Nintendo apologizes for Pokémon performance problems, promises “improvements”
arstechnica.com
arstechnica.com
But I am still very surprised they allows this to ship, unless they have less control over Gamefreak than I thought? I know its a different company and another company controls Pokemon.
But I compare this to far more ambitious games, Breath of the Wild, Mario Odyssey, Bayonetta 3. That all run nearly perfect. I mean even at launch I can't remember there being any notable bugs in BoTW.
The only explanation I can think of is the sheer number of models in a pokemon game may be much higher than those games, but I am not sure.
For Nintendo, pokemon is more and more a bad look for them when compared to the rest of their games. Don't get me wrong I am enjoying the new game, but from a technical prospective it is just disappointing.
Gamefreak and the Pokemon Company, on the other hand, need to have the games release on schedule to put out new animes, new merchandise, and so on. Nintendo doesn't see that profit, as far as I know, but it's really important to TPC, so I assume the release pressure comes from there.
That is something that keeps Nintendo games as largely day one purchases for me. I don't hesitate to preorder their next game because I (generally) know it will be quality. Even as bad as the Wii U fumbled, the games themselves were some of the best Nintendo games.
Yeah I can see on the Pokemon side that being complicated. But it does leave a mark on Nintendo's reputation. Especially for anyone not familiar with how it actually is setup and just thinks Pokemon == Nintendo (more or less).
I just feel like Nintendo needs to find a way to do something. Somehow, they have a majority stake in the pokemon company and I thought gamefreak.
I played on Switch and I feel like I don't remember that. But honestly I may have just not cared... I burned through that game. I could not put it down.
Regardless, if it was limited to a few areas and not basically as soon as you start the game I feel like it's much more forgivable. Especially for just how damn beautiful that game was.
They seem like the company that will hesitate the least to push back a game. Partially I assume because they know that no matter when it releases it will likely sell well, but also because of their reputation.
I can't think of a rushed Nintendo game in recent memory (outside Pokemon). Nintendo sets a pretty high standard for game quality across the industry.
> The only explanation I can think of is the sheer number of models in a pokemon game may be much higher than those games, but I am not sure.
They remade models in 3D for 3DS games, at the SwSh release people compared them to the 3DS ones and they turned out to be essentially same. So they already put the effort to migrate models to 3D
The explanation is "people will buy anyway". The game would realistically need another 6 months to get to the "good" state and they couldn't miss the holiday season.
Model like CoD with multiple teams working in parallel on subsequent games would probably work better but... Pokemon fans will buy it anyway so why bother ? From company's management view it's a golden goose.
They also tried to make non-pokemon game and it was pretty much a flop so really it is a mediocre developer that have license to print poke-money
I don't really think that would play into the technical issues. The models themselves look fine; nearly all the problems I've seen come down to (in order)
1. Really bad LOD logic/implementation. Just shockingly bad, entire characters popping in 20ft away from you
2. Animation/rigging issues, mainly with the player character, and seemingly correlated with network play
3. World navigation/physics glitches for the main character
I guess #1 could be affected by the number of models if it came down to "we don't have time to create proper LOD increments for every model", though surely that could be automated these days? At least a rough version, which is all you really need
Overall it just feels like an open-world game made by people who have no idea how to make an open-world game (but like others have pointed out, BotW was the first real open-world game Nintendo had made and it was a slam dunk)
The big game engines all have assets or built-in features to do this. Of course, there's still the last 10% of QA and fixing up the ones that aren't properly converted.
Yet, I've enjoyed the latest Pokemon vastly more than the previous one (Sword/Shield, didn't play legends of arceus). Personal opinion, of course, but shaking up the usual Pokemon formula did a lot more for me than performance and polish.
I recommend Legends of Arceus though; much of the shake up seen in Scarlet/Violet was tested first in Arceus, plus a fair bit more divergence still.
Splat and BotW were developed by Nintendo itself (EPD, formerly EAD, formerly R&D4).
Pokémon is developed by GameFreak and managed by The Pokémon Company. For which video games are just one aspect of the franchise, even more so mainline ones (there’s also a bunch of spinoff games sometimes developed by Creature Inc and often not).
That sounds more like the limitations of their 2015 Android phone hardware these games have to run on.
Even the previous pokemon release on the Switch (Arceus?), which is not amazing visually, looks positively fantastic when it's sitting next to the latest, hot-garbage release.
Also, you could transfer ANY pokemon from ANY game to the new game, complete with special moves and items, such as mega-evolutions. Building good teams and all the work going into it, from breeding to training, was fun. There were complex breeding mechanics to get pokemon with unusual moves or better than average stats, and there were side games like Pokemon beauty pageants that you could breed and train them for. Also there were extremely rare "shiny" Pokemon with unusual colors or markings. Those are still there, but it's a lot less meaningful to have one when you can't transfer it to other games anymore.
once (if?) the technical issues are fixed, this chapter is probably gonna be the best pokemon game that we've had for quite a few generations, imho.
I think even Pokemon Arceus had a difficult development as well. The original in house engine was replaced with a modified version of Zelda's Breath of the Wild engine.
I know Mario kart tour was written in unity I think, and it works well across my families wide range of devices.
Edit: It seems some pokemon remakes were done with unity too: https://exputer.com/news/nintendo/pokemon-brilliant-diamond-...
If all those things are shit, the output will be shit. If all those things are OK, the output will be OK, no matter what engine you choose/develop.
This is just not true in any domain.
If your tools are bad, buggy, or inefficient, it's very possible for your talented people to be unable to overcome them.
For example, if a game engine takes even 2x developer time vs. another engine, it's probably not possible to achieve a similar level of polish because the developer time isn't going to double accordingly.
I'm my own experience, good developers are able to work around bad tooling much easier than bad developers. Or even being able to recognize bad tooling requires the developer to not be bad themselves. The matter of "it's working fine, let's move on" VS "I can't take this bad experience, lets fix it before moving on" is often choices bad developers won't even engage in.
edit: At this point, I think it's clear that it's not just their tooling and engine though. I mean, clearly these are big issues, but I'd say it largely stems from a lack of technical expertise and/or investment in R&D and engine/tool development.
Videogames use tons of third party libraries with various engine support either from library developers or third parties. Alot of studios do not have the inhouse talent/money to support all the skills required for inhouse development, so they use a ton of libraries. If you have to reimplement all the third party library support in your custom engine, well you are going to have a bad time.
I think this why unreal/unity consolidation is occurring and honestly not the worst approach for practical game development.
- I'm not in the videogame industry, but I'm interested in becoming a project manager in it after pseudo-retirement
EDITED: They used Unity for Pokémon Brilliant Diamond & Shining Pearl. I'm not sure why they switched back to in-house for Arceus and these newer ones.
The main issue seems to be their release schedule as well as their inexperience with games of this scope, and likely not the engine itself (although it's likely a contributing factor).
Digital Foundry put out a good video showcasing how the game performs and comparison to Arceus https://www.youtube.com/watch?v=pBZqt7D24Zc Honestly, Arceus which was already rather ugly compared to Breath of the Wild and certain other 3rd party titles on the Switch, is a much better looking and detailed game then this new generation.
People have peaked into the nsp file structure, and have confirmed Brilliant Diamond/Pearl did but Sword/Shield was a modified version of the 3d pokemon engine.
https://en.wikipedia.org/wiki/Pok%C3%A9mon_Brilliant_Diamond...