Jump to content

Touch Screen possibility?


Just Phil
 Share

Recommended Posts

Ok i'm back on my x/y touch interface spree!!!

but this is a new idea which i'd like your input on...

So im' building a midibox and have a decent budget to work with.

i am looking to implement 2 or more really nice touch interfaces with visual feedback...

so i saw this display...

http://shop-on-line.tvielectronics.com/product_info.php?pName=touch-screen-lcd-module-7&cName=touch-screen-lcd-modules

that comes with this controller...

http://shop-on-line.tvielectronics.com/product_info.php?pName=tc51854-lcd-controller&cName=lcd-controllers

it says that "5 port pins of ATmega16 microcontroller (PA6, PD2, PD5, PD6, PD7) are available for your personal use"

and it also says at this link...

http://www.tvielectronics.com/TC_Controller_Descriptions.html#Touch_screen_controller

The touch screen controller can respond by sending out X and Y coordinates of touch screen contact location... to computer or microprossessor...

so my question is if i wrote some custom code into the midibox hardware would i be able to interpret that code as MIDI...

if so is there anythign else i'd need to make this work?

If not, would i be able to write a program into their ATmega16 chip (just like tk did for midibox) and just use that to send midi out and jsut stuff it into the same box as the midibox?

my intention is to put 4 of these as an x/y controller and program their (or midibox if neccesary) controller to display the position of the current state of the control.

i'm willing to lean to code these things eventually and any programming languages neccessary.

Link to comment
Share on other sites

so my question is if i wrote some custom code into the midibox hardware would i be able to interpret that code as MIDI...

Yes, but you could easily just interface the LCD module and touchscreen to the MBHP core and not use their custom controller.

If not, would i be able to write a program into their ATmega16 chip (just like tk did for midibox) and just use that to send midi out and jsut stuff it into the same box as the midibox?

Possibly

i'm willing to lean to code these things eventually and any programming languages neccessary.

You will probably be able to figure this out yourself when you learn the languages.... In fact it could be good practical experience, as a way to learn :)

Link to comment
Share on other sites

PIC has ample memory for either/both of those, just not in addition to the MB64 app, which is feature-packed, and therefore uses most of the RAM already. Anyway, that module wouldn't need the PS2 driver, it's analog resistive so just wire it in to the AINs

If the screen has a different resistance you will need to work around it.

Link to comment
Share on other sites

http://www.elotouch.com/Technologies/compare_all.asp

the apr model seems to be perfect for small size music applications.

using a piezo sensor under the screen, velocity sensing could be achieved...

i made this experiment wit my touch screen: http://www.kvraudio.com/forum/viewtopic.php?t=192282&start=345

take a look at the midipads plugin from insertpizhere for inspiration!

Link to comment
Share on other sites

Hi Phil,

I'd say that it's very unlikely that a touchscreen is resistance based at all, because resistancy usually requires some pressure that is measured, which would mean that you cannot look through the pad. It is most likely that these touchscreens use a capacity based input grid and these are not the easiest to interface; you need to find specialized ICs to convert the capacity to a readable voltage.

If I am allowed to give you an advice: I see you're tending to make it quite complicated, instead of trying with one touchpad, you want to connect 16 pads by PS2 and then you increase the level of difficulty even further by adding not one but two (!!) graphical LCD modules - and an even more complicated touch recognition mechanism.

See, overdoing's a typical thing from people being not that experienced (you will find plenty of posts like "I want to build a midibox with 512 motorized faders" in this forum, but to my knowledge none of these ever got built - whereas people starting their projects with four or eight faders usually also build their boxes successfully) ;)

If you do not have that massive knowledge in electronics/programming: get yourself a realistic task you can manage. It does not need to be silly or unworthy, but just a little bit more modest ;-)

e.g. try with one resistance based touchpad: see if you can find one (and I mean one, not fifty) that you can hook up straight to your MB64, make some tests and you will probably manage this without programming skills at all. Maybe you bought one or two wrong pads, but if you can find cheap used ones that shouldn't be the problem. After that you can try to add a second pad without interfering the signals with the first pad and increase the level of difficulty step by step. If you worked a bit with various touchpads, I'm sure you'll get to the point where you are able to develop a PS2 driver or something crazy like that :)

Keep it up!

Best,

Michael

btw: there might be (expensive) touchscreens already interfaced with a IIC protocol, but again, this is advanced stuff, and I do not believe someone without coding skills is able to develop an IIC driver in an ASM based application. Sorry, don't want to discourage you, but one has to be a bit realistic, too ;)

Ps: I agree to polosid; touchscreens are probably more useful on a desktop PC rather than on a microcontroller.

Link to comment
Share on other sites

I've been playing with touch films for a while now - they use them in industrial controllers.

Rather than use a whole screen, I had another idea recently.

Put the film on a translucent backing, and fit it over a bigger modulation matrix for a MIDIbox SID CS. That way I would have the equivalent of one button for each LED on the modulation matrix.  I just have this vision of a sort of 'Star Trek' all touch panel, eventually. Ther is a sort of half way option that uses a very basic screen driver to colour the spots on the matrix, perhaps to show what has been updated.

Just a thought. The worst parts of coding seem to be the noise elimination and the 'quantisation' for the zones. I need to dig into my archive, I do have some code, (in FORTH), for, resistive, HP optical, and I think one which was accoustic. THe HP was the easiest, because the low level drivers were built in to the old HP150, and you could simply use it by escape commands to the terminal that was built in. As an aside, it also did Tektronix 4010 graphics too, by issuing the correct sequences from the keyboard, you could use one as an "Etch-a-Sketch". On the HP150 mk2, the touch screen was already an optional extra. I must dig out the system I kept and see if I can drag it back to  life. Wow, the joys of DOS 3 on a green screen with  a wonder HP-IB, (IEE 488), bus to connect the drives.....

I have got the Microvitec 'Touch Tech' soemwher too, they were optical for BBC micros, and fitted onto the 'Cub' series of monitors. They're output was serial, showing that you don't need much speed for something like that. They had a on-board controller, which might be worth looking at. Again thewy were the 'grid of LED's and sensors' style.

I suppose you've all seen the Lemur.

http://www.jazzmutant.com/lemur_overview.php

As you might have guessed, unusual control systems are a pet interest of mine - I did some work for a charity for the disabled years ago, and worked with breath/mouth controllers, lightweight joysticks, a few other things. Eye movement was still impractical then, though I believe working controllers are available now.

Link to comment
Share on other sites

Rather than use a whole screen, I had another idea recently.

Put the film on a translucent backing, and fit it over a bigger modulation matrix for a MIDIbox SID CS. That way I would have the equivalent of one button for each LED on the modulation matrix.  I just have this vision of a sort of 'Star Trek' all touch panel, eventually. Ther is a sort of half way option that uses a very basic screen driver to colour the spots on the matrix, perhaps to show what has been updated.

http://it.youtube.com/watch?v=KyP4BUO_GBE    :o

Link to comment
Share on other sites

Hi Phil,

I'd say that it's very unlikely that a touchscreen is resistance based at all, (most affordable ones ARE resistance based) but it's being really difficult finding ones with 10k ohms resistance because resistancy usually requires some pressure that is measured, which would mean that you cannot look through the pad. It is most likely that these touchscreens use a capacity based input grid and these are not the easiest to interface; you need to find specialized ICs to convert the capacity to a readable voltage.

If I am allowed to give you an advice: I see you're tending to make it quite complicated, instead of trying with one touchpad, you want to connect 16 pads by PS2 and then you increase the level of difficulty even further by adding not one but two (!!) graphical LCD modules - and an even more complicated touch recognition mechanism.

See, overdoing's a typical thing from people being not that experienced (you will find plenty of posts like "I want to build a midibox with 512 motorized faders" in this forum, but to my knowledge none of these ever got built - whereas people starting their projects with four or eight faders usually also build their boxes successfully) ;)

If you do not have that massive knowledge in electronics/programming: get yourself a realistic task you can manage. It does not need to be silly or unworthy, but just a little bit more modest ;-)

e.g. try with one resistance based touchpad: see if you can find one (and I mean one, not fifty) that you can hook up straight to your MB64, make some tests and you will probably manage this without programming skills at all. Maybe you bought one or two wrong pads, but if you can find cheap used ones that shouldn't be the problem. After that you can try to add a second pad without interfering the signals with the first pad and increase the level of difficulty step by step. If you worked a bit with various touchpads, I'm sure you'll get to the point where you are able to develop a PS2 driver or something crazy like that :)

Keep it up!

Best,

Michael

btw: there might be (expensive) touchscreens already interfaced with a IIC protocol, but again, this is advanced stuff, and I do not believe someone without coding skills is able to develop an IIC driver in an ASM based application. Sorry, don't want to discourage you, but one has to be a bit realistic, too ;)

Ps: I agree to polosid; touchscreens are probably more useful on a desktop PC rather than on a microcontroller.

Audiocommander, first of all my box needs approximately 16 x/y interfaces (between 12-16 budget will dictate), where i'm really staying away from joysticks for because of the the functions they will be controlling.

i also need visual feedback of where the pointer is at any point...

i will stop at NOTHIGN to make it happen, if necessary i will HIRE a programmer to make it work. But i'd prefer to learn to code/program, (no more then 2 languages or i may go insane)

i dont care if the x/y control is PS/2 or anythign.. all i need to know is.. what touch x/y touch sensor (no velocity) do i need to hook up and make work with midibox hardware.. OR any other hardware for that matter...

IF i have to i will buy 12-16 KORG kaoss pads and be done with it and mount them to a panel..

i hope that you understand where i'm going with this.... this isn't just a fun project i have people depending on me to make this work!!!!!

Link to comment
Share on other sites

regarding the idea of 16 xy pads... a vst version could be tryed.

using the insetpizhere's midipads vst plugin ypu have a RESIZABLE plugin with amny configurations

1,4,16,64 pads presets that send xy cc's and notes

every pad can load any type of image, so visual feedback is ok

4,16 sliders configuration, so you can use it as a controller for digital mixers...

it has an option to take the velocity value and the trigger from an external midi signal (a piezo signal converted to midi)

and preset inspired to the c-thu-music axis controller!

latency: it depends on the touchscreen model. the software latency is 1ms.

if you don't have budget limits you could buy another pc... an old model is sufficent to handle touchscreens.

this solution is very very usefull, the plugin is freeware...

the problems regarding this configuration are:

* monotouch. actually the multitouchscreens are not supported by any os, and there are no kits or monitors avaible.

* latency: i have a resistive (not elo)  type and the latency is a little high. 35ms... i use midipads without any problem as a midimixer, xy controller and to control many programs... but if yu want to play complex drum solos it's not the best. i had good results using a piezo trigger on the border of the screen but the piezo signal arrives before the mouse pointer refresh , so you cannot play 16 pads at once with this precision.

latency is a big problem if you use the original ps\2 drivrs

the only touch technology that can be small sized and resistent enought to musical applicatins is the acoustic pulse recognition. if i'm correct the latency for this model is around 10ms... but the minimum for musical apps is 6ms!!

i'm very intereasted in making a big controller with a lot of xy pads with visual feedback, but i'm starting from visuals, then velocity sensing and ar last (if i'm alive) on xy touch screen or every pad

Link to comment
Share on other sites

the only touch technology that can be small sized and resistent enought to musical applicatins is the acoustic pulse recognition. if i'm correct the latency for this model is around 10ms... but the minimum for musical apps is 6ms!!

I had a chat with a technician from elo, and he stated that the Carrol touch (led-photoreceptors) technology will be faster. I have tried 3technologies from them briefly, now trying to get them to lend me some touchscreens, so i can have a look a them a bit closer. It seems to me that it will be possible to get them respond "immediately", since intervals, the controller reports to the computer, are very customizable. Either from 0 to 3000ms in 10 ms steps, where 0 and 10ms setting can flaug the computer, or you can send the controller a message from the driver to respond with currently touched coordinates, or it can immediately respond when a change occurs.

No affordable-responsive-working in daylight multitouch technology is available at present.

I am sorry for this slight offtopic. I too wanted, to go with a polytouchable controller, but I decided to give priority to visual feedback. I am still considering Led rings or RGB leds to go around this, in which case I would use samples from spectrasymbol

Link to comment
Share on other sites

I decided to give priority to visual feedback. I am still considering Led rings or RGB leds

Yeh I was thinking about the spectra touchpads with LED bars would be the cheap and easy way...

x  ,----------,

x |              |

x |              |

x |              |

x |              |

x  '-----------'

  x x x x x x x x

Link to comment
Share on other sites

Yeh I was thinking about the spectra touchpads with LED bars would be the cheap and easy way..

I am not sure about that, spectrasymbols prizes arent very keen for one off production, and the prizes I received were insane. It could be I have choosen some not standard parameters, but I did tell them to suggest any changes to reduce the cost.

Also,according to the uk distributor, the xy pads arent very reliable, I have one from them, still waiting to be tried..

Link to comment
Share on other sites

Maybe you guys can explain to us all what the technical limitation was, with the LED array sensors (as per the youtube video above) ... I remember there was one... and I remember thinking they must not be doing it the most optimal way ;) But I can't remember what the problem was exactly...

Anyway I couldn't help but wonder if it would be possible to make something like the 16F88 do that, and send the results by IIC.... I thought about it before but just don't have the time to develop it all by myself.

Edit: polo what you're talking about is visual feedback for every segment of the entire matrix, which is a different thing again... Anyway if you feel like doing my homework for me on the LED sensors...

Link to comment
Share on other sites

Anyway if you feel like doing my homework for me on the LED sensors...

sorry , i don't understand, what are you saying to me?

anyway..

i think that the limitations of a ledmatrix multitouch sensor are:

*it work only in dark ambients

*you cannot print sprites\images while you use the matrix as a xy pad

*bad resolution

but maybe i'm wrong

Link to comment
Share on other sites

sorry , i don't understand, what are you saying to me?

...

but maybe i'm wrong

Those things are all partially true... Regarding the inability to use them in bright lighting, I wondered if anyone had tried this with UV or IR leds... this would also allow you to embed a second array with visible light.

I think the limitation may be resolution or something like it (larger led arrays?)... But yeh the homework i referred to, was that there was a specific technical limitation (not a practical limitation like the ones you mentioned) that the guys who developed the led sensors encountered, which meant the controller needed more power or ram or something.. I dunno, that's why I need you to do the homework ;)

Link to comment
Share on other sites

Those things are all partially true... Regarding the inability to use them in bright lighting, I wondered if anyone had tried this with UV or IR leds... this would also allow you to embed a second array with visible light.

a second array of uv leds and a larger matrix (16x16) could solve these practical problems...

do integrated leds with uv and visible light exist?

eh eh .... it gets too complicated!  :o ;D

i'll concentrate my work on visual feedback using led matrix in the next months. probably i'll choose ARDUINO boards, because those have a friandly usb handling and i want to build a pc based app. i don't see usb compatibility so easy inside mios

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
 Share

×
×
  • Create New...