Reporting Style Question

GTT TFT Support

Moderator: Mods

Post Reply
mpetty
LCD Geek
Posts: 25
Joined: Mon Jul 26, 2010 10:31 am

Reporting Style Question

Post by mpetty »

Sorry for all the posts, I'm hitting all types of new things as I try to port the screens that I'm drawing in GDI/Windows across to the GTT displays.

For change reporting style, the manual states that:

Code: Select all

0 - no event reporting
1 - down region events only
2 - Up region events only
3 - Down and up region events
9 - Down coordinate events only
10 - Up coordinate events only
11 - Down and up coordinate events
All other values reserved

What happens when I have a screen where I want coordinate AND region events to be sent to me?

I have a series of screens that the user goes between by swiping left or right. I have some buttons/touch regions on some of those screens. If I want to be able to register the touch in that region I can't register the swipe. If I register the coordinate events, I can't tell what button I have pressed.

My screens are identical at the bitmap level between the GTT and my displays (in fact, I'm positioning all of my elements on the GTT by referencing my GDI elements, so if I move something around in Visual Studio Designer, it shows up moved around on the GTT). So, I guess that I could just set up to register coordinate events, and then either fake the click at that coordinate on my application, or bounds test the coordinate against the buttons. So, I guess that it's not a huge issue for me, but is a much bigger pain than it should be.

If I hadn't decided to draw everything pixel for pixel in GDI as I was wanting it drawn on the GTT, I'd be in serious trouble, and I only did that because I can prototype a lot faster on GDI by working with the visual designer.

My solution would be to add the X and Y coordinates to the touch region report that's coming back. I would prefer that over a second reporting style that reported both coordinate and region events back, because without the coordinate given for the region, the user could only swipe the screens left or right on areas not covered by buttons, which would kind of be a pain.

Yea, this pretty much knocks me dead in the water. Some of my displays make use of the entire screen and I don't have room for a button to move the screens left or right, so I was doing it by swipe. But since that's not really possible, I can't navigate anywhere.

Clark
Matrix Orbital
Matrix Orbital
Posts: 881
Joined: Fri Aug 17, 2007 10:58 am
Location: Matrix Orbital
Contact:

Post by Clark »

No worries Mike, good to see you're giving the GTT quite the test drive.

Like previous displays, the GTT will only report in one of two basic modes: region or coordinate. I believe you've hit on the solution to your problem in running in coordinate mode and faking buttons. I did a simple drawing app for the GLT240128 when it came out and found remembering certain coordinate regions as buttons within my own code was the easiest way to display buttons and read coordinate events on the same screen.

I'll inquire as to the inner workings of our code, but unfortunately, I don't believe it is possible to return both a region and a coordinate. If I'm wrong I'll be sure to create a formal feature request.

We're working with a relatively new medium here in the touch screen, and as you can see there isn't a direct correlation between it and older passive screen technology so it's difficult to predict what functionality will be needed. We've started with the most basic mode in coordinate reporting and have begun to build upon it with more advanced buttons. As we receive feedback from designers we can begin to see common patterns of use and think about building in more advanced features. Perhaps direction reporting in region mode may enter into the conversation, it will really depend on the standards we see emerging from the feedback we receive; we're always listening and constantly designing.

~Troy
Troy Clark
Design & Development
Matrix Orbital

mpetty
LCD Geek
Posts: 25
Joined: Mon Jul 26, 2010 10:31 am

Post by mpetty »

Clark wrote:No worries Mike, good to see you're giving the GTT quite the test drive.

Like previous displays, the GTT will only report in one of two basic modes: region or coordinate. I believe you've hit on the solution to your problem in running in coordinate mode and faking buttons. I did a simple drawing app for the GLT240128 when it came out and found remembering certain coordinate regions as buttons within my own code was the easiest way to display buttons and read coordinate events on the same screen.

I'll inquire as to the inner workings of our code, but unfortunately, I don't believe it is possible to return both a region and a coordinate. If I'm wrong I'll be sure to create a formal feature request.

We're working with a relatively new medium here in the touch screen, and as you can see there isn't a direct correlation between it and older passive screen technology so it's difficult to predict what functionality will be needed. We've started with the most basic mode in coordinate reporting and have begun to build upon it with more advanced buttons. As we receive feedback from designers we can begin to see common patterns of use and think about building in more advanced features. Perhaps direction reporting in region mode may enter into the conversation, it will really depend on the standards we see emerging from the feedback we receive; we're always listening and constantly designing.

~Troy
Yea, that's typically how buttons work in most frameworks. In my .Net framework that's emulating what I want on the GTT, I just installed a global hook that tells me everything. In the past, I've also over-ridden all the mouse_up and mouse_downs on the buttons and other areas, depending upon what I've needed to do.

It's normally not a big deal for interfaces, like on a computer, especially since you can over-ride them. But for small, touchscreen devices where you don't have multiple windows, or a mouse, or a keyboard, you have to make the input method pretty much global. Just like on Android or iphone, or garmin gps or whatever, you have global swipes, long presses, double-taps, gesture actions, etc. On a screen space that small, and with just that one method of input and navigation, you really can't just have a mode, or by default section off a part of the screen to do just one thing. Computer UI stuff doesn't translate well onto stuff this small :)

Post Reply