Sorry your browser is not supported!

You are using an outdated browser that does not support modern web technologies, in order to use this site please update to a new browser.

Browsers supported include Chrome, FireFox, Safari, Opera, Internet Explorer 10+ or Microsoft Edge.

AppGameKit Classic Chat / Doesn't work too good on Samsung Galaxy Note 8

Author
Message
Coilwinder
11
Years of Service
User Offline
Joined: 28th Feb 2013
Location: Norway
Posted: 2nd Jun 2014 10:22
I got the AppGameKit (1 and 2 beta was it?). Purchased about a month ago, tried the AppGameKit 2 a little bit then but it wasn't too sucessful. Some tests worked others not.

Just installed the APK again, version 1 this time, to try it some more. Basically the same results as last time.

It may be I have a "difficult" Android tablet? Its the Samsung Galaxy Note 8, which have both the usual multitouch screen, and a wacoom(?) tablet-type S-pen.


Testing some of the tutorial programs here: http://www.appgamekit.com/code-snippets.php

Often results in a black screen, or a non-responsive one.

The "particle effects" demo did show something, but didn't react to touch input.

The "Point sprite at mouse x,y position" had an interesting bug: It did show the white box at the center of the screen (the sprite), but did not react to the touch input Y-value (I guess). Moving a finger up or down didn't matter, but it reacted when moving sideways.

As if getPointerY() don't work, but getPointerX() do? Or maybe its the atanfull function, or something else, I don't know.

The PC version worked with the mouse though.
Markus
Valued Member
20
Years of Service
User Offline
Joined: 10th Apr 2004
Location: Germany
Posted: 2nd Jun 2014 15:31 Edited at: 3rd Jun 2014 11:17
be sure u use v108 19 + player from google play store.
v2 is alpha only, need extra agk player, its not in google play store.
also agk player must match the agk ide version.
some android os versions from device manufacturer result also in problems.
some code snippets are old, maybe not testet with never agk version.
if agk player did not work i often drag it out of screen and than it start fresh.
if touch input or some commands did not work at your device, please tell tgc.
...
about touch input, what android os have this device?

AGK 108 (B)19 : Windows 8.1 Pro 64 Bit : AMD Radeon R7 265
Naphier
14
Years of Service
User Offline
Joined: 2nd Oct 2010
Location: St Petersburg, Florida
Posted: 3rd Jun 2014 05:18
Most of the examples are a bit out of date, but the principles should still hold true.

The point sprite at mouse snippet looks to me to have some immediate errors. I've attempted to correct them here:



Try that out. Check out the values. Pointer commands might not work well on that device and if they don't it would be good to know. They should be reported to AppGameKit and the better we can describe the issue the better they can fix it.

TDavid
11
Years of Service
User Offline
Joined: 7th Feb 2013
Location:
Posted: 3rd Jun 2014 11:07 Edited at: 3rd Jun 2014 11:08
If there are problems with the pointer, it might stem from Wacom technology. I don't mean that it doesn't work well but that it's hard to work with. I know paint.net creator made a version of his program in which wacom sketching tablets were supported. A few built later their support was removed. Then he told us it would be supported again in version 2.0 but he didn't deliver. It may be a coincidence though.
Coilwinder
11
Years of Service
User Offline
Joined: 28th Feb 2013
Location: Norway
Posted: 3rd Jun 2014 12:06
Thanks for replying!

(Slight typo in my first post, it should ofc be "AGK", not "APK")


Markus:

The google play store doesn't say "update" so I guess I have the latest player. At the bottom of the player screen (when starting the player) it says "10000008 build Aug 6 2013".

If memory serves, it worked just as good (or bad) with the AppGameKit 2, I didn't know there was another player around for that one.

My Android version is 4.2.2.



Naphier:

Thanks for that! It doesn't work any better than the other one, but at least it's a starting point for figuring out things.

It seems I got it the other way around, it seemingly is the py# variable that works, except it think it is px# (reacting when moving the finger, or the S-pen, horizontally).

The px# (or getPointerX()) is reacting as if it is the getPointerState(), showing "0" (zero) when not touching the screen, and "1" when I touch the screen. No floating points, just 0 or 1.

The value for py# shown on screen is also an integer, 10 digits long. From about 1065353216 (leftmost) to 1151328256 (rightmost). This was in landscape mode. About the same if started in portrait mode, except rightmost value is "slightly" less at 1145552896.

Resolution of the screen is 1280x800 btw.



So it seems getPointerY() actually accesses some kind of X value, while getPointerX() is getPointerState().

The real getPointerState() only reacted when I lifted the finger (or pen) up from the screen. Then for a very brief moment I can see some numbers flashing by, and the sprite changing direction a bit (not the correct direction as it lacks the real px# value ofc).

It didn't seem to make any difference using the S-pen. It is pressure sensitive, but I didn't see any difference in either the px# or the pointerstate value. When I lifted the finger (or pen) from the screen it also took on a 10-digit (ish) number (see slightly modified code below - printing the pointerstate to the screen - ofc you have to have the same tablet, or a similar one maybe).


I guess (but don't know) pressure sensitivity isn't supported by the AGK. And it seems the addresses of which to access the X and Y etc positions is a bit different in this device (or possibly Android version?).


Modified your code snippet a bit. One thing I found is that I could not use



Not sure if its the "str" function or print() itself that protests. As sais I got an error "unknown instruction 649". Commenting out the print statement fixed it (the if block worked).

But changing "pointerState" to a float also works for printing:




Code snippet:



So I guess it's time to report something to the AppGameKit people.. not sure where or how to do that yet. Unless they also peruse these forums (I see you have a "AGK developer" text by your nick?).

(Hope this post isn't as messy to read as it was to write )
Naphier
14
Years of Service
User Offline
Joined: 2nd Oct 2010
Location: St Petersburg, Florida
Posted: 3rd Jun 2014 13:41
That's all quite odd. Sounds like the function just aren't grabbing the right data. The one print statement is probably failing because the value is too big or something.
You can report bugs here:
code.google.com/p/agk

They'll get to them some day!

If I were you I'd try out the actual touch commands.
Here's an example:



Then call GetCurrentTouchX/Y when you've got the right touch event.
Setting GetRawTouchCount(0) and GetRawFirstTouchEvent(0) excludes unknown touch events. Set it to 1 to include them and you might see a difference.
I recently learned that iOS sometimes has an issue with the GetPointerPressed() commands too (they were inexplicably not updating on the second finger press until you've pressed for a full second).
So toy around, see what works. It's probably best to always use Touch commands with mobile I'm thinking.

Coilwinder
11
Years of Service
User Offline
Joined: 28th Feb 2013
Location: Norway
Posted: 3rd Jun 2014 14:40
Alright, thanks for your help! I will try that.

(Also, my previous post seems to be missing, or not accepted (yet?)).
Coilwinder
11
Years of Service
User Offline
Joined: 28th Feb 2013
Location: Norway
Posted: 10th Aug 2014 22:22
Minor update. I just thought I'd mention that I tried the same (the above modified code snippet from my last post above), with a Samsung Galaxy Tab 3. Same results. I also made a note of it on AppGameKit google code page, in the bug report for the other Samsung (Galaxy Note 8) tablet.

Login to post a reply

Server time is: 2024-11-25 07:37:16
Your offset time is: 2024-11-25 07:37:16