DK Jr. springboard discussion

[gstrain]

This question is proposed to all of us. So far no specifications have ever been made either by TG or by DKF. No restrictions are placed on the use of auto fire, multiple buttons mapped for the same purpose on keyboard.

Mame is another animal. Since there is no default mame mapping for DK or DK Jr then maybe consider making it as much like arcade with one button for jump, and no auto fire. But like I said, no one can verify keyboard mapping. 

I don't think we need any official rule accept maybe to mention what is allowed so that this question can be anticipated and be already answered for newer members or anyone else for that matter.

Thank you to any and all who chime in on the question.

Auto-fire is against rules at both TG and MARP and should be here too.  For DK it doesn't matter, but for springboard on DK Jr and possibly DK3 it could.  In MAME its possible to detect probable auto-fire usage by analyzing the inp file for overly consistent input patterns.

TG doesn't have rules on multiple mappings, but MARP has a rule forbidding it, and the use of "macros" where a single key is mapped to a specific sequence of timed input.  For DK games I can't think of a case where it would help you, but for fighting games it can make certain combos much easier since you can use a special key to always press two buttons at the exact same time or always time a button press to occur in the exact same frame as a movement.  In MAME it is pretty easy to detect this sort of multi-mapping when it is pervasive by analyzing the inp.  Human input will never have the frame perfect timing every time that you get with multi-mapping.

Like I said, for DK itself, I don't think either of these matter at all.

-George

[Graham Wolfe]

Ok for speed on springboard I've been multi mapping jump. I guess for some people this is irrelevant and when I play arcade I just double tap the jump to create the same scenario as pressing jump on two buttons on my keyboard. The reason I don't double tap keyboard as fast is mainly because arcade buttons have a spring inside which makes someone with big fingers like myself able to press it faster. On keyboard I get no such luxury and I prefer using two keys side by side to make my double tap. To be honest I don't even know if it actually works but to me it feels easier maybe it's a mental thing. In either case I wouldn't use anything far from normal settings as if one did that such as rapid fire they would be ineffective in an arcade setting. Since TG and DKF don't mind multi mapping then ill continue using that at home for the springboard jump and double Tap in arcade settings. Obviously if this was an advantage I wouldn't have hit 550k in an arcade as I would lack a key function to passing the springboard stage.

In any case I appreciate the quick feedback.

[up2ng]

Definitely some interesting questions from Graham.  Mapping multiple jump buttons would be very helpful to deal with the broken Springboard IMO.  Hopefully no one is doing this on the MAME side currently -- I think that would open a can of worms . . .

Graham may not realize that the videos he is probably watching of top players playing DKJR often feature a "Track and Field Style" approach to using the jump button during the Springboard Jump -- the button itself is unmodified but the players ARE pressing the button incredibly quickly.

Note:  Just saw your last post -- you mention "I don't even know if it actually works" . . .  You may not be aware of some of the technical (code level) details that have been discovered with the Springboard -- long story short, it actually IS broken -- there is a small window of frames that you can press the jump button to get it to function properly, BUT every other frame within this window fails.  Therefore, pressing the button incredibly quickly during this time window IS advantageous to the player since it will increase your chances of landing on a "good frame". . .

[gstrain]

Definitely some interesting questions from Graham.  Mapping multiple jump buttons would be very helpful to deal with the broken Springboard IMO.  Hopefully no one is doing this on the MAME side currently -- I think that would open a can of worms . . .

Graham may not realize that the videos he is probably watching of top players playing DKJR often feature a "Track and Field Style" approach to using the jump button during the Springboard Jump -- the button itself is unmodified but the players ARE pressing the button incredibly quickly.

Note:  Just saw your last post -- you mention "I don't even know if it actually works" . . .  You may not be aware of some of the technical (code level) details that have been discovered with the Springboard -- long story short, it actually IS broken -- there is a small window of frames that you can press the jump button to get it to function properly, BUT every other frame within this window fails.  Therefore, pressing the button incredibly quickly during this time window IS advantageous to the player since it will increase your chances of landing on a "good frame". . .

Note that this sort of "multi-mapping" (two different keys/buttons to the same input) is not actually detectable in MAME .inp files since the file just records inputs, not the source.  The .inp file would be indistinguishable from someone using "Track 'N Field" style presses on a single button.

-George

[Graham Wolfe]

Well what I've been doing with decent success (not 100%) is press two jump buttons twice, so 4 total inputs on keyboard. Funny enough is when i'm on, my single presses are much more accurate but due to having to have near perfect timing which i'm known for in the fighting game community.

I was the first player to find Jin's Lightning God Fist (1 frame in 60 frames in a 1 second window) in America, every other Tekken player I worked with thought it was random like ryu's red fireball in original street fighter. I did it successfully 7/10 times when asked to perform it for a strategy guide we were writing and the chief editor finally gave in when I told him random doesn't happen 7/10 times when they believed it was 5 or 10%.

Anyway back to DKJR, If what Dean said is true then I am not gaining any any benefit from the method I'm using but psychologically it makes me feel as if I am since I'm foolishly pressing both jump at the same exact moment but twice real fast, similar to how I press the arcade jump. I guess I can unbind the left jump button since I'm not even rolling the buttons like I would on an arcade stick due to the lack of a bevel on keyboard buttons.

Btw what is the framerate for these games?

[ChrisP]

You may not be aware of some of the technical (code level) details that have been discovered with the Springboard -- long story short, it actually IS broken -- there is a small window of frames that you can press the jump button to get it to function properly, BUT every other frame within this window fails.  Therefore, pressing the button incredibly quickly during this time window IS advantageous to the player since it will increase your chances of landing on a "good frame". . .

Dean, you may have gone into this elsewhere at some point in the past, but I've been curious why you interpret this as "broken."

It seems to me that not being able to get the superjump every time was intentional. In fact, I would actually find it stranger if it was designed such that you COULD get the superjump every time. To me it just seems like another deliberate random element added to make the game more dynamic, like the 300-500-800 smashes in DK.

Is there something about the way the code is "phrased" that would imply that having to hit the button on a good frame was an accident?

[Graham Wolfe]

well hitting 1*3*5* isnt really too bad. In Street Fighter 2 we have a 1 frame window to hit reversals but what we do to compensate is roll the buttons for certain moves. Funny thing is this my brother prefers single press and he almost always hits his reversals. Obviously no one is perfect but with enough practice (and want to succeed) you can really narrow down your margin of error.

[gstrain]

Well what I've been doing with decent success (not 100%) is press two jump buttons twice, so 4 total inputs on keyboard. Funny enough is when i'm on, my single presses are much more accurate but due to having to have near perfect timing which i'm known for in the fighting game community.

Anyway back to DKJR, If what Dean said is true then I am not gaining any any benefit from the method I'm using but psychologically it makes me feel as if I am since I'm foolishly pressing both jump at the same exact moment but twice real fast, similar to how I press the arcade jump. I guess I can unbind the left jump button since I'm not even rolling the buttons like I would on an arcade stick due to the lack of a bevel on keyboard buttons.

Btw what is the framerate for these games?

Framerate is 60 frames a second and inputs are read at the same rate.

If you're actually pressing both buttons at the exact same moment, the extra button isn't doing anything.  MAME is just reading the input port each frame, and it will be on if either button is pressed.  You can't turn the input port on more than once in a frame regardless of how many buttons you map to it and press simultaneously.

[Graham Wolfe]

yeah it was just nice because my hands are kinda big for a keyboard and im on a laptop. in either case after i got wolfmame all working I just tried the jump using one mapped jump instead of 2, im having identical results as before so it was just a psychological thing.

yeah i was actually mashing both buttons simultaneously not rolling before, just mashing 2 buttons at the same moment because it was easier for my hands than pointer pressing a single key on a keyboard. I can still do this but having one mapped will just make sure no one thinks there is any unfair advantage.

[up2ng]

You may not be aware of some of the technical (code level) details that have been discovered with the Springboard -- long story short, it actually IS broken -- there is a small window of frames that you can press the jump button to get it to function properly, BUT every other frame within this window fails.  Therefore, pressing the button incredibly quickly during this time window IS advantageous to the player since it will increase your chances of landing on a "good frame". . .

Dean, you may have gone into this elsewhere at some point in the past, but I've been curious why you interpret this as "broken."

It seems to me that not being able to get the superjump every time was intentional. In fact, I would actually find it stranger if it was designed such that you COULD get the superjump every time. To me it just seems like another deliberate random element added to make the game more dynamic, like the 300-500-800 smashes in DK.

Is there something about the way the code is "phrased" that would imply that having to hit the button on a good frame was an accident?

That's strange, I've never heard anyone assume that before.  I feel the exact opposite -- to me, something like a "springboard" in a video game is something that I would intuitively believe should "always" work if the player executed it correctly.  There really should be nothing "random" about it.  Now, if you intended to just sort of bounce off of it or if you legitimately "missed" the timing of your jump then I can see why it should be programmed to give no "spring" to your jump in these cases, but when a good player intends to "spring" off of a springboard, they should be able to do so pretty much every time.

Think about other Nintendo platformers / Mario games.  Super Mario Bros. is a great example.  In that game the springboard is "fixed" -- as long as you jump within a reasonable period of time, you spring off of the springboard 100% of the time.

Keep in mind that a 6-frame window is really not a very long time anyways.  It would still take some knowledge and skill to press the button at the correct time to be within a 6-frame window (1/10 of a second).  Now, I haven't actually seen the code for DKJR, but I do have some knowledge about software code and logic in general and I know that these games were made very quickly and are full of bugs, poor design and poor code and based on that I have a very strong hunch that this is a bug.

When you slow it down and look at it frame-by-frame, what happens is that Junior is in contact with the springboard one frame and then Junior moves up a little bit the next frame, but the springboard does not -- they are now NOT in contact.  Then, the next frame, the springboard moves up and Junior does not and so they are back in contact again . . . and so on until the springboard is fully extended.  To me, this just screams huge bug.  The software probably performs a check of Junior's y-position to see if he is in contact with the springboard at the moment the button is pressed, assuming that he will be in contact throughout the motion -- but actually he is not because the two objects animate out of synch with each other.  This is silly anyways since there are probably dozens of other ways this software could have been effectively written, such as at the moment the springboard is fully retracted, simply count off 6 frames and make Junior "super jump" if the button is pressed -- why do we need to check Junior's position?  The ONLY time the springboard should be retracting is if Junior is ON it!  lol. 

I've seen some explanations that the out of synch animations might actually be done on purpose purely for aesthetic reasons when animating these motions -- it might not look as good if the two objects were rising together.  I don't buy it, but I suppose that's possible.  If so, they should have known to account for this in the super jump algorithm, but they didn't.  My belief is that they didn't even realize the objects were animating out of synch with each other and when the game was tested by novice players, the springboard did occasionally work, so they just assumed that it was working and that the testers just weren't very good at it.  But actually, it's broken.

Is it possible that this was actually done on purpose as a design choice for Junior to fail his super jump attempts a large percentage of the time?  Technically I'd have to admit that this is possible since there is no proof to the contrary, but realistically -- no, absolutely not . . . this is a bug, and an incredibly annoying one at that.

[marky_d]

For those who play Junior, what would you say your averages are on making the superjump on the first attempt over the course of a game? I would say mine is probably around 90% using the 'scratch' technique. I heard that Ben made it through an entire game without missing one, and I have done that one time as well (18 of 18). Admittedly making ALL of then will require a bit of luck haha.

[marinomitch13]

I do the double flap and get about 80%, I'd say. If I did a 3 finger scratch with a little timing, I think I'd do better, but I'm in no rush for 10% or so more.

[up2ng]

It still surprises me that people can press a button quickly enough for it to matter in this case, although I have seen plenty of evidence that Track and Field players can manage to get a button press to register incredibly quickly -- without doing that, it would surprise me if just pressing the button twice is really effective. 

Just for fun, I encourage folks to do a Yahoo search for "Online Stopwatch" and click the first link.  Select Stopwatch (not Count Down).  Now, use your mouse and "double click" on the button. . .

When I use a normal (fast) double clicking style of the mouse button, I generally get around 0.165 seconds.  That's pretty fast.  But, it's not even close to fast enough for this particular issue in DKJR.  I'd have to dig around to double check, but I think there's only 3 "good" frames for the springboard -- if there's a 4th good frame, obviously it makes things a lot easier.  But, with only 3 good frames, we're talking about frame #1, #3 and #5 and that's it.  So, if you "miss" on your first button press and land on frame #2, you need to be incredibly fast to press the button again to catch frame #5.  We're talking about pressing the button 0.05 seconds later.

Now, go back to the online stopwatch and see how long it takes you to manage a time of 0.05 seconds.  Just now, after about 15 tries, I managed exactly 0.05 on the stopwatch by using two different fingers on my other hand to press the left mouse button (most of the time my mouse just would not register the 2nd button press).  Using one finger to hit the button twice the best I could do after 7 or 8 tries was 0.067 seconds -- again, not good enough.

If you happened to "miss" by landing on frame #0 instead of frame #2, you'd have a bit more time and with very fast presses I could see catching the frame #5, but you'd have to press the button very quickly.  The point that I'm trying to emphasize is that it's a VERY small amount of time to try to push a button twice and I think a lot of people are doing it but it's not actually effective.

Still, a ton of people are reporting "mastering" the springboard jump and/or are getting success 80 - 90% long term -- honestly, that just doesn't make sense to me.  I feel like it really should be about 50/50, and maybe slightly better like 60/40 if we feel like a gamer really can press a button on an exact frame more often than not (Graham reports 7/10 success with this in fighting games, which I would think is around the upper end of what anyone could expect long term).  But, unless someone is an unusually good Track and Field player and can consistantly get the button to register once every three frames it seems to me that long term success rates should remain frustratingly low for all other players.

[Graham Wolfe]

Dean, if you watch my submission I am pretty steady on my jumps, I was able to land them as a child pretty consistently with a single press. The thing is if its a constant then it shouldn't be too bad. But here I'm using a double tap. I look at it this way, I know the single press timing well as I've practiced that. I suggest if you can master single press it will definitely help you master the jump as you can double tap to double your chances but knowing the single press makes it a lot more successful. Just make a save state of l4 springboard and just keep jumping over and over until you get the feel of it with 1 press, then toss in double tap or triple tap if you are quick enough.

Keep in mind in street fighter we often are forced to press a 1 frame move to live. I was playing the best zangief player in the world and basically if I missed this 1 frame reversal I lost the match, I landed it EVERY time knowing it was death otherwise. Part of making more success is having your adrenaline running you'll notice your palms get sweaty if you are really focused and this happens. When you are playing like that your reflexes and accuracy does increase. Basically when you get a fight or flight reaction from your body you can do things that are not normally possible. If you can train yourself to get in this mindset when playing you can improve your game, if you don't do this already. If this all sounds crazy you should see some of the stuff I've seen done by my brother and myself. I landed a 1/100,000 chance move, probably first ever done in any tournament setting when I was dead if the move didn't land. It was an impossible move. This stuff sounds like science fiction to some people but it is actually quite normal if you see it happen enough you begin to understand that it is something you can apply as a way to improve performance.

[marinomitch13]

Dean, a few things about button mashing that would affect the expected numbers:

1) The arcade button is bigger and allows for more ability to get more fingers on the button in less time. I don't know what button you have been using on the keyboard, but if it is just the normal 'up' button people use (such as the up arrow, or 'w') I'd imagine it gets a little cramped.

2) People that do the 'pressing' technique of the 'Double Flap' tend to utilize it (and tend to do best) if they start doing the flap a bit ahead of time. This allows some time for the fast-twitch action of the fingers to 'get some momentum' of sorts, and end up doing more mashes per second. Getting the groove going with your fingers is much different than just starting and stoping with one finger -or even two- fingers.

3) Most people that do very well at the super jump (that I've seen/heard of) tend to use the 'scratch'. This is drastically different than the 'double flap' in that it's not as much of a 'pressing' technique, rather, the 'scratch' is more of a 'sliding' technique. People that get very good at the 'scratch'  almost drag their fingers across the button rather than actually pressing in a more downward direction. I think that this allows for the mashes to be executed significantly faster, since it comes down to space, rather than fast twitch muscle reaction. Even if you're able to pull off the double flap on a keyboard, I don't think you'd be able to accurately replicate the 'scratch' on a keyboard button.