Risk assessment can add a lot of depth to a game.
Of course though hecatom. The devs put in BBhoods infinite on purpose in vsav just becuase.
Real talk, people that have been playing games at a higher level and already understand “the game” (the tactics the mindset the timing, you’ll understand if some of you guys actually play) can break down the execution argument much better than you guys that don’t.
Keep in mind that the people you are arguing with are the ones PLAYING THESE GAMES. So don’t you think that to back up their argument they would be using the fact that all scrubs gtfo, hadoukens should be SRKs? No, they are giving GAMEPLAY REASONS as to why easier execution does affect games. I know that all mvc2 players would agree if I said that making sentinels refly easier would hurt the game. Allowing strider to call ourobouros with no waiting for the robots to leave the screen would affect the game. Being able to mash out srks in ST whenever you want would hurt the game. I have played ST and I know that there are many times where i just wanna mash that SRK on dhalsim in the corner.
Nothing has to be insanely hard like 2 frames to input an srk but fuck son, it really looks like you guys want fighting games to stop being fighting games.
i’m glad you think that it is just so easy to get a list of recoveries out of a fighting game engine
If you had access to the source and data I don’t see why this would be challenging at all. You whip up a timer, put in a few log commands and collate the data. Why do you think this would be hard?
You think every one is going to have the access the source code laying around for every game?
Edit: Not every game uses the same language some use C, M68k assembly, SH2 assembly, even assembly for custom made processors, PPC Assembly, and I can go on.
“Frame Data”.
Seriously, wtf? You could probably get frame data for fighting games that 5 people played back in 1953
If the frame data isn’t actually 100% accurate, do what is in Ilthuain’s post. Still, the published data is probably close enough for you and I.
I believe this conversation was about game development, and discovering if something links or not during the development process.
You think the people not on your side aren’t PLAYING THESE GAMES?!
so you want to do a program that runs a permutation on every normal and special of every character to make it sure that they don’t have links or the links have wider frames of opportunity
that task is already overwhelming if the cast is big and has a good amount of moves, how would you factor screen positioning and height on the jumping moves?
how would you factor the differences of hitboxes between each character?
do you know what is an np problem?
the logic of doing a program that finds the answer is not difficult
getting the answer on human time of that problem, when the number of factors are not polynomial is
I’ll tell you this much every fighting game while Poo was working at Capcom there was no plan for Intention for 2 frame links. I say 2 as the turbo would skip frames making 1 frame links almost impossible.
just using the frame data on its own and manually testing the results to filter out potential links that don’t work because of other variables wouldn’t take much time at all. aren’t there already tools online that will give you lists of potential punishes based on frame data alone? that would be the same logic. it’s at least a decent approximation to start with. would be n^2 complexity i think?
Yeah it would. It’d be timeconsuming but doable as a person (for instance Ryu has just about 50 different moves in SF4AE, about 2500 possible links to check, some of which you can throw out out of hand, like his ultra1 being -25 or things like it being impossible to have both ultras at once), but would actually be trivial using a computer.
On the subject of lab time, I think there is a big difference between “experimental” lab time and “repetition” lab time.
I find experimental lab time fun. I was frustrated the other day because a guile was sweeping me constantly and I didn’t have an answer for it other than blocking the second kick and giving him meter. A few minutes in the lab taught me that I could do a quick focus attack or light SRK between the first and second kick to punish. It was fun labbing it and figuring out a strategy. I have done this numerous times (“How do I beat blanka’s slide? How do I beat Guy’s stupid elbow drop?”) and it is an entertaining experience. Same goes with figuring out if X move stuffs Y, or if you can use Y as an anti-air against Z.
Spending many hours pressing the exact same button combination is just not that fun to me by comparison.
Sure.
No it isn’t. It might be overwhelming if you’re not an engineer, but a gameplay engineer wouldn’t have a problem with this. Personally, I wouldn’t have a problem with this, and I’ve created data tables from runtime output with far more variables.
I don’t think I would need to. If we look at the first possible cycle where player input is accepted and compare that with the last possible cycle that a hit could be registered, we would have enough data to create the table. If a move can execute and impact on the last cycle before the player moves from a jumping state to a standing state, then we’ll use the jump-to-stand transition as our “ready in X time” variable.
You’re making the problem far more complicated than it needs to be.
We don’t need to unless the timing returns a positive on a “can potentially link” test, which would assume that the target could always be hit on the first cycle of potential contact. If it does, and we want certain moves to link to other specific moves based on distances and hitboxes, that would be a different problem.
Yup, and this is not one of them. There aren’t enough moving parts in a fighting game for me to believe that this would be impossible (unless it’s done in Unreal. That would be difficult because engineering in Unreal is difficult regardless of the task ;-))
I’m not so certain I would take “Poo”'s word for it, but in general, Capcom’s engineering doesn’t exactly blow my mind.
Still bugging on ‘HOW CAN YOU POSSIBLY KNOW MOVES STARTUP AND RECOVERY?!?!?! THAT’S JUST MADNESS!’
C’mon man. It’s insanely easy to make a game without links but allows combos. Make all moves at least three frame startup. Make all moves have at maximum +2 advantage even after counterhit. Bam! No links possible. Then allow chaining (magic series or Mortal Kombat patterns) and special cancels and now you have a game with combos and no links.
tebbo tennis sounds awesome but hyperbole gets old.
ideally a game scales (sort of like smash bros melee).
you can enjoy it at every level along the way. the more interested you are, the more there is to learn strategy wise AND execution wise.
but really that is already true for everything. you play at your level with people around your level. there is nothing wrong with execution heavy elements in games. some of us like that aspect and enjoy mastering those things. why is that an issue? we shouldn’t have an advantage for putting in the time?
honestly what game works like that. where the person who puts in the time isn’t rewarded? crocodile dentist i guess? hungry hungry hippos?
it’s not like people are winning solely based on their skill with the controls. so i just…don’t see the issue.
That’s one way, sure, but you can still do it and keep wide move diversity (if you wanted to).
That’s the funny thing, it’s gone beyond the ‘hey should links be included’ which is debatable to ‘can links be prevented?’ which is obvious.