Suggestions for Reliable Touch Screen Displays?

seigniory asks: "I'm looking to put about 20 touchscreen displays into operation as information kiosks in low-traffic areas (i.e. private waiting rooms, not malls). The application is written, but is mouse-driven, so a conversion to touch screen is necessary. At present, I'm imagining a 12-15" LDC with active-matrix, but will entertain other suggestions. Without much experience with the quality and durability of most touchscreen solutions, or the 'gotcha's' of converting an existing app to touchscreen, I'm at a loss as to where I should begin. I can't start the touchscreen w/o the actual screen, and I don't have a ton of money to play with yet, so I can't afford to waste money trying to find my ideal solution. Searches for 'LCD Touchscreen Display Reviews' has turned up not much more than ads. What experiences or advice can Slashdot offer?"

Comments

Don't go cheap - go for quality, or get headaches!

[Midnight Ryder]

I'll start off by saying that I typically don't recommend overly expensive (because I'm a tightwad), but in this case I'll make an exception.

Allen Bradley uses Elo Touch products for thier industrial control touch screen PC's. These things are wonderful IMHO - I didn't want to see the ones I had leave when the project was done, and they had to go be installed on site - I love mine as a nice 21" desktop montitor and touchscreen combo :-) These critters are a bit hefty price wise, but I haven't seen one fail yet in the field (and these are industrial areas - either gritty dirt or greasy type stuff). They are pretty nice displays, but a little more faded color wise than I'd like to normally use for a desktop application like Photoshop. Incredibly durable, lots of different styles, and they have excellent accuracy. I recommended 'em.

What I DON'T recommend is going with the lowest priced one you find. It will fail. Seriously. Don't be afraid to take your time, and find out what has the longest time between failures -vs- ruggedness (they don't always go together) -vs- the features you need. Anything less than what you need for the application is going to REALLY hurt you later.

As for programming for one - if it's on Windows, no problem. It's just a one-button mouse. They also have Linux, Mac, etc. drivers - it's just sending ya' mouse coordinates and a mouse click. For most applications, it's ready to roll before you even see the touch screen, it's that simple.

Enlarge and Limit

[digitalvengeance]

I recently wrote a traffic stop statistical program for a sheriff's office that used their laptop touchscreens as the primary mode of input. From that short 2-day project, I learned a few points that I'll pass on now.

First, make everything larger than you need it to be. I've found that the size of buttons I need on a touchscreen is only about 1/3rd as large as some of my less adept users. (Of course, they are also using these in laptops in often-moving vehicles.)

Secondly, try to severely limit the items on any given screen. I've found that users tend to be faster visual-scanners when they can just "tap" as they move their eyes, providing too many options on one screen can lead to more false-clicks.

As far as drivers go - just process the input as a one button mouse.