McDuff wrote:That's got the potential to be a really good feature if the AIs can handle things like marketing. Or a really annoying and crappy one if they can't.
Personally I'd love to see it done well. Wait and see on that one...
The idea ties in with my thoughts on an
infinite-variety crafting system; with such a system in place, different corporations would (hopefully) naturally tend towards specialising their products towards being good in a particular area where their competitors' products aren't so that they have a unique selling proposition. Think of the different weapon manufacturers in Borderlands, for instance; all of them have a distinct feel and set of strengths and weaknesses to them.
If that is the case, I would hope that NPCs would be intelligent enough to not only specialise their products in a particular way but to
recognise that they're specialising their products in that particular way and to factor that into their marketing.
How could you design the system so that NPCs would naturally design products that address market needs? I suggest that corporations developing products for a particular market will hire or subcontract the task of R&D to research-proficient NPCs. These NPCs would design different blueprints, and the corporation's engineering department would produce products based on these blueprints. The NPC would receive royalties on every product sold that used one of its blueprints, and it would be kept aware of which of its blueprints were generating the most money for it. Then, by using genetic algorithms where each blueprint counts as a "solution", the set of all blueprints it has developed counts as the "population", and the money generated by each blueprint is factored into the fitness function, the NPC will develop more blueprints. This means that it will base future blueprints on the most successful blueprints it has developed so far.
In a market where consumers favour low-AD, high-ROF plasma cannons over high-AD, low-ROF variants, NPCs developing blueprints for plasma cannons that have a high rate of fire at the expense of alpha damage will find that those blueprints earn it a lot of money and gear future research in that direction. Since the products based on those blueprints would be in high demand, the corporation hiring or cooperating with them would likely continue to do so in the spirit of mutual self-interest, which over time would lend that corporation a distinctive feel for its products: "We don't hit 'em hard, but we hit 'em fast".
Other competing manufacturers would eventually get the message and start gearing their products that way as well, but if we model "brand loyalty" (and I really think we should, since it would be a natural extension of a "reputation" mechanic that I and many others would like to see in Limit Theory), then that would act as a market barrier that would limit the amount of homogeneous products. On a smaller scale, we'd see other firms that might sell high-AD, low-ROF plasma cannons to meet the smaller, niche demand for those kind of weapons in the system.
This is all a small part of the thoughts I've had on business management, but didn't get around to posting yet.