).

 

However,it usually extracts too much (especially on level 10), and the frame rate goes to hell.

 

 

I hit upon a new algorithm when I was about half finished which would have boosted the frame rate by about 50%, but we didn't have time to rip the guts out of a working program and replace it with something new.

[Lost Dragon]

Monster Manor was completed from start to finish in six months.

This is an extremely short time frame when you consider the amount of art,

music, sound effects, coding, debugging, debugging, memory budgeting,

debugging, change requests from the publisher... (Did I mention debugging?)

 

We were also trying to hit a moving target. When Monster Manor was

started, the 3DO system wasn't finished yet. The OS was undergoing constant

revision, the hardware slightly less frequent revision, all of which had

bugs against which we had to defend (and then pull out when the bugs were

fixed).

 

We also had a few design goals. Among these were: It had to be

done by Christmas :-), and it had to show off some of the special features

specific to the 3DO. For the latter, it meant using the cel engine

hardware. I didn't object to this because it's blazingly fast.

 

My homework for Monster Manor was to sit and stare at Wolf-3D for

hours, looking at their art, field-of-view, map size and layout, colors,

animations, number of different objects, etc. etc. etc. However, I didn't

spend time with the Wolf-3D code in a debugger. I don't speak Intel and

have no desire to learn (I'm an Amiga bigot). So I was unable to figure out

what cleverness they used to do the image mapping so well. (And I was

trying to stay away from the Net so I'd have time to do actual work.)

 

 

Based on what I learned from staring at Wolf-3D, I formed the basic

underpinnings of Monster Manor, and chose to go with a polygonal model.

There were two primary reasons I didn't go for raycasting (assuming I had

known about it at the time)

 

 

I didn't know how to write a sufficiently quick raycaster, whereas

I had a fair idea how to write a decent polygonal system (remember, we're talking time constraints here).

 

 

A raycasting system would not (easily) have shown off the effects

possible with the cel engine hardware.

 

Yes, I saw the polygon deformations while developing Monster Manor,

and wished there was something I could do about it. When I finally did

learn about raycasting, incorporating it into Monster Manor would have

required a complete re-write from scratch. This is not possible in a

commercial environment when you're staring down deadlines, and your

publisher (the one who's paid you $$$$ to finish the thing on time) expects

to see forward progress every $(INTERVAL).

 

Readers are posting accounts of motion sickness, so I must have done something right. But if the polygon deformation detracts from the game that much for you, then please accept my apologies. I'm older and wiser now, and if I had to do the whole thing over again, I'd do it very differently (and budget more time).

 

(Just think: I could have prevented this entire discussion if I'd

simply tilted the camera when you got whacked by a monster... :-) )

 

_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_

Leo L. Schwab

[+chicgamer]

I recently discovered his YouTube channel, and it's been interesting listening to him talk about his work and the environment at 3DO.

 

I'm one of those people that get motion sickness from Monster Manor, which is a shame, as I would like to play through it (I love how they designed the monsters in the game).

[Lost Dragon]

I recently discovered his YouTube channel, and it's been interesting listening to him talk about his work and the environment at 3DO.

 

I'm one of those people that get motion sickness from Monster Manor, which is a shame, as I would like to play through it (I love how they designed the monsters in the game).

 

He and the 3DO itself, are very new grounds to myself.

 

I find him fascinating to read up on.

 

Prior to looking into the 3DO, discussions had always been set around how much further the Jaguar,32X and Saturn could be pushed, how dissapointing games on them could of been written better..

 

It's most refreshing to dive into the 3DO waters at last.

[+chicgamer]

Yeah, it would be really interesting to see what could have been done with some 3DO games if they weren't under the constraints they had.

 

Schwab talks a lot about how much money and time was given to Jurassic Park Interactive (which turned out to be horrible). If he had been given the same money and time for Monster Manor, who knows how much better it would have been? I think, as it stands, it's one of the highlights of the 3DO.

[Lost Dragon]

I'm starting to view the 3DO in a whole new light.

 

It had always impressed me with:Starfighter 3000, Space Hulk, NFS, Return Fire, SSF2, Samuari Showdown, The Horde, Madden..even games in genres i was not personally interested in, it's titles seemed a lot more polished than what Atari were offering me as a Jaguar owner at that time.

 

 

Finding comments from those involved in 3DO software development has really opened my eyes to just what they were up against. .i am so used to developers complaining of working for the Tramiels and restrictions imposed.

 

Another new source to myself.

 

 

 

Bill Baker:

 

EA had nothing to do with Monster Manor other than marketing it. It

was designed and programmed by the titling nucleus of ex-NTG folks who

got rolled into the 3DO Software group, specifically Leo Schwab, Kim &

Stephen, some other folks. I test-played some of the later beta's.

Pretty good for the relatively meager resources given to the team. All

those guys are long gone.

 

 

Some of the early ports were *really* ugly. Conversely, I always

wondered how some early titles done by third-party houses managed to

look better than ours, even though we had the original designers of the

system at hand.

[trapexit]

Speaking of his YT channel: https://www.youtube.com/user/BolsEwhac/videos

 

It takes a while to go through his playthrough videos but there's a lot of good info.

[Lost Dragon]

Yeah, it would be really interesting to see what could have been done with some 3DO games if they weren't under the constraints they had.

 

Schwab talks a lot about how much money and time was given to Jurassic Park Interactive (which turned out to be horrible). If he had been given the same money and time for Monster Manor, who knows how much better it would have been? I think, as it stands, it's one of the highlights of the 3DO.

 

I did see Bill Baker talking about J.P.I:

 

JPI was not a cheapo job, so 3DO wasn't

trying to rip anyone off. It's just that a lot of that money was wasted,

along with much of the schedule, leaving a great deal of the content to be

crammed together at the last minute.

 

For instance, it was decided to hire a "hotshot" film school grad from

UCLA, at outrageous contract rates, to write the JPI "script". Nevermind

that 3DO already had two trained screenwriters on staff with more

multimedia experience. Said UCLA grad spent weeks just transcribing,

literally retyping, Criton's book into screenplay format. And charging

3DO for every minute. But hey, she's from Hollywood and wears black a

lot, so she must be a whiz scribbler, right?

 

(Same story for Twisted. Went out-of-house and hired a "comedy writer"

who sold himself as gaming savvy. End product was equal parts cornball

and racist stereotypes, very little of which made it into the final

version. I met the yutz at SCES, and he couldn't get a laugh out of a

drunken hyena.)

 

Y'know, the early betas of JPI actually held a lot of promise....

[+chicgamer]

In his YouTube videos, Schwab talks about how they would see new iterations of the Jurassic Park Interactive game, but very little had been improved in each build.

 

It really does sound like they totally mismanaged the license, which is why the final game seems to be a cobbled together mess that in no way showcases the power of the 3DO. It's really a shame, as it could have done so much more to give people a reason to invest in a 3DO.

[Rick Dangerous]

LOVE Escape from Monster Manor. One of the exclusives that will always make me hang onto my 3DO. Thanks for all the cool background info.