Project diva homemade controllers

Yes

 

If I understand right, you use it with PS4, so there, especially how you connect it - there will be no colors. On PC PDAFT - no problems, you can find some pages ago something about how to connect it.

 

Yes, I already connected the panel to my pc.

For the ps4 part, two colors are displayed :

- white only when playing and when ps4 dualshock is correctly paired with the fpga.

- red color if fpga does not communicate with the dualshock.

 

This thread is amazing and I joined the forum because of it! I'm also looking into making a controller. For mine, it's probably going to be in a 15 inch wide enclosure with 60mm cymbal buttons. For the slider, I was looking into this thing linked below, can anyone tell me if they think it will or won't work? It is advertised as being for gaming purposes...

https://www.seeedstudio.com/Grove-C...C4014LQI.html?utm_source=blog&utm_medium=blog

 

Too short. Otherwise okay.

 

Hi everyone that makes a lot of time that I didn't checked the last result about sliders on PS4, I checked previous pages and saw that a lot was done, but without getting where we were now ^^'


Can someone recap actually what are the best solutions in therm of latency, feeling, cost ?
And what is good for PS4, what is good for PC ?

Thank you ^^

 

Please elaborate. I'm really interested.

On PC (AFT) it would be official controller with buttons hooked up via a good PC xinput fightstick board and slider hooked directly to the serial port. 99.99999% authentic experience.

On PS4 it would be (theoretically) official controller+a PS4 controller emulator that emulates touchpad, rumble and IMU (gyro/accel) and runs TGRF algorithm for the slider emulation. TGRF captures all the cases that I could think of that would happen on a "normal" arcade playthrough (including swiping the slider with 5 fingers to get super sanic speed, quite dumb but you can't unsee once you've seen it in action :). The problem is it's theoretical and I'm still too lazy to implement it. Meanwhile ATRF is your best bet since it captures almost everything a normal arcade game play requires (including accurate slider position tracking).

 

Thanks for the explanation, but I was more on the Homemade possibilities, as I'm thinking about upgrading my homemade controller ^^

Also, saw the Hori latest demonstration, they clearly did NOTHING about sliders speed... that's clearly the same as Divaler, I'm so done with this one x)

 

Well it's actually a really deep rabbit hole if you plan to DIY a controller that doesn't suck. I started the project with a custom PCB as a f-you to B***k's monopoly on homemade controller market but now I'm so grateful that I made the irrational decision since I'm now depending my slider emulation algorithm entirely on my custom-rolled board and B***k just doesn't offer the feature I wanted. Same thing goes for the slider board. I ended up rolling my own but with constrainted tools I simply cannot make a module-like thing that just works and contribute it to the community.

So all I can really say is good luck and have fun learning a lot of things that you previously don't think you would ever need :)

PS: I think I should quit being lazy and finish my controller (spaghetti) upgrade (eating). Also I got a new idea on the (mechanical side of the) slider recently and might actually be able to make something module-ish using tools that I have access to. Although who knows if I will sell it or not :D.

 

With your softpot approach, are you able to put anything on top of the sensor, like some really thin plexi? Or does the sensor have to be pressed on directly?

 

I just sticked the SoftPot directly on the acrylic panel. Didn't try putting it under the panel but I think it would require a lot more force to operate if it's under the panel.

Also I'm moving to capacitive as well since it's just way nicer. The sensing area of SoftPot is pretty thin (about only 8mm compared to the official slider's ~4cm)

 

I didn't see it stated earlier in the thread unless I missed it, can you tell me what the serial port settings need to be to use the slider? I'm using PD loader and arcade future tone on PC. It's port 11 right, but what is the baud rate, data bits, parity, stop bits, and flow control?

 

You can leave eveything default (9600,8,N,1 etc) on the PC end, Just the COM11 setting matters. The game handles the rest when using a real slider, so using 9600 on a custom one should work.

 

I think there is everything much difficult, then you expect. And I do not have an ideas how to implement this..

 

@steelpuxnastik Yeah for sure, it is pretty complicated looking from the surface.

Gentlemen, it appears at least several of us are engineers here. If not, this is pretty deep stuff for a hobbyist so I commend you.

Any chance we can get a summarized list of the things necessary to build and implement a touch slider? So far, I've been able to absorb the following:

• If using the original arcade slider, you just need to send the serial data to your computer (via serial to USB converter if necessary) and configure to COM port 11 in device manager. Is that literally it? I thought I read some people talking about interfacing through a microcontroller IC....
• If building your own capacitive touch slider, it's recommended to use flooded x design with both coplanar and resistive interpolation (see http://ww1.microchip.com/downloads/...r-Design-with-PTC_AT11805_ApplicationNote.pdf) because of the long length of the slider. What did you guys do?
• Interfacing your slider to your controller PCB - Ok so the signals coming from the slider will be changing voltages, correct? Then a microcontroller IC with touch signal sensing is needed to interpret those voltages, correct? What are our options for the MCU/IC?
• Now that the slider is being read properly, the data needs to be formatted for your console or PC to understand. That's where the coding comes in right? And not only that, but how to configure the output signals that go directly to your console/PC?
• Any other information in between those high-level blocks I stated above that is necessary to make this work?
• Can I get recommendations on what software to use to design the slider PCB and where I could send the file to have it made into a real thing? Or even how I could create my own at home? I saw something about copper clad R34 as a possibility that could be etched...I have no experience in doing this as of yet.

Yes my brain hurts trying to research all this....Hopefully when all is understood, that it isn't as difficult as I'm assuming.

 

Original controller works literally through RS232 on COM11, but I think it can work through UART-to-USB converter too.

I built my variant of slider for a Cypress CY8CKIT-049 with 5 pieces of PCB and it has, may be, more-less different requirements and advices in building sliders, but I think, there are no strict rules, more important are the conditions under which the slider itself will be used, on basis of which you either use or not certain recomendations.

Difficult question, because, I think, that's not the question you really need to think (only, may be, in some special cases), because that's the problem of MCU to think about it. For example, Cypress MCUs has IDE where you only set pins that will be used for a CanSence (for your touch-panel), some little settings about it's interactions with hardware (PCB, for example) and program.

I used project of DIVA-controller of this japanese guy http://pol.dip.jp/diva/ and he used to set slider movements as analog-sticks movement (because his goal was to use controller with PS4, but I haven't PS4, so I do not know other variants that can be used there) and this variant also can be used with PC too, it's pretty simple. He used 5 data-blocks, that Cypress (slider) send to Arduino Micro (main controller with buttons, it works as gamepad on PC), 1st data-block shows itself movement left-right on slider and one or two hands used and this block are used in control of analog-sticks, and other 4 data-blocks shows what exactly slider sigment are used to be toched at this moment, not really used by adruino, but I found how to use it for some extra-moments I need.

I really think if you don't really know how to make it - you can use some ready solution, http://pol.dip.jp/diva/ for example, because it's really not so hard to complete.

I used KiCad for making slider PCB ( https://github.com/steelpuxnastik/PDArcadeSliderPCB ) and after that I sent it to China PCB factory, 5 pcs for $2$ in result. You can develop your controller design and after that make my changes to my PCB as you need for your design and also to send it to China's factory.

 

I appreciate all the help dude!
I wish I could open that website you linked, but my work blocks it for some reason, so I'll have to read it at home.
I would just buy and use the official controller, but I want to make a smaller controller that uses a slider of around 12"x2" size and 60mm note buttons. Not giving up! I feel like I understand most of this and it's a couple small gaps that are hanging me up.

 

AFAIK that's good for single touch slider (like volume slider) but not multi-touch slider like the one used by FT.

 

Can you elaborate why though? I do wonder myself how the design would detect multitouch when it's just 2-dimensional electrodes. It is similar to what this guy did though: https://tt3d.xyz/2018/07/28/divaslide-revisited-its-happening/

 

The real thing does multitouch by simply sensing the array of "capacitive switches" and see which one is on or off. Interpolation works fine if you got only one point of contact because the controller can estimate where your finger is based on how long it takes to charge up the sensor (i.e. the raw sensor reading). Now imagine what will happen if you put multiple points of contact within one interpolated sensor...

Also for anyone who wants to learn more theory, I recommend this youtube channel:

https://www.youtube.com/channel/UC92R1ulUgUewSuEEnX9n-yg

After watched some of his videos and read bunch of design guides from touch controller manufacturers the concept of capacitive sensing gradually started to make more sense to me.