Pokémon Thread XXIX: Sword, Shield, and Spoilers

[OracleofWuffing]

Then you have no reason to be pissed off, because I never said that.

All is I asked is that if you claim that "x is totally easy to do", then provide some actual examples of x actually being done. For you to prove your point.

Yet nobody can actually name a single other game that does all that they're asking of pokemon upgrade hundreds of characters, leave absolutely nobody behind, do it in a timely manner with just a staff the size of Game Freak's, funny that.

And yet when I provided picture and video evidence of it having already been done, it is dismissed because I am clearly oversimplifying the game design process and am missing something arcane about polygon textures that cannot be seen or named. If your argument is genuine, we must both be right because nobody can give me a name, either.

[deuterio12]

And yet when I provided picture and video evidence of it having already been done, it is dismissed because I am clearly oversimplifying the game design process and am missing something arcane about polygon textures that cannot be seen or named.

One picture and a short video do not a complete game make.

The arcane things you're missing are called:
-"limited resources" (including time). Just because somebody has time to upgrade a few animations doesn't mean Game Freaks has the time to upgrade all the animations before the release date.
-"monsters are not just graphics". They also have stats and text and whatnot. And again, that runs into the arcane concept of "limited resources" when you have to repeat it countless times.

So with your confusion clarified for good, pokemon masters trailer up. Collect both the mons and the trainers!

And bringing back mega evolutions too.

[Bartmanhomer]

One-Hit KO moves like Horn Drill, Guillotine and Fissure really stinks. The accuracy is pure garbage. Like why do they even exist in the first place?

[Qwertystop]

The arcane things you're missing are called:
-"limited resources" (including time). Just because somebody has time to upgrade a few animations doesn't mean Game Freaks has the time to upgrade all the animations before the release date.
-"monsters are not just graphics". They also have stats and text and whatnot. And again, that runs into the arcane concept of "limited resources" when you have to repeat it countless times.

They wouldn't have to upgrade all the animations. They didn't upgrade many (or any?) animations between X/Y and OR/AS, between OR/AS and S/M, or between S/M and US/UM.

Meanwhile, almost nothing has ever had stat changes since gen 3 (when they split Special and rewrote the underlying mechanics, so everything got that). Most dex entries have just been copied up from an earlier game for the last few generations, as well.

Yes, it would be a lot of work to do all of that – but Game Freak has repeatedly demonstrated that they're entirely willing to just not do any of those things.

And it's not as though they've claimed the reason to be that it would be too much work, at least that I've heard. Their stated reason is "balance". I don't think it makes much sense as a reason (Pokemon has never had balance and this almost certainly won't change that), but frankly if they were going to lie about their reasons they could have come up with a better lie than that.

[deuterio12]

One-Hit KO moves like Horn Drill, Guillotine and Fissure really stinks. The accuracy is pure garbage. Like why do they even exist in the first place?

There's other moves that raise acuraccy to 100%.

And well it's kinda tradition for RPGs to have a few "one hit kill but good luck actually connecting" abilities.

They wouldn't have to upgrade all the animations. They didn't upgrade many (or any?) animations between X/Y and OR/AS, between OR/AS and S/M, or between S/M and US/UM. That's the thing about 3D – it doesn't take nearly as much upgrading.

Meanwhile, almost nothing has ever had stat changes since gen 3 (when they split Special and rewrote the underlying mechanics). Most dex entries have just been copied up from an earlier game for the last few generations, as well.

You realize that Bulbapedia has the pokedex entries for each pokemon for each game and they're different, right?

And lots of stats get changed. As an example Pikachu learns Double Team at levels 13, 15, 18 or 21 depending of generation, so they weren't just copy-pasting the stats blindly. Actual human beings got around and decided "we should move Double Team to this level this time for this pokemon". Good luck actually finding a pokemon that didn't get their learnset changed in any way between games. And that's just old moves, every time a new move gets added they need to decide where it goes on old pokemon too. And then the moves themselves get changed too often enough. Tackle's power oscillated between 35 and 50 over the generations.

Yes, it would be a lot of work to do all of that – but Game Freak has repeatedly demonstrated that they're entirely willing to just not do that.

That was when they were doing 3DS games and thus most probably working with the same base engine.

But Sword and Shield are for the Switch, meaning a completely new base engine so everything needs to be converted to said new engine.

[Bartmanhomer]

There's other moves that raise acuraccy to 100%.

And well it's kinda tradition for RPGs to have a few "one hit kill but good luck actually connecting" abilities.


You realize that Bulbapedia has the pokedex entries for each pokemon for each game and they're different, right?

And lots of stats get changed. As an example Pikachu learns Double Team at levels 13, 15, 18 or 21 depending of generation, so they weren't just copy-pasting the stats blindly. Actual human beings got around and decided "we should move Double Team to this level this time for this pokemon". Good luck actually finding a pokemon that didn't get their learnset changed in any way between games. And that's just old moves, every time a new move gets added they need to decide where it goes on old pokemon too. And then the moves themselves get changed too often enough. Tackle's power oscillated between 35 and 50 over the generations.



That was when they were doing 3DS games and thus most probably working with the same base engine.

But Sword and Shield are for the Switch, meaning a completely new base engine so everything needs to be converted to said new engine.

Ok. What are the moves that raises accuracy?

[OracleofWuffing]

The arcane things you're missing are called:
-"limited resources" (including time). Just because somebody has time to upgrade a few animations doesn't mean Game Freaks has the time to upgrade all the animations before the release date.

I will reiterate: All of the animations work. All of the models work. They had this available during X and Y, where it worked fully there. They have less resources and less money than Game Freaks, and they did this on their free time, outside of their regular jobs. Your assertion is incorrect.

-"monsters are not just graphics". They also have stats and text and whatnot. And again, that runs into the arcane concept of "limited resources" when you have to repeat it countless times.

First, let's review some points made previously throughout the thread:

• There has been no indication that the stat system is changing in Sword and Shield, so they can reuse the same stats.
• Even if the stat system changes in Sword and Shield, there is no reason they can't use the same algorithm they used to change the pokemon that were included on the pokemon that weren't.
• You can re-use the existing pokedex entries, which has already been done in previous generations.
• When you re-use existing pokedex entries, you can re-use the existing translations, so there is no need to re-translate existing text.
• The previous two points don't even matter because the standard in Sun, Moon, Ultra Sun, and Ultra Moon is to just not have the pokedex entries for non-native pokemon, that's for Pokemon Bank and/or Pokemon Home to provide.
• Reading, parsing, and mass-editting of delimiter-separated value files is a well-established, frequently used, core tradition in many computer-based fields.

For clarification on how I'm going to harp on this point, ASCII had control codes for this purpose in 1963, grep was released in 1974. Don't make the human type the thing over again, have the computer box automate it. Higher-end models of computers are even capable of running more than one task at a time, it is quite possible that they can do all of this while the human uses the computer to write the tutorial on Dynamaxing.

The stats have been taken care of. The text have been taken care of. The models have been done. The animations have been done. So that just means your "name" for the thing that I'm missing when I provide an example is... "Whatnot." Well, there you go, that's also probably the name of a series that releases two games every year featuring thousands of monsters and they always keep bringing back their whole cast in every release while also adding new monsters.

Edit:

Good luck actually finding a pokemon that didn't get their learnset changed in any way between games.

Unown. Smeargle. Regardless, DSV point.

That was when they were doing 3DS games and thus most probably working with the same base engine.

But Sword and Shield are for the Switch, meaning a completely new base engine so everything needs to be converted to said new engine.

A new engine that works the same way as the old engine on a new system that runs the same language as the old system.

[Bartmanhomer]

I will reiterate: All of the animations work. All of the models work. They had this available during X and Y, where it worked fully there. They have less resources and less money than Game Freaks, and they did this on their free time, outside of their regular jobs. Your assertion is incorrect.

First, let's review some points made previously throughout the thread:

• There has been no indication that the stat system is changing in Sword and Shield, so they can reuse the same stats.
• Even if the stat system changes in Sword and Shield, there is no reason they can't use the same algorithm they used to change the pokemon that were included on the pokemon that weren't.
• You can re-use the existing pokedex entries, which has already been done in previous generations.
• When you re-use existing pokedex entries, you can re-use the existing translations, so there is no need to re-translate existing text.
• The previous two points don't even matter because the standard in Sun, Moon, Ultra Sun, and Ultra Moon is to just not have the pokedex entries for non-native pokemon, that's for Pokemon Bank and/or Pokemon Home to provide.
• Reading, parsing, and mass-editting of delimiter-separated value files is a well-established, frequently used, core tradition in many computer-based fields.

For clarification on how I'm going to harp on this point, ASCII had control codes for this purpose in 1963, grep was released in 1974. Don't make the human type the thing over again, have the computer box automate it. Higher-end models of computers are even capable of running more than one task at a time, it is quite possible that they can do all of this while the human uses the computer to write the tutorial on Dynamaxing.

The stats have been taken care of. The text have been taken care of. The models have been done. The animations have been done. So that just means your "name" for the thing that I'm missing when I provide an example is... "Whatnot." Well, there you go, that's also probably the name of a series that releases two games every year featuring thousands of monsters and they always keep bringing back their whole cast in every release while also adding new monsters.

This is extremely advance computer programming right here.

[Lord Raziere]

deuterio, all your argument seems to be is "no stop, they're a small indie company thats been doing this for 23 years and making billions of dollars, how can us randos possibly do better than them?" and taking the company at only at their word, when the best principle when observing or discussing any company is to take their word with a grain of salt.

the reasons they state they're doing this make no sense while the reasons people can work out (Pokemon Go suddenly being incredibly profitable, sudden shift towards more mobile games by pokemon, Game Freak fearing its place as the sole pokemon game developer is threatened so its putting less effort into it, and putting effort into TOWN which is of course a game no one has heard or cares about because its just a more generic animal crossing.) make far more sense.

companies rarely state the real reasons they do anything because often they believe they aren't profitable to state. and it doesn't seem like they thought about this long term, if they're announcing this after the great showing they did at E3. there are other factors at work here other than what Game Freak admits, thats how it always goes with corporations.

[OracleofWuffing]

Ok. What are the moves that raises accuracy?

Lock-On and Mind Reader are the big ones. There's also the No Guard ability, but that's a bit less easy to get around with OHKOs.

[deuterio12]

Ok. What are the moves that raises accuracy?

Lock On for one. There's a bunch of other moves that will also guarantee a 100% hit.

I will reiterate: All of the animations work. All of the models work. They had this available during X and Y, where it worked fully there. They have less resources and less money than Game Freaks, and they did this on their free time, outside of their regular jobs.

I will reiterate: graphics alone do not a game make. And you only showed me just a short video. Not even all the animations.

And great that those dudes are willing to waste their free time for no profit, but Game Freak employees expect actually be paid for their work time. You can't just expect Game Freak to overwork their employees for no pay.

First, let's review some points made previously throughout the thread:
[LIST][*]There has been no indication that the stat system is changing in Sword and Shield, so they can reuse the same stats.[*]Even if the stat system changes in Sword and Shield, there is no reason they can't use the same algorithm they used to change the pokemon that were included on the pokemon that weren't.
based fields.

That's already wrong right there because I already showed how moves get changed between games in a non-standardized way so you can't just throw an alghorhytm at it and hope for the best.

The stats have been taken care of.

They haven't.

The text have been taken care of.

For previous games, not the new ones.

The models have been done. The animations have been done.

One picture and one short video from random dudes on youtube do not count as "done" I'm afraid.

Unown. Smeargle.

Nice try, but their one moves did get their stats changed.

A new engine that works the same way as the old engine on a new system that runs the same language as the old system.

How do you know? Do you work for Game Freaks by chance and have direct access to the source code?

[Qwertystop]

That's already wrong right there because I already showed how moves get changed between games in a non-standardized way so you can't just throw an alghorhytm at it and hope for the best.

They do have the option to just not change learnsets, or leave them mostly the same. And they certainly could – an algorithm is just a set of rules, for all that some tech companies like to hype themselves up by claiming theirs is (figuratively) magic.

For each type, pick a few physical, special, and status moves that generally fit with each ten-level range. Assign each mon a random selection of these for each level based on their type, with a higher share of physical or special moves according to their higher attack stat, and more status moves if they have a higher defensive stat, and a few off-type moves mixed in. It won't be perfect but it'll give you a reasonable baseline, take much less time to write, and it's not like existing movesets always make sense anyway. You can go in and tweak the list aftwerwards for specific mons you want to focus more on (e.g. things that can be caught wild in that game, things used by Gym Leaders and other important NPC battles, legendaries and fan-favorites and anything with a signature move from older games).

For previous games, not the new ones.


For previous games, not the new ones.

New games can use the same animations and models and dex entries and stats. I don't think anyone expected them to be otherwise – they didn't the last few games.

How do you know? Do you work for Game Freaks by chance and have direct access to the source code?

The game is visibly doing the same things as every prior game, and these things are not complex as games go – moving around on a fixed map (no non-cosmetic jumping or flight to need to deal with actual collision, even) and the same battle system they've been iterating on since the start. There's no reason for them to have done it differently; they've got more than twenty years experience on this, the games aren't technically demanding to need heavy re-optimization for new hardware, and even if they were programming everything directly in assembly instead of a higher-level language for some mad reason, they're still on ARM which would make porting relatively easy (to the limited extent that any assembly work can be easy, which is admittedly not much).

You realize that Bulbapedia has the pokedex entries for each pokemon for each game and they're different, right?

I'll grant this one. I remembered mostly re-use from those times when I had incidentally read the dex entries from there, but it seems I misremembered the scope – there is a lot of re-use (and very minor rephrasings) but it's not total. Still, there are species that didn't get any new entries in a given generation – Riolu has the same two entries in English for all eight games of gen V and VI, and one of those two was also its entry for Platinum, for example. I don't think many people would fault them for copying all the entries up.

[enderlord99]

Nice try, but their one moves did get their stats changed.

Ditto, then.

[deuterio12]

Ditto, then.

Transform got recoded plenty of times too. Short list (there were actually a lot more changes):
-Generation II transform had to take in account shiny pokemon.
-Generation III transform had to take in account pokemon base abilities and multiple forms.
-Generation IV transform needs to take in account Quick Powder.
-Generation V ditto gainst his own hidden signature ability, Imposter, that interacts specifically with Transform.
-Generation VI transform needs to take in account mega-evolutions.
-Generation VII transform needs to take in account normalium Z.
-LGEP transform needs to take in account the unique partner pikachu/eevee.

So as you can see Ditto works in different ways each game as Transform (or even ditto itself) changes. Like, how will Transform interact with Dynamaxed/raid boss mons?

Some people going all "lol just copy-paste everything it's what they always do and what's the worst that could happen" and makes me seriously wonder if they bothered playing more than a single pokemon game or any at all.

There's big sweeping changes like splitting special into special attack and defense (with different ratios for each mon so just going with a blind alghorytm is a big nope), then there's more minor but still important stuff like Struggle and Shadow Tag being changed because two Wobuffets with leftovers would lead to an endless slapfight where they would struggle forever unable to hurt each other faster than they were healed and unable to switch pokemons either because of shadow tag either, with the only solution being turning off the game. So Struggle was changed to deal recoil equal to 1/4 max HP regardless of how much damage you actually dealt and Shadow Tag was changed to cancel itself if both pokemons have it.

So if they just do a rush job of trying to ram all the old pokemons in whitout a care for reviewing their abilities and moves one by one, the cumulative chance of something nasty slipping through the cracks increases drastically. Maybe some amatauers on youtube can afford to do such mistakes with their free time, but not an actual professional company like Game Freak while preparing a product to be sold for money. They need time to check each and every pokemon individually and make sure they meet a minimum level of balance. And yes the first games had plenty of bugs and problems, but they've been working hard at improving things with each game. Including again adding whole new types of pokemons and moves like Dark/Steel/Fairy.

Eventually a level would be reached where they simply don't have the resources to properly update every pokemon at the same time, stats and graphics and all. That level has been reached. But they still leave the door open to patching the pokemon left behind later. It's the best of both worlds really. The game still gets released whitout delays (and remember they have the anime/TCG companies and whatnot are breathing down GF's necks to deliver it in time) and the other pokemons may be become playable someday still. The alternative being a bugggy mess where nobody is playable because the game is constantly crashing from unforeseen bugs or delayed for Arceus knows how many years.

[Lord Raziere]

"No Plans to do that" is not "open to doing that" it in fact sounds like the exact opposite: they didn't take this into account in the least and put no thought into whether they should announce a patch, and thus never thought they would HAVE to, or SHOULD. if they were open to this DLC hope, it'd be a no brainer to just announce it, say that they're going to get in there eventually. they have not said this. They have in fact gone out of their way to say that they are going to cherry pick the mons from now on depending the games theme, implying that they're not going to bother.

nothing you say matters about this, deuterio. no amount of positive spinning will convince me, because they said that this was a clear permanent policy change, not a one time deal where we can hope they'll patch it in later. and I'd much rather have the pokemon in like all the other generations than any nonsense about dynamax or whatever stupid minigame they come up with, or whatever else you think that is so worth giving up the games core functionality, because you have not said a single thing that is worth giving up who knows how many pokemon so far. we're not getting a good deal, its less than what we got, and I truly don't care if your "good riddance" to who knows how many pokemon thinking that all the "bad ones" are going to be cut.

I'd much rather have the waiting for years honestly. a delayed product will be good someday, but a bad release early is bad forever. first impressions are everything, and they have not put their best foot forward. and perhaps the casual pokemon players don't care about this- fine. but don't expect someone like me, who has been playing since Generations 1 and 2 to be happy about it or see it as positive. I love new pokemon designs, I want to see more, I want to a new pokemon game, but its with the expectation that it expands, that it is open to my efforts, that all my catching them all hasn't been in vain- and they just shot that in the heart.

because sure, in previous generations you'd never truly be able to catch them all for long, there would always be some mythical pokemon you haven't gotten yet, maybe some temporary stretches of time where you finally did get them all before the next games out, but there is a difference between that hope just being out of reach and suddenly getting that hope crushed entirely. and perhaps you don't understand WHY its important to Catch Them All, why I'd be so adamant about this, but you don't have to. I don't expect you to, because its not romance that you have. its just not an idea you ever bought into. thats okay, but no matter how much sense you think your talking, that dream of Catching Them All is dead on the ground and no amount of defending Game Freak will resurrect the corpse.

It doesn't matter the reason why that dream is dead, it matters that it is. That its gone. and perhaps I can hold out a little longer to see if the death was just a temporary coma or something, but given how Game Freak still has not given any definitive plan on how they're going to be moving forward in the wake of the fiasco, just ignoring the people who do have a problem with it for their new mobile game, I'm not inclined to look at anything they're doing positively. whats their plan for Home? how do they plan on winning me back? what is their direction, and is it a direction I'm okay with? the more the I know sooner, the sooner I can make a decision and either stick with them for sure or just cancel my order and leave. your word deuterio? means squat for that, because your not Game Freak, all you can do at best is speculate. any savvy customer of anything knows to temper their expectations so that they don't get burned by falling for too good to be true ideas like "oh maybe they'll patch it when they haven't said definitively they will yet!".

so unless you magically have 100% accurate knowledge of what Game Freak will do next to solve or not solve this problem-because to me its Most Definitely A Problem and not anything else for what reasons it exists, good or bad I don't care- the dream is dead. does this make me crazy unreasonable hardcore fan? I don't know, and I still don't care. not all of us can shrug it off, because maybe I don't want to. maybe I'm just the kind of person that really does get value out of the collecting aspect that you'll never understand, or whatever, maybe my enjoyment doesn't take the shape of the kind that yours does. whatever the reason, all I know is that the National dex is a vital experience to pokemon for me regardless of whatever anyone, ever says.

[NatureKing]

Transform got recoded plenty of times too. Short list (there were actually a lot more changes):
-Generation II transform had to take in account shiny pokemon.

If opponent = shiny, use shiny texture
-Generation III transform had to take in account pokemon base abilities and multiple forms.
Multiple forms are different pokemon.
-Generation IV transform needs to take in account Quick Powder.
Quick Power = Check Mon = Ditto, Apply Defense bonus in X order
-Generation V ditto gainst his own hidden signature ability, Imposter, that interacts specifically with Transform.
- as in, imposter refuses to trigger if the opponent is transformed? Check if opponent = transformed if yes, do not trigger
-Generation VI transform needs to take in account mega-evolutions.
- No, they are just another form of pokemon
-Generation VII transform needs to take in account normalium Z.
No, Transform-Z was a new move entirely.
-LGEP transform needs to take in account the unique partner pikachu
they are just another pokemon in the code

So as you can see Ditto works in different ways each game as Transform (or even ditto itself) changes. Like, how will Transform interact with Dynamaxed/raid boss mons?
Hard to say without having the code to hand. If it works like it currently does with mega evolutions, then it will transform into a mega version. If it works like imposter does with transformed pokemon, it will check if a mon is dynamaxed, then fail to trigger if so. If the mon is

Some people going all "lol just copy-paste everything it's what they always do and what's the worst that could happen" and makes me seriously wonder if they bothered playing more than a single pokemon game or any at all.

Having existing bad coding is not an excuse.

There's big sweeping changes like splitting special into special attack and defense (with different ratios for each mon so just going with a blind alghorytm is a big nope),

Phys/Special split was a 1 time thing. The only difference is that during design process is now the balance team can decide 'do we want this pokemon to be specially offensively weak/strong and or specdef weak/string instead of generically weak/strong. These stats already exist in game now, so the only thing that needs to be adjusted are the 150 or so new knee, and of those, let's be honest, maybe only 20-30 of these actually matter.

then there's more minor but still important stuff like Struggle and Shadow Tag being changed because two Wobuffets with leftovers would lead to an endless slapfight where they would struggle forever unable to hurt each other faster than they were healed and unable to switch pokemons either because of shadow tag either, with the only solution being turning off the game. So Struggle was changed to deal recoil equal to 1/4 max HP regardless of how much damage you actually dealt and Shadow Tag was changed to cancel itself if both pokemons have it.

What you are looking for here is 'game breaking bugfix patch' downloadable by Internet connection.

So if they just do a rush job of trying to ram all the old pokemons in whitout a care for reviewing their abilities and moves one by one, the cumulative chance of something nasty slipping through the cracks increases drastically. Maybe some amatauers on youtube can afford to do such mistakes with their free time, but not an actual professional company like Game Freak while preparing a product to be sold for money. They need time to check each and every pokemon individually and make sure they meet a minimum level of balance. And yes the first games had plenty of bugs and problems, but they've been working hard at improving things with each game. Including again adding whole new types of pokemons and moves like Dark/Steel/Fairy.

Making a type isn't much trouble. Getting the balance right can be tricky (re Steels OP resists), but making a new type is no more difficult than adding the code to a table, which xrefs a damage multiplier. In theory, you could create new degrees of effectiveness, where the multiplier isn't x2 or x0.5, or x0.

Eventually a level would be reached where they simply don't have the resources to properly update every pokemon at the same time, stats and graphics and all.

The games graphics of the pokemon's 3d models are significantly changed? Really? That noticeably? You mean, like, by shaders and lighting? You want to see what shaders and lighting can do, have a look at Minecraft.

That level has been reached

Depends how much money they put on it. If you have a 10 hour job but pay someone 5 hours to do it, you only get 5 hours done. Pokemon has money to make the works take place. They (Game Freka) still project this meme of small independent studio, when they are not.

But they still leave the door open to patching the pokemon left behind later.

By patching them in, you mean paying £45 for a new game in a year or two right? Are you really going to care that skitty has the same polygons in this game, rather than having 2000 more?

And you could use the exact same response to putting the models in the game. 'we realised we didn't have the time to release all the new models on release, apologies, so we used some of the older ones, but we are now releasing the hifi models and updated animations in gradual patches'

It's the best of both worlds really. The game still gets released whitout delays (and remember they have the anime/TCG companies and whatnot are breathing down GF's necks to deliver it in time) and the other pokemons may be become playable someday still. The alternative being a bugggy mess where nobody is playable because the game is constantly crashing from unforeseen bugs or delayed for Arceus knows how many years.

The alternative being using two pokemon in an endless battle clause fixed by turning the power off and on again to restore to your most recent save point?

I am astounded by your brown nosing, can do no wrong to this scalping. If you could pose more convincing arguments, you could conceivably be declared a shill, but there isn't even that there for you. I'm sorry, just stop man. It's embarrassing.

The fact that all of what you have said is an argument you have conjured as well and made up on the spot when the defense given is 'we have done it for balance' has also completely passed you by.

[Fable Wright]

If opponent = shiny, use shiny texture

You, uh. You're not a programmer, are you? Or even had to do database management?

Let's say that someone already had every pokemon have a shiny flag baked into their data in a way that's difficult to replicate, based off the hidden TSV and ESV. When you transform...

...How do you *temporarily* change Ditto's TSV/ESV to make it shiny, and then undo that change? Finding an appropriate temporary data storage is nontrivial.
...How do you ensure that your temporary application/removal of shininess doesn't persist?
...How much does the code responsible for Shinies change during development?
...How many QA hours do we put into ensuring that Transform works appropriately?

Reworking the structure of the move can change the entire data structure holding what moves are and can do. This is a non-trivial change that could easily cost 40 man-hours... for one move.

There are 742 moves.

In theory, if you planned this out ahead of time and had no existing code base, the change would be easy. But in practice, it gets expensive fast.

Having existing bad coding is not an excuse.

It's not a matter of existing bad coding.

It's a matter of having legacy code.

The further along in development you change something, the more it's going to cost to change. That's just a fact of development, due to how code architecture expands into all facets of development. Changing something late in development is expensive.

Overhauling a system while trying to otherwise inherit the old structure of the existing game is really expensive.

Developing a whole new, better system from scratch is monumentally expensive.

And this issue, as well as expectations of pokemon games as well as just general time, have compounded through 8 generations, over 20 years, and at least four consoles. It's not sustainable.

They're going to need to change, or the games are going to be too expensive to make.

It sucks, no one is to blame, and it needs to happen. If you disagree, that's your opinion. Don't buy the game. That's a risk that the development team accepted when deciding to pull the trigger. But the fact of the matter is that you are grossly misassessing how software development works. The enormous number of updates needed are both expensive and a factor that limits innovation on pokemon games, and that the enormous changes that come with moving to the Switch made this the natural breaking-off point to integrate the changes that have been so badly needed for years, no matter how much you try to abstract it.

I have a printout of this at my desk at all times for a reason.

[Bartmanhomer]

This whole argument isn't going anywhere and I really don't have an opinion on either side because it's irrelevant to me.

Anyway in an unrelated Pokemon topic. I'm doing well with Pokemon Let's Go Overused Format games. I've been switching so many Pokemon teams.

[deuterio12]

In theory, if you planned this out ahead of time and had no existing code base, the change would be easy. But in practice, it gets expensive fast.

Quite well put. The difference between theory and actual practice. It's easy to point to a short video and picture while claiming "why don't they do that for a whole working game", but turns out that a full working game is much harder to do than a short video and a picture.

It's not a matter of existing bad coding.

It's a matter of having legacy code.

The further along in development you change something, the more it's going to cost to change. That's just a fact of development, due to how code architecture expands into all facets of development. Changing something late in development is expensive.

Overhauling a system while trying to otherwise inherit the old structure of the existing game is really expensive.

Developing a whole new, better system from scratch is monumentally expensive.

And this issue, as well as expectations of pokemon games as well as just general time, have compounded through 8 generations, over 20 years, and at least four consoles. It's not sustainable.

They're going to need to change, or the games are going to be too expensive to make.

It sucks, no one is to blame, and it needs to happen. If you disagree, that's your opinion. Don't buy the game. That's a risk that the development team accepted when deciding to pull the trigger. But the fact of the matter is that you are grossly misassessing how software development works. The enormous number of updates needed are both expensive and a factor that limits innovation on pokemon games, and that the enormous changes that come with moving to the Switch made this the natural breaking-off point to integrate the changes that have been so badly needed for years, no matter how much you try to abstract it.

Extremely correct. It would be wonderful if inheriting all the old features of a previous version was fast and easy... But programming simply doesn't work that way. Everything you add/change can and will have repercussions accross the whole code, and it's simply impossible to future proof for all possible cases, even more for in a timespan of decades and while switching hardware. Eventually one needs to discard some old things to stay in budget and meet deadlines. Everybody that actually works in the professional programming world does it.

This whole argument isn't going anywhere and I really don't have an opinion on either side because it's irrelevant to me.

Then you're in our side since you indeed agree that trying to brute force every bit of legacy code is simply not a priority.

Anyway in an unrelated Pokemon topic. I'm doing well with Pokemon Let's Go Overused Format games. I've been switching so many Pokemon teams.

See, you show that there can be good fun with a smaller pokemon pool as long as said pool is properly polished.

[Bartmanhomer]

Quite well put. The difference between theory and actual practice. It's easy to point to a short video and picture while claiming "why don't they do that for a whole working game", but turns out that a fully working game is much harder to do than a short video and a picture.



Extremely correct. It would be wonderful if inheriting all the old features of a previous version was fast and easy... But programming simply doesn't work that way. Everything you add/change can and will have repercussions across the whole code, and it's simply impossible to future proof for all possible cases, even more for in a timespan of decades and while switching hardware. Eventually one needs to discard some old things to stay in budget and meet deadlines. Everybody that actually works in the professional programming world does it.


Then you're in our side since you indeed agree that trying to brute force every bit of legacy code is simply not a priority.



See, you show that there can be good fun with a smaller pokemon pool as long as the said pool is properly polished.

I'm not on your side or theirs. I'm neutral for taking either side because this whole argument is ridiculous. But I'm really enjoying the game in PokemonShowdown.com.

[deuterio12]

I'm not on your side or theirs. I'm neutral for taking either side because this whole argument is ridiculous. But I'm really enjoying the game in PokemonShowdown.com.

Fair enuff, you're enjoying the version of the game you prefer instead of just demanding the impossible.

So, how many of your Pokemon Let's Go Overused Format teams use the special Eevee and what are your favorite picks for Mega Evolution out of curiosity?

[Qwertystop]

You, uh. You're not a programmer, are you? Or even had to do database management?

Let's say that someone already had every pokemon have a shiny flag baked into their data in a way that's difficult to replicate, based off the hidden TSV and ESV. When you transform...

...How do you *temporarily* change Ditto's TSV/ESV to make it shiny, and then undo that change? Finding an appropriate temporary data storage is nontrivial.
...How do you ensure that your temporary application/removal of shininess doesn't persist?
...How much does the code responsible for Shinies change during development?
...How many QA hours do we put into ensuring that Transform works appropriately?

Reworking the structure of the move can change the entire data structure holding what moves are and can do. This is a non-trivial change that could easily cost 40 man-hours... for one move.

There are 742 moves.

In theory, if you planned this out ahead of time and had no existing code base, the change would be easy. But in practice, it gets expensive fast.

Well, that's certainly a way it could be done, and you're right that it would be a huge pain, on the other hand, what about this possible model:

- There are two kinds of moves: simple, and complex
- Simple moves are represented by a data structure containing power, accuracy, chance of each status effect, chance of each stat change, magnitude of each stat change. Also an animation. Many, perhaps most, moves can be represented in this way.
- Complex moves, in addition, have one extra field containing a function, executing arbitrary code on gamestate.

That model eliminates your "reworking the structure of the move can change the entire data structure holding what moves are and can do" point, entirely. The structure would only need to be changed if it's decided that some new class of effect, previously considered "complex", is common enough to need to get a new field in the simple-move structure. Even then, unless someone's doing pointer arithmetic on move definitions, any necessary changes can be handled by the compiler. Modular code and data, that's the ticket.

As for the other points regarding shininess: You've found an unnecessarily complicated way to do things. Persistent sprite changes in-battle, triggered by moves or abilities, which end when the battle ends or the mon switches out, are already a thing that exists – Substitute is as old as Ditto is, plus Castform's and Cherrim's abilities in later generations. Use that same mechanism, set the user's frontsprite and backsprite to those of the target, whatever they happen to be.

You're right about the need for some QA, but I don't think it's on the order of many hours

If you're in a situation where any change to any little thing needs reworking of large portions of code that ought to be unrelated, either you've done some very low-level manual optimizations regarding memory layout and/or pointer arithmetic, or you've built your system poorly without spending nearly enough time on design. I might have believed that Pokémon would be like that early-on, but by now it shouldn't be straining the hardware at all (except perhaps with the visuals), and they've had plenty of time to think things through and set them up cleanly.

I might not be surprised to see a setup like the one you propose, where changes to the in-battle state of a mon need to be explicitly rolled back at the end of the battle to keep them from persisting, in a project where the developers had no expectation it would need to be touched or re-used beyond the ship date. At this point, though... no. Pokémon is far too well-established for anyone to proceed that carelessly, and regardless of how much code might be re-usable we know there's been at least one major engine restructure (Gen 3) since it was that small or otherwise might not be expected to continue indefinitely. If anyone working at Game Freak was thinking that short-term about a main-series Pokęmon game within the past decade, I wouldn't trust them to design so much as a linked list.

[Bartmanhomer]

Fair enuff, you're enjoying the version of the game you prefer instead of just demanding the impossible.

So, how many of your Pokemon Let's Go Overused Format teams use the special Eevee and what are your favourite picks for Mega Evolution out of curiosity?

First off, I don't use the "special" Eevee. I use all three Eevee evolutions. Second my favourite Mega Evolution is Mega Charizard X.

[Lord Raziere]

So what?

It still doesn't have all the pokemon.

its not a more polished game experience, its a more shallow one. at least the Lets Go games were being honest about how they only care about the casual player. I'll start liking any of this when they announce you can catch them all once again and not a moment sooner.

cause here is the thing, things like this snowball. the next game after this is going to be worse. they are just going to keep doing stuff like this, and the game will suffer for it. how long until they reintroduce the lets go mechanics and make them permanent? how long until pokemon is so watered down that it loses what makes it good at all? they've lost trust that cannot be regained easily.

because they failed to live up to expectations, and thats all that matters. this just convinces me more and more that I should play fan games rather than deal with a game company that clearly cares less and less about me and people like me who actually went through and valued an idea that they apparently decided was a lie all along.

rather than try to hope that they will somehow fix this or that things will somehow magically change in five months when november rolls around but guess what, that has never happened. if you don't care about national dex, fine, whatever, but I find it insulting that you are still trying to convince me to be fine with it.

and your using someones unwillingness to be caught in up in argument as implicit agreement to your view which is the most scumbag thing I've ever heard of. Let the person speak for themselves and leave them alone, you jerk. they want nothing to do with it, let them not be involved rather than making them apart of it with their absence. have some respect, geez.

[tgva8889]

For me, I've never cared about having access to the whole Pokedex because it was never really that attainable anyways. It's not hard for me to believe removing some Pokemon would make a better game experience; it's certainly been my experience with other games that less is often more. I wouldn't mind seeing what they do with it. But clearly I'm not the kind of people this decision matters to, even though I've played every single game up to this point and played some competitively.

[deuterio12]

First off, I don't use the "special" Eevee. I use all three Eevee evolutions. Second my favourite Mega Evolution is Mega Charizard X.

That's surprising with fully skipping the "special" Eeevee since they offer plenty of unique utility in the format. So do you use all three Eevee evolutions at the same time (that would be quite the thematic team) or do you rotate them or pair them up in twos?

Less of a surprise with Charizard, they're like the most popular pokemon right behind pikachu itself.

For me, I've never cared about having access to the whole Pokedex because it was never really that attainable anyways. It's not hard for me to believe removing some Pokemon would make a better game experience; it's certainly been my experience with other games that less is often more. I wouldn't mind seeing what they do with it. But clearly I'm not the kind of people this decision matters to, even though I've played every single game up to this point and played some competitively.

The lesson of "less is more" is also quite important. More just for the sake of moar is simply not good design.

[Bartmanhomer]

That's surprising with fully skipping the "special" Eeevee since they offer plenty of unique utility in the format. So do you use all three Eevee evolutions at the same time (that would be quite the thematic team) or do you rotate them or pair them up in twos?

Less of a surprise with Charizard, they're like the most popular pokemon right behind pikachu itself.



The lesson of "less is more" is also quite important. More just for the sake of moar is simply not good design.

Yes. I do use all three Eevee evolutions at the same with the Dragon Type Pokemon and Legendary Birds. I also used Mega Blastoise, Mega Aerodactyl and Mega Gyarados.

[Lord Raziere]

News of official response from Masuda on the national dex controversy

it is tepid and doesn't address concerns in the least. saying that the pokemon MIGHT be in other games in the future. so no, no patch is coming and pokemon will be cut from now on. if I don't cancel my order of Sword, it'll be the last main pokemon game I buy.

[NatureKing]

You, uh. You're not a programmer, are you? Or even had to do database management?

Let's say that someone already had every pokemon have a shiny flag baked into their data in a way that's difficult to replicate, based off the hidden TSV and ESV. When you transform...

...How do you *temporarily* change Ditto's TSV/ESV to make it shiny, and then undo that change? Finding an appropriate temporary data storage is nontrivial.
...How do you ensure that your temporary application/removal of shininess doesn't persist?
...How much does the code responsible for Shinies change during development?
...How many QA hours do we put into ensuring that Transform works appropriately?

Reworking the structure of the move can change the entire data structure holding what moves are and can do. This is a non-trivial change that could easily cost 40 man-hours... for one move.

There are 742 moves.

In theory, if you planned this out ahead of time and had no existing code base, the change would be easy. But in practice, it gets expensive fast.



It's not a matter of existing bad coding.

It's a matter of having legacy code.

The further along in development you change something, the more it's going to cost to change. That's just a fact of development, due to how code architecture expands into all facets of development. Changing something late in development is expensive.

Overhauling a system while trying to otherwise inherit the old structure of the existing game is really expensive.

Developing a whole new, better system from scratch is monumentally expensive.

And this issue, as well as expectations of pokemon games as well as just general time, have compounded through 8 generations, over 20 years, and at least four consoles. It's not sustainable.

They're going to need to change, or the games are going to be too expensive to make.

It sucks, no one is to blame, and it needs to happen. If you disagree, that's your opinion. Don't buy the game. That's a risk that the development team accepted when deciding to pull the trigger. But the fact of the matter is that you are grossly misassessing how software development works. The enormous number of updates needed are both expensive and a factor that limits innovation on pokemon games, and that the enormous changes that come with moving to the Switch made this the natural breaking-off point to integrate the changes that have been so badly needed for years, no matter how much you try to abstract it.

I have a printout of this at my desk at all times for a reason.

1. It's my job to ensure that non compatible accounts systems and databases of over 250 businesses in my career, ranging from SME to multinational, multi billion dollar companies can have their various account systems set up and run during accounts switches. I know what I'm about.

2. You are comparing one of the single most complicated moves in the game to the majority of Moves which are X Power, Y Accuracy, Z Type, with A Rider, B% Chance to inflict on C Target. Not a strong argument.

3. Legacy Code/Bad Code: Which costs money to correct. Which pokemon has plenty of. Which game freak has plenty of.

4. No, they do not need to change. They are happy enough turning out the the previously linked animations video, Battle mechanic of '3 turns of big pokemon', poor overworld models and textures, cut content games, become some poor sap will be happy enough to suck their dusty teat and try to defend them.

5. Changing systems is the perfect opportunity to turn out a polished capable game showing off the best of your product and a fantastic baseline, and not use the 'oh, we are stuck with what what we are doing because we have 20 year legacy code we wrote on a 4mb flash cart in a garage somewhere'

[tgva8889]

How many of us actually know anything about animating? I don't know that much about it, but from what I've seen it's a lot more work to make a 3D model that looks at all like a living creature than one might guess, so I wonder how much work it actually takes to translate some 900 creatures like that into a game, each with distinct body types, skeletons, etc. Like, I wonder if making all of them even slightly better would add a year to the animation time.

This is not a defense, just something I think about when thinking about Pokemon, which is mostly unique in this realm. Like, they're making the models anyways, they exist, but I wonder, how much time and effort would it take to increase the fidelity on every single one of them? Probably a long time, which is why they've kept the style for so long.