A great example was someone did this with Skyrim a while back. In the dialogue they convinced the NPC to join their party. But there isn’t any code logic to allow that, so the NPC is talking like they joined the person’s party, but the gameplay itself doesn’t support it.
That’s the exact type of scenario I was thinking as well. I had seen another video for Skyrim with AI dialog where they used it to haggle with a merchant who agreed to drop the price of an item in the shop. But an item’s gold value is baked into the game itself. An NPC can say they’ll lower the price, but it will still cost the exact same (barring the normal modifiers based on skills/quest completion/disposition/etc.)
It’s very big on the open-world model, and the party being a lot more dynamic opens up a lot of choice for how you want to build your standard batch of characters.
There are still times during the story where members of the party will split up, essentially to spotlight each character at least once, but most of the game gives you a lot more party composition choice than Remake.
They simplified some of the character progression, but expanded the Synergy feature from the Remake DLC that allows characters to do special attacks with other characters, which is cool and helps mix things up. Certain combos of characters can be good just for their synergy abilities. And the new party members in this game are just fun.
Everyone also has more capability to deal with flying enemies or enemies at range, just built into their standard moveset. I found flying enemies to be really annoying in Remake, but fine in Rebirth. And everyone has the option to obtain a set of elemental damaging abilities that help with staggering foes to avoid having to use as much MP on spells.
On the topic of pressuring/staggering, they also improve that a lot, where the conditions to pressure an enemy are more varied and easier to pull off, which you can learn by using Assess on an enemy just once.
Everything feels familiar to Remake, so I’m sure if someone simply doesn’t like anything at all about Remake, they may still not like Rebirth. But for anyone who likes remake except for a few peeves with combat or how limiting the game feels in terms of exploration/story railroading, Remake vastly improves all of that.
If there’s only one potential gripe specific to Remake that I may not like as much, it’s just that a lot of the open world mechanics feel a bit Ubisoft-y, but it didn’t really feel as tedious to me to do them all. It’s worth doing enough of them to upgrade the BGM for each zone, at least!
Love this game. I will say that not all of its systems are perfect, but I do believe that it is worth pushing through to get to Rebirth, which is simply bigger and better in every way (including photo mode).
What threw me about the remake too is that a lot of the FLUDD mechanics are more annoying when you can’t partially press the trigger. It felt like it made more sense when you could “regulate” the flow with how strongly you pushed, but triggers on Switch controllers are only off/on.
The Switch 2 is actually decently beefy for what it is—give or take certain specs, it’s about comparable to the PS4, which Elden Ring launched on and ran fine on. But Elden Ring is simply a poorly optimized game overall. It ran like shit on PC after it launched, though they eventually got it into a mostly good state years later (or maybe people just upgraded hardware to the point they could brute force it to be stable).
But I guess trying to port it from x86 to Tegra for Switch 2 is another thing entirely that they apparently weren’t prepared for. If all they did was shove it behind an emulation layer or something (yikes if so), I can see why it’d suck. But given just how held together by duct tape the game is in general, I wouldn’t be surprised if they simply lack the resources or expertise to really optimize for a different architecture, since they barely support one to begin with.
Seconded. Switch 2 at this point is mainly just worth it if you have a backlog of Switch 1 games that you want to play in better quality.
Donkey Kong is the first true “must buy” (MKW is good too but it’s mainly just for people who have played MK8 to death and want something new). It’s gonna be a bit for another tentpole franchise to carry the console further towards being a compelling purchase.
I’m not sure Air Riders will be that game either. I love the original Air Ride to death and I’m really looking forward to Air Riders, but I don’t think it carries a console. Metroid Prime 4 is probably the next big decider for a lot of people.
I wouldn’t say it’s “slow” per se, but it does feel different, and in some ways I believe it’s not as good as its predecessors.
One consideration is that it does not have the 200cc mode that MK8 added after the fact. It’s currently (maybe permanently?) at the default max speed of 150cc.
The biggest difference for me though is the courses. Previous MK games use circuit courses, where you start at the finish line and you race in 3 or more laps in a circle that returns to the same finish line. MK8 fleshed out a bit more by incorporating lengthy straightaway courses, where you start at point A and race to point B, with laps being more like checkpoints along the way. But the majority of tracks in MK8 were still circuits.
Mario Kart World, on the other hand, is primarily straightaway style tracks with only a small smattering of circuits, because it’s attempting to integrate everything with the open world map they made, and those tracks also feel like they have less character. The majority of races feel harder to pace because most of them do not repeat themselves, and there’s less opportunity to learn a lap and do better on the next one within the same race.
There’s also the fact that they doubled the number of characters in each race compared to MK8. MK8 had 12 racers per course, MKW has 24. All of those racers are still picking up items, still tossing red shells and blue shells everywhere, still spamming lightning, etc., so it feels a lot more chaotic.
Accommodating that aspect is the fact that it now takes 20 coins to hit max speed instead of 10, because they assume you’re going to get hit by more things that you can’t avoid, so it can take longer to ramp up your speed from the beginning of the race.
Final notable difference that may contribute to a feeling of “slowness” off the top of my head is that you no longer choose parts of a kart like you did in MK8. You simply choose a racer and choose a cart, and your stats are based only on a combination of those two factors. It is more difficult to optimize for things like acceleration, max speed, and turning because you can no longer mix and match parts that exactly fit your stat preferences.
All of this is just my opinion from having played it, but I think that MK8 is still the better Mario Kart game. Just considering how content-dense it is after years of DLC, and the fact that it still runs well on Switch 2, tells me that it’s still worth keeping around and still a good go-to for Mario Kart nights with friends. MKW is still a fun game, and I’d recommend it for Mario Kart fans looking to change things up a bit, but it tried a lot of new things and not all of them work as well as I think they could have.
I was thinking this recently when watching footage of Dread Delusion, a 2024 game that looks like something out of 1999.
It’s a visually interesting game, maybe not profoundly so, but it gave me a passing thought about what makes a game more “artistic”. I was looking at a rocky wall texture, low res enough to count the individual pixels, but I still recognized it as rock. And then I asked myself what takes more skill: a high fidelity AAA game that just megascans a real rock surface to capture as much detail as possible, or a game like Dread Delusion trying to convey the idea of a rock in as little detail as possible.
Developers back in the day would have absolutely killed to have the hardware capabilities we have today. No longer needing to worry about fitting games on a tiny disc or cartridge measured only in MB, not even in GB. Even Dread Delusion, despite looking like a PS1 game, could not have fit on even 3 PS1 discs. But it was those very limitations that made developers really have to think carefully about their content, the total scope of the games they wanted to make, how much detail they could afford to include, etc.
I don’t think those limitations necessarily made games inherently better, because there were still a lot of bad games back in the day. But it meant that everything had more deliberation to it, where a developer would create a game that was one really good idea instead of a game made of 20 just “okay” ideas.
I think it works best sitting down. The scenes are generally a fixed perspective, but you do at least want to give your head a bit of room to look around because sometimes there’s small details hiding behind parts of the environment you can peek around, and honestly it’s just a beautiful game to take in.
I am not sure what the user above is thinking, but to play devil’s advocate:
One thing that modern AI does well is pattern recognition. An AI trained on player behavior, from beginner level all the way up to professional play, would be able to acquire a thorough understanding of what human performance looks like (which is something that games have been developing for a long time now, to try to have bots more accurately simulate player behavior).
I remember someone setting up their own litmus test using cheats in Tarkov where their main goal was just to observe the patterns of other players who are cheating. There are a lot of tells, a big one being reacting to other players who are obscured by walls. Another one could be the way in which aimbots immediately snap and lock on to headshots.
It could be possible to implement a system designed to flag players whose behavior is seen as too unlike normal humans, maybe cross-referencing with other metadata (account age/region/sudden performance anomalies/etc) to make a more educated determination about whether or not someone is likely cheating, without having to go into kernel-level spying or other privacy-invasive methods.
But then…this method runs the risk of eventually being outmatched by the model facilitating it: an AI trained on professional human behavior that can accurately simulate human input and behave like a high performing player, without requiring the same tools a human needs to cheat.
I don’t think that’s a bad thing, though. It’s not like I’d be bothered if someone with a GPU 5 years older than mine is able to play the same game on PC.
This is what consoles just should be. No longer locking games to specific generations, letting newer hardware run older titles better, and letting developers continue developing for lower hardware targets to include more people.