4 ms·

Have to say competitive optimization existed long before AI and it kinda feels like the parallels are drawn just because that’s the current buzzword. Real life

Have to say competitive optimization existed long before AI and it kinda feels like the parallels are drawn just because that’s the current buzzword. Real life sports are slightly more resistant to it because of a couple of factors, but video games have been dealing with competitive optimization to the point of destroying the game since people put money on them. Since the games were not easily patched early on there was a “do what you can to win” mindset/game theory outcome with rare community added rules for tournaments. Plenty of games just died because they aren’t fun to play seriously due to major imbalance or some hard to police technique. In short it’s just the nature of putting a monetary prize on something. You have to design and adjust your system in such a way to handle the idea of people taking every edge they can get

hinkley 4y ago
I mean, motor racing, modulo the clever cheating on engine designs.
llamaLord 4y ago
I think the difference today though is the developers do have the ability to tweak every single tiny little detail of the game in order to tackle these kinds of things, whereas in a physical sport there's really a limit to what you can change to "fix" issues that don't fundamentally change the entire game. Games like DOTA are great examples. There are countless times where the devs have done something like "moved single tree 0.25 in game metres to the left" (noting that there's probably >1000 trees on the map) to address ridiculously tiny optimisation "exploits". You just can't control the environment in physical sports the same way, so finding real "balance" that optimises for skill rather than just stats optimisation is going to be nearly impossible.
- Eji1700 4y ago
  I've been a huge fan of dota since 04/05 and yeah it's a great example, but on the real life side another reason things don't change as fast is just the huge barrier to entry, and to some extent the entertainment value. The cost, time, and physical demands to be in the top 10-20% of many real life sports is just a huge gate, and so you get a lot of people who become complacent. It often is the lower rungs that are trying to innovate, but it's hard to see that flow all the way to the top because you don't know if your strategy is actually viable vs the best of the best. Vs something like a video game where playing near the top level is much more achievable. There's also still an aspect of "entertainment". I'm not certain on this but to memory there's decent evidence that the best way to shoot a freethrow is an underhanded granny shot, but VERY few NBA players ever even attempt it.

hinkley 4y ago

I mean, motor racing, modulo the clever cheating on engine designs.

llamaLord 4y ago

I think the difference today though is the developers do have the ability to tweak every single tiny little detail of the game in order to tackle these kinds of things, whereas in a physical sport there's really a limit to what you can change to "fix" issues that don't fundamentally change the entire game. Games like DOTA are great examples. There are countless times where the devs have done something like "moved single tree 0.25 in game metres to the left" (noting that there's probably >1000 trees on the map) to address ridiculously tiny optimisation "exploits". You just can't control the environment in physical sports the same way, so finding real "balance" that optimises for skill rather than just stats optimisation is going to be nearly impossible.

Eji1700 4y ago
I've been a huge fan of dota since 04/05 and yeah it's a great example, but on the real life side another reason things don't change as fast is just the huge barrier to entry, and to some extent the entertainment value. The cost, time, and physical demands to be in the top 10-20% of many real life sports is just a huge gate, and so you get a lot of people who become complacent. It often is the lower rungs that are trying to innovate, but it's hard to see that flow all the way to the top because you don't know if your strategy is actually viable vs the best of the best. Vs something like a video game where playing near the top level is much more achievable. There's also still an aspect of "entertainment". I'm not certain on this but to memory there's decent evidence that the best way to shoot a freethrow is an underhanded granny shot, but VERY few NBA players ever even attempt it.