Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - spacefractal

Pages: [1] 2 3 ... 235
its would been a very great help, if you can do a simple gamepad testing app. Now im have used a bit too much time in this one (hehe). Etc its must been your turn to help!

Etc its should show a gamepad and the default mappings like this screen:

And then do a "remap" function for the image, if the controller does not working.

Then its would been possible to people to send images to us as im have no changes to testing all controllers that is out to the market.

Im is happy if its just using with KEY() values im have used by default, not by using JOY commands (except for the naming).

For auto remapping, then JOY() commands for those joypads would works out of the box.

Howover for most cases today, right stick and both triggers is seen often been swapped.

You can also get the controller name now with the newest update (Currectly with GETJOYNAME$(0), but later when im update, its would been GETJOYNAME$(1) instead as tilt default, allways enabled, uses GETJOYNAME$(0). So dont hardwire it by number, but with a variable instead.

Last time, you said its still broken, but then im might have oversigt it. but phew :-).

AS is quite stable here throught. correctly Using 3.5.3 with only using the Android Studio JDK. Howover using about 1.5gb of memory (im do have 24gb ram, so not a problem).

Im do prepare the gamecontroller code to possible to use multiply gamepads with own values seperated include Tilt (as own joypad). Tilt example now routed to GETJOYX(0) and GETJOYY(0). Normal controllers just use another joynum.

this was not what im asked, please confirm the assest files now works again? gamepads is toally different beast, which is hell off annoying on Android. Trust me.

Im have only a ouya controller to test on a wireless device, so you must wait here to next week. That it. im actuelly newer tested with x-box controller as its dont follow the google standard, while ps4 does.

Also its been changed quite a bit with the JOY commands communication with Java and c++, so multiplayer and tilt can been worked.

So except this to been changed to a better tight with the JOY commands and KEYS() would been expands to much much more than 256!

its look like there might still have been assets loading as reported by mr plow. Im have no idea why, while assets files in my games works without issues.

For best practical, allways uses SETCURRENTDIR("Media") start on your game.

But its should not matter you use full path or relative path, but its clearly best using relative path throught.

Also in this version, JOY VALUES has been updated (only have a OUYA controller to test throught) and you can now get the name using GETJOYNAME$(0).

Currectly im can only garentee OUYA, X-BOX and Playstation controllers to work, but its under testing and im waiting for a Android TV. But if you do a remap/binding feature for your game, pretty much all controllers should works.

As there is still issues im hold back for the steam release, until im make sure its works as intended!!! (then im remove Beta again). This why im diddent have updated the steam version yet.

anyway, while im newer got a change what its happens with the assests and sound, and im newer say the projects (only what im can help now, if still issues), then im now worked a bit on JOY using the OUYA controllers. Im cant test with other controllers before the Android TV arrive.

The values that is on the sticky thread is actuelly a bit wrong. Please update it.

Im used this test code for glbasic:

Code: (glbasic) [Select]
LOCAL xx% = GETJOYX(0)*100
LOCAL yy% = GETJOYY(0)*100
LOCAL zz% = GETJOYZ(0)*100
LOCAL rx% = GETJOYRX(0)*100
LOCAL ry% = GETJOYRY(0)*100
LOCAL rz% = GETJOYRZ(0)*100
LOCAL dx% = GETDIGIX(0)*100
LOCAL dy% = GETDIGIY(0)*100

DEPRINT("name: "+GETJOYNAME$(0)+" x:"+xx+" y:"+yy+" z:"+zz+" rx:"+rx+" ry:"+ry+" rz:"+rz+" dx:"+dx+" dy:"+dy)

Im uses a DEPRINT function for SDOUT here really. Just change it what you want.

Yep, GETJOYNAME$(0) now works with the new c++ version (you should no need recompile it).

in c++ im used those values:

Code: (glbasic) [Select]
        Data->lX = 32767 * (AndroidFloatCode(244) - AndroidFloatCode(245)); // FIX 2020 GOOGLE STANDARD
        Data->lY = 32767 * (AndroidFloatCode(246) - AndroidFloatCode(247)); // FIX 2020 GOOGLE STANDARD
        Data->lRy = 32767 * (AndroidFloatCode(250) - AndroidFloatCode(251)); // FIX 2020 GOOGLE STANDARD
        Data->lRx = 32767 * (AndroidFloatCode(248) - AndroidFloatCode(249)); // FIX 2020 GOOGLE STANDARD
        Data->lZ = 32767 * (AndroidFloatCode(252) - AndroidFloatCode(253)); // FIX 2020 GOOGLE STANDARD
        Data->lRz = 32767 * (AndroidFloatCode(254) - AndroidFloatCode(255)); // FIX 2020 GOOGLE STANDARD

PS. Tilt on Android has been very much offtypic now, but changed the subject to more ontypic.

GLBasic - en / Re: Tilt on Android
« on: 2020-Feb-20 »
Again: im can first fix it when I'm get android TV, then I'm checking with both xbox and ps controllers. So im have no currectly focus with game controllers at all. Im have no devices recently to testing, only a OUYA controller, which uses totally none default values.

But if you do remapping/binnding feature to the game, any sticks etc should works again. This is very important anyway as YOU CANNOT relay on the defaults values anyway. Im will only garentee xbox, ps3/4 and Ouya would works out of box when im fix it using Android TV (as well of course the remote), when im also get the devices. Im orderd one recently, but its a bit delayed.

Those kind of devices have been some kind of heavy selling in Denmark recently, due a recently TV War between to companies (That include Apple TV).

Im is aware the JOY() is currectly broken, but use the KEY() instead. They should been uneffected.

Its much much much more perfered to im know the assets and the sound is working again, so im can push this verson to steam. Im have hold it back until im know its working as intended.

PS. Im have done a mini "review" how users can fix the game.

GLBasic - en / Re: Tilt on Android
« on: 2020-Feb-20 »
If still issues you need to send the project. But not home in later today. I'm is out not. See you.

GLBasic - en / Re: Tilt on Android
« on: 2020-Feb-20 »
Now I have absolutte no idea, except its that cradle issue and or you did not update as.

Checkout logcat what it's try to do.

There was c++ change to fix sound thing.

Now I'm out soon and is in bus.

GLBasic - en / Re: Tilt on Android
« on: 2020-Feb-20 »
I'm  don't have a jdk installed at all, but using that open source version  android studio has been installed by default. So I'm not seeing it. This could been the reason for the long compile time.

Grandle should use same path as as uses.

You should also not need to clean for a c++ update as well.

GLBasic - en / Re: Tilt on Android
« on: 2020-Feb-20 »
its too late. allways testing before doing it launch. currecly should have set it as a beta rather than official.

here its dont take that long time throught. about 3-4min for Greedy Mouse (which is quite big) and much less for the Break Bricker game (howover need do a new AS for that eventuelly for testing). Im also have a I7 CPU here.

You only would need to unpublish for a day or two, then republish the app again. people that do have it can still use it.

Howover im do long time unplushed all my games and property do complete new id for those when they are back again.

Now im is out for 5 hours or such. So im cant help you codewise until im home, if im is not tired......

Hopefully this one work, im tried all fail cases im could find for the assets loading. This is a blindfix, so hopefully works.

PS. Im also updated Android Studio to newest version without issues.

GLBasic - en / Re: Tilt on Android
« on: 2020-Feb-20 »
DONT USE JOY() commands. Here Im do cant testing it until im got my Android TV.  Property first next week!

Only changes was im slightly changed for analoge deadzone controls from 0.1 to 0.001, due tilt. but a Log.i im inserted to void reportkeycode() might have crashed it for spamming it too much? Remove that log in

You should take the app off until it fixed, and then reenble it again when its fixed.

PS. Im accidently edited your post. sorry.

GLBasic - en / Re: Tilt on Android
« on: 2020-Feb-20 »
Im have just see a issue in as AndroidStudioMakeApp.exe might delete all assets file in the AS project without updating it again. That is not cool. That is fixed as im just uncommed the killtree, as its not needed at all and im perfer doing this kind of thing manual really. So make sure the assets files actuelly exists in the Android Studio project!

All needed is just recompile it from glbasic and they are back again.

try this version:

This version logs much more info as well fixed LOADSOUND as well (fullpath should no longer required, etc if its fails first time. then its try again with fullpath).

This version does a better filter on various cases as im missed some few cases. Now even the internal font should now loads again, which was called before the internal setcurrentdir()).

Now its should not matter you uses full path or not, nor DOESFILEEXISTS().

GLBasic - en / Re: Tilt on Android
« on: 2020-Feb-20 »
About a half hour now. Also later today I'm is out to play some pinball. It's long time ago last time....

GLBasic - en / Re: Tilt on Android
« on: 2020-Feb-20 »
Pm me the as project. I'm have no idea how you load the files as they are all working here. I'm cannot blind fix a issue I'm cannot reproduce here.

Also I'm is not home before around 2 hours.

GLBasic - en / Re: Tilt on Android
« on: 2020-Feb-20 »
Yes, eventuelt add a does file exists as well.

It's a very old issue really.

Pages: [1] 2 3 ... 235