King's Quest: Mask of Eternity Development

stub page. You can help the Sierra Wiki by expanding it.

 
 

Developer Notes

Wikipedia wrote:

Mask of Eternity went through two or three main development phases, in which Roberta Williams' ideas changed, the engine was finalized, and the graphics were finalized. In Fall 1996, Williams showed off some of her first screenshots of the game's levels, a few enemies, and the placeholder for Connor. Video showing the level development of this version of the game was included in the Roberta Williams' Anthology. There are scenes for two or three levels that were cut from the final game (including an undersea area, and a green forested area with a village).^{Citation Required} By July 1997, the game had gone into its final phase, and appeared much like it does in the final product. The first gameplay footage of this version appeared in the King's Quest Collection II. This version was not completed until December 1998.

Williams' team was forced to build the engine from an earlier version of 3Space (as Dynamix was behind schedule finishing the updated Red Baron II version of the 3Space engine for their own games).^{Citation Required} 3Space had been designed for running flight simulators, and so was inappropriate for a 3D adventure game in several respects: it could not effectively display texture-mapped terrain up close, it could not animate 3D objects, and it had no means of streaming voice clips. Modifying the engine to serve the purposes of Mask of Eternity took nearly a year. Producer Mark Seibert explained that this was necessary because 3Space was the only game engine that could run both 3D terrain and interiors, thus enabling characters inside buildings to view and interact with characters and objects that are outside the building.

Buildings and other structures were constructed with a proprietary object editor called ZED, while terrain was created with another editor, called TED.

Ken Williams

Ken Williams wrote:

KQ8 is a wild story.

KQ8 was in development at the same time that the company was sold. Basically, Sierra went through changes during the development of the game, and those changes are reflected in the game. During the first half of the game, I was the CEO - during the last half of the game my status shifted to "reasonably nice guy who used to work here". My way of doing things was different than the new way of doing things.

My #1 issue was always to maintain the "clarity of vision" of the game designer. A Sierra project, like KQ8, has nearly a hundred highly creative people on it. Many of these people were working at Sierra because they wanted their shot to be a game designer. It was not uncommon for everyone on a project to seek opportunities to "put their mark" on the game. This is a delicate issue. I recruited people who could be designers, and I was a huge supporter of creativity. Roberta wanted ideas from the team, but at some point, if you accept too many ideas, the product can become a muddy mess. There were dozens of people on KQ8 who could have been the designer, any of which would have made a great designer. But, unfortunately, if this tendency, on the part of developers, to add their creativity to a product, isn't carefully controlled, the product starts to veer into "design by committee". Roberta had her vision for the product, as did almost every person on the project.

When I lost control of Sierra, Roberta's ability to maintain her control over KQ8 was also eroded. The product that shipped is very different than what would have shipped had the company not been sold.

There was another issue at work on KQ8. Roberta is a perfectionist (I'm guilty of the same sin). Whenever she would play the game, she would turn in lists of hundreds of "bugs". Perfectionist can be a pseudonym for nit-picker. When a development team gets a long list, the natural tendency can be to look at some bugs as nit-picky. I always supported my designers. I wouldn't let a game go until the designer was happy (with a couple of exceptions that I regretted later), even when it seemed like we were spending lots of money to fix stuff no one cared about. It was critical to me that the game our customers played represented the game our designer wanted produced. When I left Sierra, Roberta's ability to get bugs fixed diminished.

Ultimately, the last year of KQ8 development was a tough one for Roberta. For a long time, she refused to let the game ship and there was threatened litigation floating around.

This is not to say that the game that shipped isn't a good game. Roberta was reasonably happy with it at the end - but, it reflected a much wider product vision, than Roberta's alone. People other than Roberta influenced its development, in a greater capacity than in her previous products. There will be some gamers who see the change as positive, and some who wanted a Roberta product more consistent with her prior products.

There is an example I used to use on this point. One of my favorite authors is: Steven King. I also like Peter Straub. Each alone is a bestselling (mega-selling in Kings case) author. They cowrote a book; the Talisman, which bombed. Either alone could have sold plenty of copies, but together, the whole becomes less than the parts. KQ8 had wonderful people on it. This message should not be construed as being derogatory to anyone (other than that I am definitely critical of the management changes that took place.) My belief is that if the new owners had taken a couple of days to ask about "what made Sierra special" in the days after acquiring it (they could have asked me, or better yet, its customers) before dramatically changing things, things would have gone a lot smoother in the transition.

-Ken W

Jeff Pobst

Jeff Pobst wrote:

I was on the game. It wasn't anywhere close to done in 1997. Maybe one of the four (?) realms was stubbed in at that time. The challenge of taking Sierra's adventure game teams and bringing them to 3d development, with movement, action, cinematics, dialogue, combat, monsters, climbing (don't forget all that grappling hook fun), puzzles, 3d graphics, 3d collision systems, follow cameras in 3d, buildings, pathing, monsters, etc. was a daunting task. I joined the programming team in July 1997 and the hope was that the game would be done within a year. It became clear by January though that we were very far behind. Several staff changes were made in the Fall/Winter, and I experienced my first (and most significant) crunch in the game industry (along with most of the rest of the team). It went for 6 days a week, 12-14 hours a day from Feb to July in 1998 to build all the systems and content needed and then in July switched to 7 days a week until launch in November 1998. It was grueling, but also there were some great times. It was a special group of people and we really bonded, but it was also a huge challenge to try to bring adventure games into a world that was moving 3d, wanted more action and combat, and wanted deeper exploration without pixel hunting. While the general story, dialogue, and puzzle outlines were likely there in 1997 (before a lot of iteration to get it to work for launch), the actual development of those systems and implementation of the design and all the thousands of details that had to be figured out, filled in, or invented were the things that took more than a year and a half more - and likely would have taken longer if not for the immense amount of personal time the 33+ person team (huge at the time) contributed of their own to the development process.

 

Jeff Pobst wrote:

The design documents of those days covered maybe 10% of what it took to build the game, so having a "completed" design doc isn't much. I actually don't think I ever saw a design doc on KQ8 (I remember the very intensely details one on Gabriel Knight III, but KQ8 didn't have anything like that). What I actually remember most is Roberta's hand drawn map that then producer Mark Seibert used to update, annotate, describe and detail over the years. Near the end of the project it was hung up on his office wall. When I joined in July 1997, there wasn't a 3d camera that followed the main character so that was my first task to build. Then we worked a lot on controls - character movement was already there, but selection changed, how you "clicked" on something in 3d changed, and how we set KQTraps (invisible boxes that would initiate scripts or sequences or monster AI when the player intersected with them) was all developed during that time. We then continued the scripting system that had started developer and over time built all the scripts that would control the puzzle logic and conversation logic, and game states. Then I programmed all the puzzles that required direct programming - the jumping puzzles, the cinematic sequences, etc. I remember working on the use crossbow verb, and shooting things for the first time, I remember working on the use grappling hook verb, and coming up with the specs that we needed for anywhere you could climb so it would work and that the animations would work so you could climb. So much about game development is so far beyond the puzzle and location design and the initial voice script. That's such a tiny part of game development. Over time, the story got updated, lines were re-recorded because of changes, areas were added, others were cut, puzzles were changed (keeping the initial intent, but working differently than initially planned), conversation systems were built and changed. Making games and writing a design document are pretty much two completely different things. Most designers I know work through the entire game iterating and updating their design over time. I think Mark ended up taking on most of that responsibility during the development of KQ8 as Roberta was often out of the country or on travel during a lot of the development time. She definitely laid out the plans for the game though, and I think she returned to write most of the dialog. I remember her critiquing and giving feedback on puzzle mechanics, but so many other parts of the design came from Mark or Mark and her, or from other members of the team contributing to help solve problems as they came up. Its been a while of course - 22 years now since I started on the project, so it's hard to be sure of exact things, but it was intense while we lived it.

 

Jeff Pobst wrote:

I remember thoughts about an underwater level, but yeah it likely got cut because of the tons of technical problems that would introduce for what we were doing.

Gary Stevens

Template:GaryDStevensQ wrote:

We were on the top floor in the Bellevue office and were testing it against 98 and ME. Actually we had a small patch ready before it finished manufacturing to fix a bug we knew would happen rarely in 98 but was a blocker in the ME beta build. I vividly remember to this day, having to go talk to Mark Hood about needing a patch because we didn't lock that bug down until we were already neck deep in manufacturing. Mark Seibert was the EP <executive producer>, and Roberta hadn't been involved in a LONG time.

 

Template:GaryDStevensQ wrote:

Mark Hood was VP of Production at that time, replacing Scott Lynch when he left Sierra to work for Valve post Half-Life 1. HL1 released in November of 1998. This means that KQ8 was after that. As far as RW's involvement, I remember her being involved peripherally in the early product reviews but Mark S was making all the decisions while the game was in QA. The only Sierra alum working with us in 98 was Al Lowe on LSL Casino. K&R were pretty much out of the picture by 98. Mark Hood might have more info on her involvement. I guess i could be wrong.. but looking at the people involved and the fact we were testing it on an ME beta build from 99.

The Making of

Load video

YouTube

YouTube might collect personal data. Privacy Policy

ContinueDismiss

The Making of King's Quest: Mask of Eternity

Related Links

• King's Quest Series
• Two interviews with Roberta Williams During MoE Development