Functional Diagrams of 4/8/49-way Sticks

[NoOne=NBA=]

I wanted to post pictures of how the different sticks function because some people are still questioning why 8-ways don't work well in 4-way games.

Here is a functional diagram of the signals sent by the different sticks in a 4-way game.





----  4-way Sticks ----

Notice that a properly adjusted 4-way stick, with a 4-way actuator will NEVER send a bad signal to the game at any time while it is held to the edge of the restrictor.
Likewise, the 49-way in 4-way mode will not send bad signals.

The dead spots on both controllers are small enough that they are passed through quickly, and produce no noticable lag while travelling through them.

The 4-way base with the 8-way actuator is actually capable of sending 8-way signals, but only in a small portion of the diagonals.
This is why people have complained about the performance of the Chinese "Wico-STYLE" sticks.
They have the 4-way base, but use the 8-way actuator, and DO hit diagonals in a 4-way game; but do not have large enough diagonals to function properly in an 8-way game.



----  How Diagonals Are Handled In 4-way Games----

The problem with the diagonals is that they are an "unknown" signal to a 4-way game.
How the game handles this information varies from "guessing" which direction you wanted to go (as is the case with Pacman), to just flat stopping until it gets a signal it can process (as is the case with Donkey Kong).


Pacman

If the dead spaces are small enough, you may not notice them in slower levels of a game because you will be into the right area by the time the game needs to make a decision on which way to go at the next intersection.

This is not the case on higher levels though.
If you need to do a quick right/up/right on Pacman for example, the game may read the first right, and the up, but miss the second right because you hadn't gotten out of the up-right zone yet, and it "guessed" that you wanted to keep going up.
These deadspots are a coinflip each time, and you can't afford to lose ANY of them--if you want to be successful.


Donkey Kong

Donkey Kong is the ultimate test to see why an 8-way doesn't work in a 4-way game.

Mario should NEVER stop moving while the stick is away from center--unless you push sideways while on a ladder, or up/down while on a girder.

To test this, rotate slowly from Right to Up, and then back through Right to Down, while Mario is running on the first girder.
That will show you exactly where the deadspots on your particular stick start.
For the up/down directions, you will need to be on a ladder to test them, but the procedure is the same.



----  8-way Sticks ----

The reason that 8-way sticks do not work well on 4-way games is that the size of the dead spots are exaggerated by the way the sticks are designed to work.

Here is a functional diagram of the different 8-way sticks.




Round 8-ways

If you look at the diagram for the round 8-way with 8-way actuator, the diagonals are designed so that the arcs that send the diagonal signals are approximately the same size as the arcs that send the cardinal signals.
In a game such as Robotron, this will result in approximately the same number of shots in each direction on a sweeping movement.

Notice how much larger the deadspace is on the inside of the 8-way base with 4-way actuator.
That is because the 4-way actuator is smaller to prevent the 4-way sticks from hitting the diagonals in a 4-way base.
One result of this is that the 4-way actuator has to travel farther to hit the switches than the 8-way actuator.
The other result is that the 4-way actuator will have larger cardinal areas, and smaller diagonals, than the same stick with an 8-way actuator.

Also notice that there is virtually NO difference between the 49-way in 8-way mode, and an actual round-restricted 8-way.
There may be some difference in the size of the deadspot in the middle, and the actual size of travel however--depending on which round 8-way you are comparing it to.
Functionally though, both sticks have eight almost even arcs defining their directional signals, causing their operation to be identical.


Square 8-ways

The amount of diagonal signal sent by the square 8-ways is even greater than that of the round 8-ways when measured as distance of stick travel around the perimeter.
This is because the stick is allowed to travel a little farther into the corner.

This makes them good for games that require quick, and precise, hits to the corner (some fighters); but makes them harder to control than round 8-ways in games that require large sweeping movements (Robotron, Time Pilot, other fighters).

[FractalWalk]

Excellent descriptions!

Looking at the 8-way pictures, it should also become apparent why making a DIY 4-way physical restrictor would be difficult to do. The restrictor must "just miss" the dead zones, which takes extreme precision.

[Tiger-Heli]

Great write-up!!!

[daywane]

Excellent descriptions!

Looking at the 8-way pictures, it should also become apparent why making a DIY 4-way physical restrictor would be difficult to do. The restrictor must "just miss" the dead zones, which takes extreme precision.

well.... wish me luck. I am trying just this on a tron stick and a xyenaphobe.
well, I am having a machinest do it... so I see you are right it is not a DIY.

[NoOne=NBA=]

It shouldn't be too hard for you.
You have a Tron one to use as a baseline, and there is some room for error on a Tron stick.

The reason for this is that a properly adjusted Tron stick will function like the diagram of the 4-way base with 8-way actuator.
It needs to be ABLE to hit the diagonals, for the Spider/Cone stages.
But it must not be PRONE to hitting them (like a true 8-way) to avoid the zig-zags on the Light Cycle stage.
As long as the restrictor gets you out of the "bad" zone quickly, it should function properly.

[RandyT]

The reason for this is that a properly adjusted Tron stick will function like the diagram of the 4-way base with 8-way actuator.
It needs to be ABLE to hit the diagonals, for the Spider/Cone stages.
But it must not be PRONE to hitting them (like a true 8-way) to avoid the zig-zags on the Light Cycle stage.
As long as the restrictor gets you out of the "bad" zone quickly, it should function properly.

I guess this is as good a time as any to bring this up....

The GP-Wiz49 has an intentional very small area in each corner of the 4-way DRS mode for diagonal operation in just such circumstances.  You never hit them in 4-way mode because of their extreme positions, but when you want to for reasons stated above, you can and it works quite well.

FWIW.

RandyT

[Tiger-Heli]

I guess this is as good a time as any to bring this up....

The GP-Wiz49 has an intentional very small area in each corner of the 4-way DRS mode for diagonal operation in just such circumstances.

[RandyT]

Interesting.  Will this cause any problems if you are playing a 4-way game and moving the stick around the perimeter?

I would guess pacman wouldn't care, and you probably wouldn't use that motion in Donkey Kong, but just curious?

Absolutely not.  Due to the fact that the extreme corners are pretty hard to hit on a 49-way to begin with, you never know they are there...unless you need them and know where to find them.

Also, the shortest distance between two points is a straight line, and moving a stick from absolute "RIGHT" to absolute "UP" through the rounded corners would be a very difficult and unnatural way to play a 4-way game.  Even so, the areas aren't "dead" so it's pretty much impossible for them to create a problem in the unlikely event that they were hit.

RandyT

BTW, kudos to NoOne=NBA= for the diagrams.  Very well done!

[SirPoonga]

No, think about a real 4-way joystick.  There is a point when you are going in between say, Up and Left, that no inputs are pushed.  Do deadzones in the corners is very realistic.

And your center deadzone, to be correct, isn't going to be square if you have a round accuator is going to be round.  That's the way it works geometrically.  But that's nitpicking.  The whole idea you are making is still the same.

[RandyT]

And your center deadzone, to be correct, isn't going to be square if you have a round accuator is going to be round.  That's the way it works geometrically.  But that's nitpicking.  The whole idea you are making is still the same.

Hmmm...I don't think it would be round, rather somewhere in the middle.  The larger the dead zone, the more square it becomes, obviously with corners that have the same radius as the accuator.

But it would be round if the switches were in contact with the accuator at the center point, i.e. null deadzone.

RandyT

[SirPoonga]

Yeah, it must be the way I have my actuators on my supers setup.

[RandyT]

But my main point was that on a real 4way there is a small deadzone between direction changes.

This is true, but it should probably also be pointed out that a deadzone is an undesireable trait of a 4 way.

The perfect scenario is a case where absolutely no deadzone exists, and also no overlap of switch actuation occurs.  Like infinity, this  condition can exist only as a concept on 4-way sticks relying on mechanical switching principles.

However, the 49-way in 4-way DRS mode can, and does,  actually achieve that goal because it can be either-or with nothing in between.


RandyT

[NoOne=NBA=]

You can tweak leafswitch 4-ways to get that condition.
It requires some patience, and needs to be checked/readjusted to KEEP them there though--unlike the 49-way, which will stay there, once set.

---------------------------------------------------------------------------

I figured my dead zone charts above based on the center of the sticks, rather than the outside of the actuator, so they should be a theoretical square.

If anything, they would require slightly MORE movement from center as you get away from the center of the switch, causing the ACTUAL pattern to push OUT slightly in the corners, and closer resemble a 4-pointed star.
The shape of the actuator has little to do with the travel necessary to trigger the switches, so the inside deadzone should be the same for both the round and square actuators.

The only time where this would not be true is if you used a Competition, or similar stick, that has no lever on the switch, with a round actuator.
I did that on one of my Tron hacks to see if I could get the "almost 4-way" action out of it.
The dead zone on that stick is star shaped, as described above.
It functions as a 4-way, except at the extreme diagonals, where it will contact two switches at once.