A short time ago, at a computer terminal far, far, away, Doug Gould wrote:
>Tim Morten <tmorten@activision.com> writes: >> Doug Gould wrote: >> > I remember hearing the same crap about the stupid joystick/turret >> > control way back when MW2 was first released. Has there been a fix? - >> > NO. I'll be very surprised to see that you deal with the mech lab any >> > differently. >> >> Actually, our user base was so divided on the joystick control issue >> that it didn't make sense for us to change that, despite the verbosity >> of those that didn't like it.
(Since my news server appears to have already expired your article, Tim, I'm going to have to reply to this reply; I'm assuming it's been quoted accurately.)
I'm sure you remember I was one of those more verbose critics of the joystick forced-centering when Mech2 first came out. You, as Doug quoted from my FAQ, said that you would be "actively looking" at making the centering optional.
I must say I'm _very_ disappointed you chose not to address this problem. Even if your user base was divided on the issue, providing non-forced-centering as an _option_ would clearly please the critics, while just as clearly continuing to please those who claim to have no problem. It's really a win-win situation, as I see it.
If it was at all close to a 50/50 division, that's a hell of a lot of dissatisfied users. Even if it was 20% who didn't like it, considering the popularity and success of Mech2, that's a very large _number_ of displeased users, regardless of the percentages.
It's _always_ preferable to provide an option for users. This is one of the great advantages of selling software rather than physical objects - it's almost infinitely flexible. With a single keystroke, it can change its appearance or behaviour entirely. You just can't do that with physical objects. If you have different markets, or different consumer groups, that prefer different colors of items, or a different layout of buttons on a control face, you have to decide _which_ you want to offer, or you have to commit to manufacturing multiple designs, and you have to know that there is a sufficient market to buy them.
With software, it's entirely different. You _can_ offer several different colors, a variety of layouts, and any number of designs, and you can let the _user_ decide which one of the many possible combinations he or she prefers. Before, you had to find and target identifiable, large markets to design towards. Now, suddenly, you _can_ make a single, but flexible, product that will target _all_ those major markets, not to mention a non-trivial number of much smaller, and perhaps previously unidentifiable markets - all with the addition of a small investment in code additions.
I would also argue that the increase in market is not necessarily linear to the number of options - one more option doesn't add just one group of customers. Instead, I submit that it's an _exponential_ relationship, where the total number of combinations is what's important, not the total number of options. You add one option, and it provides not only just one new option, but an additional _number of combinations_ based on the number of already existing options.
And of course, there's an asymptote, where you approach a maximum number of options, at which you reach diminishing returns, and consumers either don't notice the additional options, or it becomes just too many to deal with. If there's a noticeable number of customers complaining about the lack of an option, though, I think it's clear that you have a obvious opportunity for some significant returns by adding the one single option, and that you're still a good distance from that asymptote.
The concept of the "configuration file" is, I think, a rather modern concept, and a particular phenomenon of the software market. As consumers become more aware of the vast possibilities and great flexibility that software offers, I think we're going to see this becoming more and more important a feature, and a crucial deciding point for consumers. They're going to seek out the product with the most options - and even if they like one product better, a competing product may offer enough options that they can configure it like the former. In fact, I think we've already seen this. Just look at MS Word - it had a "WordPerfect" mode that mirrored WP's keyboard and menu options, and was thus able to cash in on a large pre-existing, yet competing, market.
For gamers, too, this is an issue, and it becomes a criteria for differentiating and choosing products. Controller config is a prime example. Not long ago, games had pre-defined keys that were entirely designer-dependent, and completely unchangeable by the user. They almost always varied from game to game, according to each designer's preference, which more often than not didn't match the user's in one way or another. Users had to hunt-and-peck for desired commands on the keyboard, and the best solution of designers was to include the "keyboard overlay" which tried to label keys so that users could find them more easily. It was an unsatisfactory situation at best, and more often frustratingly difficult.
Now, of course, most games provide some kind of control configuration, and users have come to expect this. Some systems are better than others, though, and I think in this respect Activision's controller-config pales in comparison to others on the market. Descent's, in particular, was the one that first impressed me. Select a function, and then _press the key_ or button you want to assign to it. There's none of this searching through lists for the desired key that Mech2's config requires. Mech2's is very unintuitive, not to mention the fact that only _some_ functions are configurable and only _some_ keys are selectable. It's a good first try, and I'm disappointed that Activision didn't seek to improve this in the followup MW2 offerings. They've really fallen behind others in the industry.
Recently, in fact, I've been extremely impressed with some other config systems on games I've played. Interplay/Bioware's Shattered Steel is foremost in this respect. Not only can you select all functions for config, and indicate the assignment by merely pressing the desired key or button, as well as reversing the direction of axes, but you can _also_ choose between "absolute" and "incremental" modes on joystick axes. I almost, literally, jumped for joy when I saw that! I only wish Mech2 and Mercs had made me feel that way.
Another particularly impressive config system was in Kesmai's Online Multi-Player BattleTech: Solaris game. This had the same key-press input control, and what impressed me was that, in a matter of seconds, I could use this to re-configure the game's keys to match those already programmed into my joystick, and be ready to kick ass in combat only moments later. And, of course, it didn't have the forced-centering on the torso control, so I could go into battle and not only kick ass, but take names as well. :)
Overall, Activision really missed the boat in terms of their control configuration, and made a very poor decision not to respond to consumers' desire for one additional option for joystick control. It's another example, I think, of the unnecessary focus on graphics at the loss of needed improvements in gameplay.
===
Will Day