3ds analog>psp>else?

[ShockSlayer]

Yes. Range hasn't been an issue for me so far, I used ssbm to test, the char select screen is where I notice it, I haven't tried the program yet or WW, those are on my list. I should be getting everything together soon, might do a livestream just so other people can be there.

SS

 

[SonyQrio]

The 3DS analog stick is BIG TIME design flawed from a impedance standpoint. Long story short, there is no way you are going to be able to hook this directly up to a GC or an N64. Any combination of resistors can combat this issue, but will NEVER give you full range.

As some of you might have read, we have combated this problem with an Arduino. i.e. You feed the X+ and Y+ from the analog stick into the microcontroller and you get flawless X+ and Y+ out of the microcontroller. We were going to design a board that will take care of all the 3DS analog stick problems, and it will be about the size of the joystick itself.

Furthermore, there will be an FFC socket that you can plug the analog stick directly into, and all you will have to solder to is the nicely sized test pads for +5v, GND, X+ out, and Y+ out. There will be models for the N64, i.e support for the analog stick, and models for the GC, i.e. support fot BOTH the analog stick and C stick.

Do you still think it's a dumb idea, guys?

 

[Zero]

Yes.

 

[Mako321]

 

[ShockSlayer]

First, you copied what you said originally but changed GC joystick to 3DS joystick. You've changed your original still unproven theory.

Second, there is no "BIG TIME" design flaw. It wasn't intended to work with GC controllers. It is a 3DS stick, it is designed to work on the 3DS, and honestly nothing else. Same with the PSP stick. Sure, it's similar. But touch screens are similar too. We can get them to sort of work, yes, but as you can see, it doesn't work fully.

The same applies to all swapped joysticks on all controllers. Normally that's a mod only seen on n64s, and it usually works out fine due to the n64 joystick era. An era full of broken joysticks. An era where programmers used bigger deadzones. Meaning? Joystick swapping allowed for more error in values. That said, actual joysticks are waaay more balanced than nubs, so...that has a lot to do with it too. That's where range becomes more of an issue than deadzone.

Third, resistors have nothing to with the range, they never did, although I think they can. Again, things I'm testing. They had to do with balance in my application.

Fourth, the microcontroller thing could work, but why when we can change the range and the deadzone through easier methods. Go ahead and make a bunch of boards, populate them and sell them. If they work great, great. Do that. I don't care. I cant stop you.
---

Let's talk about the problems we are facing, and a bit about what I learned last night.

The 3DS nub doesn't have the full range of the stock GC stick. As ashen said, he was getting like 111's with the GC stick and 90's on the 3ds stick. That's pretty Dang close, and I seemed to get good results when I played my games, so I'm thinking range is probably good, a little improvement probably wouldn't change much, but it'd be nice.

The 3DS nub doesn't have equal resistance on each axis. This isn't too much of a problem, because of how the controller works, and how it marks the "centered" voltages. Tests last night with Herma confirmed that even though each center voltage on each axis was different, pushing it along the axis each way(ie, fully up and fully down) still produced even voltages at the ends of each axis. The problem? The centered voltages change. This is where deadzone comes in.

Let's say you boot up your GC, and the controller recognizes that the 3DS stick's center voltages are 1.5v for the Y and 1.7v for the X. Not a problem, right? Okay, we move the joystick, doesn't matter where or how(well it kinda does). It returns to the center. Y is now 1.7v and X is now 1.5v. The controller now thinks, based on the center points it registered at bootup, that the joystick is being moved. EVER SO SLIGHT but still moved. This. This is crawling/ghost movement. And it's not a flaw in the joystick, because it wasn't designed to be a joystick. It was designed to be a slide pad. And it was designed to be used in a 3DS, which takes this into account. A GC controller doesn't.

How do we solve this? I'm testing out tons of theories, but so far I'm pretty sure it /should/ be achievable through a combination of resistors and the right voltage. I've got a ton of numbers to work with. Essentially, the goal is making the 3ds stick more like the original GC joystick, which is what the controller and games are programmed for.

At the same time, taking a page from the N64's book is totally a viable option, using a 3rd party controller can potentially produce completely different results. Essentially, my theory is that there are two deadzones. There is one coded into the controller, and there is one on each game. The objective is to get the joystick compatible with the controller, thus compatible with every game.

SS

 

[Zero]

I mean no offense intended Qrio, but with your rep I really doubt people would throw money at you for this, especially without you showing a prototype board or similar. Plus, even if we assume you're right about the range not being fixable any other way, Ashen already said that in terms of range 3DS sticks are close enough for it not to be a big issue. And if the voltage thing works, then your entire argument is thrown out.

 

[S5114]

I say make one prototype. If it works, is convenient, and solves the problem easily, then people who don't want to do what SS will do might buy it. However, it would probably only sell to a few people depending on how hard SS' fix will be. So all in all, make one if you want, keep it to yourself, and market it after SS is done with his research. I don't care what your rep is, if Hitler says 2+2=4, it's still true, and if whatever you're doing works, I may or may not care.

 

[ShockSlayer]

Despite my best efforts, I couldn't get it working, and I pretty much gave up once I realized I could probably get away with using these other joysticks that I had, which are the same height as the nub(about a 1/4 the length and width though), but actual joysticks and full range.

The principal is as follows: The higher the voltage, the higher the range, the smaller the deadzone. The lower the voltage, the lower the range, but the higher the deadzone. I could eliminate all crawling by using a lower voltage, at the cost of range. And I could get more range and more crawling, too! Using a lower voltage(say, 3.1v) might be a good idea for the 3ds nub, but I didn't specifically test that honestly, I was still getting a degree of crawling using 1.7v, the controller's deadzone is just too small and the 3ds nub just doesn't recenter properly.

I agree with what herma told me, if you require an extra circuit in order to get the joystick working properly, it's time to start looking for a better joystick. Qrio's on the right track, some sort of circuit would be what's needed to get full range and eliminate crawling, I'm just assuming the circuit would allow you to set a deadzone.

Like I said, I'm using some of these. They are ungodly tiny, and I love them. I don't know what I am going to do for caps yet. I'll see if I can find some.

SS

 

[Ashen]

I'm going to play with the 3DS sticks a bit more tonight. Honestly I didn't notice any crawling when I tested them before, but I didn't test SSB either. I did test using the pad tester and about 10 other games though. I'll let you know how I make out.

 

[S5114]

Are there any caps that fit it?
EDIT: That's like $20 including shipping for one! I'll just see if I can fit a gc stick in...

 

[SonyQrio]

Qrio's on the right track, some sort of circuit would be what's needed to get full range and eliminate crawling, I'm just assuming the circuit would allow you to set a deadzone.

MM and I already did all the testing, SS. We already came to this conclusion. We did do something.

 

[mymixed]

Ok, i'll make a board .

 

[S5114]

Now that SS' fix didn't seem to work, if you guys make a working product, I'll probably buy it. Played around with the 3DS today and got a feel for it's stick.
EDIT: No, dude, TOTALLY not what I meant....

 

[ShockSlayer]

I never had a fix, I had tests to run and theories to confirm. I came to the conclusion that I have better joysticks. Thanks, Mario.

Honestly the ghost movement isn't really a problem unless you are really anal about the joystick, and seeing as how the XGC is literally me going full anal ( ) about a project, I can't take it, it'll nag at me in the back of my head that there is a tiny imperfection. Seriously, I literally had to trigger ghost movement. Its a little more visible in the controller testing app, but it's not that significant numerically speaking, and it can be slightly reduced with a lower voltage, again at a small range cost. I ought to test at 3.1v to see what range I get.

I'll let you know how I make out.

Ooh, kinky.

SS

 

[SonyQrio]

Well, there's just no way that anybody will be able to fix this without an IC. As I have been saying, the 3DS stick is design flawed, meaning that when you push the stick all the way down, you get -111, and when you push the stick up, you get +60. It's not some funny glitch, it's how the 3DS stick is made. Instead of fixing the hardware, Nintendo just fixed the problem programmatically. End of story, period.

Oh well, maybe MM and I will make that IC after all.

 

[ShockSlayer]

Well, there's just no way that anybody will be able to fix this without an IC.

That's true. But only if you sincerely feel the need to fix it. Aside from that it's perfectly fine.

As I have been saying, the 3DS stick is design flawed,

No it is not. I already said that. Stop arguing with fact.

meaning that when you push the stick all the way down, you get -111,

No, you get about -85 to -90

and when you push the stick up, you get +60.

No, you get about 85 to 90

It's not some funny glitch, it's how the 3DS stick is made.

I already explained exactly what the problem was and why it is.

Instead of fixing the hardware, Nintendo just fixed the problem programmatically.

ITS NOT BROKEN GOD Dang. Nintendo/Developers set a higher deadzone because that's how nubs *Can'tSayThisOnTV*ing work.

End of story, period.

SHUT THE *Can'tSayThisOnTV* UP YOU DO NOT KNOW WHAT YOU ARE TALKING ABOUT

Oh well, maybe MM and I will make that IC after all.

Good for you, until you do that you get nothing!

I don't want to have this exchange again.

SS

 

[Hermaphroditus]

Oh well, maybe MM and I will make that IC after all.

No offense, but I'm not sure you could. At least with your current knowledge of circuitry and with no help.

I mean, I'm not trying to stray you away from trying, it'd be a good learning experience, and with an adequate amount of resources I'm sure you could figure something out.

And you likely wouldn't be making an IC unless you are programming a reprogrammable IC, in that case you're not really making the IC, just putting a purpose to it. You may make a circuit out of other ICs and various other components, but that's a technicality.

Personally I feel the issue of deadzone versus range really shouldn't be messed with for the 3DS stick. If you're talking about saving space you would be adding even more space to add a correction circuit in there, and in this case you may as well just use a larger analog stick.

For the 3DS nub follow this flow chart:

Deadzone bother you? -> Yes -> Can you sacrifice range? -> Yes -> Lower the voltage.
No                                                      No
v                                                        v
              Leave flax alone.

 

[S5114]

Haha, you know how much we've been discussing this and how much it's helped me? Well I also posed this question on MBB and you know what? It took a few days to respond and all I got was a link from Bacteria that lead to Zenloc's gc analog to digital slider diagram. Showing that he clearly didn't read the whole post and he hasn't responded back at all. I mean seriously, I even put it in the n64 section

 

[LOCtronicz]

I want to add that I have used the 3ds sliders with 4 different controllers and on each controller I got different results including mirrored axis's.

However with the original 1000 psp analogs I have different results on a 3rd party analog controller. I don't get full range but it does work in every game I've tried. No crawling and no range problems. I didn't even need to power it with a higher voltage.

Which games gave you range problems? and what exactly didn't work?

 

[S5114]

This post is a little late but I thought I should say. I did own a mini wireless ps2 controller that used sliders. It was small, red, and gamestop brand so it was obviously flax. I took it apart and it looked like it was a psp slider. If I find it anywhere, I'll say so.