>
> My app is 4.9 rating since a very very very long time and all very very
> few bad rating always had comments.
> Since 14 days I start to have lot's of 1 or 2 stars rating without any
> comments and without having released a new version or the app having
> problems.
> There's still the same
> > Yesterday I moved my app which was quite highly ranked from Paid to Free
> app
> > with in app billing. The app instantly dropped to literal obscurity the
> > second I did it. It was almost as if my ranking value in the database
> had
> > been reset!
>
Well you are is now positioned aga
>
> 1. Share and find some practical solutions, if there are any.
>
There isn't any, for some specific use cases (games) best solution is to
use higher level API (game engine).
2. Bring the seriousness and span of this issue to Google's attention - as
> there seems to be complete ignorance and
> The best solution to your problem is probably to "bite the bullet" and
> rewrite your code to use shaders in OpenGL ES 2.0. All major phones and
> quite a few minor ones support it now.
> It may be your only solution, if the Galaxy S3 has to move too many more
> bits than the P350 did, or
>
> .I've decided to move to iOS development, and the only way to convince me
> otherwise is to give me a decent, reliable way of dealing with fragmentation
There isn't any reliable way of dealing with fragmentation - devices and
platform bugs has to be workaround (if possible) or device marke
> Well, I never said I was catching everything indiscriminately. I catch
> things that I can do something about. Sometimes there's a silent fall-back
> and the app doesn't even need to bother the user about it. Sometimes I need
> to involve the user in what to do next. Sometimes I simply tell
>
>
>
>> Assuming you are doing all that and giving the best user friendly message
>>> to the user possible, you still might have a vested interest in the nitty
>>> gritty details of those exceptions to fix the problem. Original poster's
>>> query is very legitimate.
>>>
>> Why? How can you fi
>
> The idea of exceptions as "unexpected" falls away once you are doing
> anything like network access or even file access.
>
Java distinguish between this two situation with checked and unchecked
(runtime) exception. Runtime exceptions should never be cached (unless you
can propose some val
> The alternative is to start removing catch clauses so that the app will,
> in fact, crash and the developer can at least see something. This seems
> like a silly approach, though.
>
Quite the contrary: if you are catching runtime exceptions then you are
doing it wrong (most of the runtime e
W dniu wtorek, 25 czerwca 2013 13:06:35 UTC+2 użytkownik reaktor24 napisał:
>
> Hi. thanks for the reply. I support Android 2.2 plus so don't really want
> to use that method. Also I don't like the look of this:
>
> 'the EGL context *may* be preserved when the GLSurfaceView is paused'
>
> EGL c
By default GLSurfaceView destroy context on pause, you can modify this
behavior with:
http://developer.android.com/reference/android/opengl/GLSurfaceView.html#setPreserveEGLContextOnPause(boolean),
if you app supports API levels < 11 you can use one of many GLSurfaceView
replacements with this
> Hmm. I did not know that OpenGL managed itself that way, last thing I want
> is to mess up the VSync. But anyways, I realize that using a game engine
> would make this significantly easier, but where's the fun in that?
>
In making game itself?
> But I am stumped, please explain to me ho
You are queening events from constructor so they may be executed before
context is created and bound (make current in render thread - if you check
GLSurfaceView you will notice that renderer thread is started when renderer
instance is set but egl will not be initialized until surface is attached
> I haven't found workaround so until then I wanted to disable the feature
> from ui.
> Could you be more specific about 10-bits-mediump fragment precision
> workaround?
> I'm interested.
>
As I said that really depend what is your issue, I use several different
techniques for this, as I ment
> My app is using some graphic filter with GLSurfaceView & OpenGL API
> and I recently found my algorithm depends on shader precision.
>
> Now I want to my app to show the GLSurfaceView only if the device supports
> proper shader precision.
>
But what's your issues exactly? I'm guessing you hav
> I am trying to add AdMob to my project, but I have a big problem during
> the compilation part. Basically I have a script which does all the steps to
> generate the APK. When I don't add the AdMob jar, my script works
> perfectly, except that javac will throw an error that AdMob is not
> de
>
>
> call javac -verbose -classpath
> %APK_SDK_PLATFORM%\android.jar:C:\Programas\GoogleAdMobAdsSdkAndroid-6.4.1\GoogleAdMobAdsSdk-6.4.1.jar
>
> -d bin\classes src\%APK_PACKAGE_PATH%\TrueDemocracy.java
> gen\%APK_PACKAGE_PATH%\R.java src\com\jjoe64\graphview\*.java
> src\com\jjoe64\graphview
W dniu wtorek, 2 kwietnia 2013 19:19:41 UTC+2 użytkownik bergstr napisał:
>
>
>> First thing first, you should go to the console and check your app
>> settings. It will also tell you what devices are being filtered and why...
>>
>
> thats what I did. All my devices (currently, a Nexus 7 and Ne
W dniu wtorek, 12 marca 2013 18:13:16 UTC+1 użytkownik Jason napisał:
> Bashing eclipse is popular because eclipse is a horrible experience, from
> start to end.
Exaggerating a bit, don't you think? Sure 4.2 isn't such a great release
(new UI framework is still buggy, but SR2 helped a lot - a
W dniu środa, 6 marca 2013 10:24:29 UTC+1 użytkownik latimerius napisał:
>
> On Wed, Mar 6, 2013 at 7:48 AM, a1 > wrote:
>
>>
>>> My minSdk is 7. Preserving GL context is a hint. You call it and pray
>>> it works. See the official docs at http://de
> Oh my god, really? We are in 2009 again or something? Since Android 3.0,
>> you can call setPreserveEGLContextOnPause method to prevent GLSurfaceView
>> from releasing context,
>>
>
> My minSdk is 7. Preserving GL context is a hint. You call it and pray it
> works. See the official docs
W dniu środa, 6 marca 2013 00:06:32 UTC+1 użytkownik RichardC napisał:
>
>
> On Tuesday, March 5, 2013 10:20:38 PM UTC, a1 wrote:
>>
>>
>> Oh my god, really? We are in 2009 again or something? Since Android 3.0,
>> you can call setPreserveEGLContextOnPaus
I'm not quite sure about the rendering thread though - that's controlled by
> GLSurfaceView and I'm afraid I can't just pause it with no other
> side-effects. I think the only way would be to
> call GLSurfaceView.onPause() which however also means a lot of other
> undesirable things includi
> But this means you can only send a maximum of 24 sprite positions with
> each update. That isn't good enough. Mario's code is sending 100 sprites
> every frame, not 24. Surely it should be easier under OpenGL 2 and not more
> complicated. Anyone got any advice?
>
You'd be better off if you
It was never an issue, GLSurfaceView was a bit eager to destroy context on
pause, since 3.0 you can use setPreserveEGLContextOnPause or if you need to
support older version you could just backport GLSurfaceView from 4.0 or use
one of many existing implementation (eg. the one from replica island)
>
>
>
>- Use Canvas to create a texture map with all used characters for the
>current locale (and I agree, that looks like a big PITA unless you're
>supporting just a single or a few languages)
>
> It's usually done offline (see BMFont
http://www.angelcode.com/products/bmfont/ a
W dniu niedziela, 27 stycznia 2013 04:04:22 UTC+1 użytkownik Justin Buser
napisał:
> I did actually, not that it's really relevant. I also looked back through
> the aosp commit logs and found a grand total of 1 commit that he made, but
> that's not really relevant either as it doesn't really PR
W dniu wtorek, 8 stycznia 2013 20:53:49 UTC+1 użytkownik bob napisał:
>
> Yeah, I thought I remember something about 1.5 allowing you to do this.
>
>
> But, I built it for 1.5, and it complains on the Nexus 7.
>
Because it's a bug and in fact a really bad one. GL take ownership of
pointers pas
W dniu wtorek, 8 stycznia 2013 18:39:38 UTC+1 użytkownik bob napisał:
>
> Well, I was getting this error:
>
> ... called a GL11 Pointer method with an indirect Buffer
>
>
> As a long shot, I was hoping maybe OpenGL 1.0 would just let me use
> indirect Buffers. Then maybe I wouldn't have to chan
> GLES20.glUniform1i(mTexSamplerHandle, 0);
>
> I would think it should be:
>
> GLES20.glUniform1i(mTexSamplerHandle, texId);
>
> But that doesn't work.
>
> Can someone help me understand this line?
>
Each GPU has multiple texture units (at least 8 as required by GL ES 2.0
specification), sam
>
> Did you check with MAT? Also you may try to verify this theory by
> forcing process termination with System.exit.
>
> I've doing that since half hundred of versions ago, more or less, so I am
> pretty sure about it.
>
No you don't only you only kill on explicte exit (probably in
onBackP
W dniu piątek, 23 listopada 2012 12:31:22 UTC+1 użytkownik Fran napisał:
>
> On 11/23/2012 10:30 AM, a1 wrote:
>
>
> As told to Mark, I do not think so: the game is just starting, loading
>> the first Screen at all.
>>
>
> Really?
>
> Well, th
> As told to Mark, I do not think so: the game is just starting, loading
> the first Screen at all.
>
Really? Data suggest quite the contrary: your heap is heavily fragmented,
your process have allocated 48MB and released a lot (since there is 15MB),
that's suggest lots of memory operations
W dniu środa, 17 października 2012 00:27:27 UTC+2 użytkownik Kristopher
Micinski napisał:
>
> The NDK is for *pieces* of an app. Typically very high performance
> pieces, but not the actual app itself.
>
Not really, since 2.3 there are two legitimate scenarios for NDK use:
1) as you mention
W dniu środa, 25 lipca 2012 23:22:41 UTC+2 użytkownik Matt Terry napisał:
>
> While 16:9 is a common aspect ratio for TVs, displays, and monitors...it's
> not a "common" ratio, or dimensions, for Android's screen qualifiers...
>
Yes it is. All high end motorola phones has 16:9 ratio (854x480 -
W dniu poniedziałek, 25 czerwca 2012 16:05:30 UTC+2 użytkownik Miha napisał:
>
> Hi!
>
> In my app, I am showing an alert dialog and in it an EditText for the
> password with code like this:
>
> _input = *new* EditText(BaseInfoActivity.*this*);
> _input.setSingleLine();
> _input.setInputType(I
> When trying to blend the full-size camera image with an equal-size
> bitmap for the filter, unless I reduce the image size, I get an
> out of memory error. I really would like to allow users to get the
> maximum resolution possible.
>
After scanning through the thread I've noticed that no one
> Thanks for pointing that out for me.
>
This is wrong of course: r.nextInt(0) will throw IllegalArgumentException.
First of all try to remove try/catch block - you are suppressing all errors
that way and none of the function that you are calling inside it should
throw. My guess is that you
I honestly think that changing from "Just In" to "Top New Free/Paid" helped
indie developers. Just before scrapping, "Just In" was in sorry state -
lot's of "developers" gaming it (either through constant "updates" or even
constant republishing of same content: http://goo.gl/7A17r) so anyway you
>
> Could you release an update and offer something for the rating? Additional
> levels, characters, some way to personalize the game for the user?
Which is, of course, violation of market content policy terms (
http://www.android.com/us/developer-content-policy.html see "Spam and
Placement"
At first glance: internalformat in glTexImage2D do not match format in
glTexSubImage2D. You should either define texture as RGB565 or change data
format passed to glTexSubImage2D.
--
Bart Janusz
--
You received this message because you are subscribed to the Google
Groups "Android Developers"
It's not a problem of passing such a big array but rather a problem with
processing it. You didn't show your code but it probably looks like that:
void native_function(JNIEnv *env, jobject thiz, jobjectArray jarray) {
...
for (int i = 0; i < env->GetArrayLength(jarray); i++) {
jstring
Using obfuscator is always a good idea because it shrink code (ant build
system from ADT supports proguard), on the other hand do not really count
on making bytecode temper proof, it's really easy to crack any licencing
solution even if code is obfuscated.
--
You received this message because
Signature should be "(II)Landroid/graphics/Bitmap;" colon is used as class
name string terminator not as type list separator (types in signatures are
single uppercase letters, except of classes which are represented by
L; where colon is terminator)
--
a1
--
You received this messa
Just a small correction, constant to check VTF support name is:
GL_MAX_VERTEX_TEXTURE_IMAGE_UNITS.
--
You received this message because you are subscribed to the Google
Groups "Android Developers" group.
To post to this group, send email to android-developers@googlegroups.com
To unsubscribe from
While specification states that texture lookup are available for both
vertex shader and fragment shader not all GPU supports VTF (vertex texture
fetch), you can check if GPU supports VTF
by querying GL_MAX_VERTEX_TEXTURE_UNIT (which is allowed by GL ES
specification to be 0).
--
You received
And did you correctly setup attenuation equation? Attenuation is controlled
by glPointParameter function (
http://www.khronos.org/opengles/sdk/1.1/docs/man/) with pname
GL_POINT_DISTANCE_ATTENUATION, where you suppose to provide array of 3
floats with attenuation equation coeffs (a, b and c).
You may try to set even lower max streams value (like 4). This is basically
trial and error - I've tested several combination of codecs and streams count
(eg. raw mono pcm worked fine in my case but they are huge). Make sure you test
on orginal samsung galaxy with 2.3 firmware as it tend to have
Lower number of max streams to 8 and re-encode sounds to mp3 96k+, generally
speaking sound drivers and player implementation on samsung device with
android 2.3.x are nightmare, what I found out:
- some OGG files that play on every other android device fail to load
through MediaPlayer on SGS wi
I really doubt it has anything to do with this malaware report, judging on
appbrain page your app/account was removed because you violated Market
Content Policies (http://www.android.com/us/developer-content-policy.html):
"Product descriptions should not be misleading or loaded with keywords in
Good day,
With recent release of SGS2 I've received several notifications (market
comments and mails) that my app crashes on this phone. Unfortunately it's
looks like crash in native code so no error report from market, what's more
important few user contacted me directly, so I ask them to use o
Looks like algorithm used to determine position in top list had been
reworked as there is a lot of changes in top through last few days
--
Bart Janusz
On Dec 18, 6:29 am, TreKing wrote:
> On Fri, Dec 17, 2010 at 8:34 PM, Zsolt Vasvari wrote:
> > Has anybody seen anything like this with their ow
I think you should first take a look at:
http://developer.android.com/reference/android/view/ViewGroup.html#onInterceptTouchEvent(android.view.MotionEvent)
and here:
http://android.git.kernel.org/?p=platform/frameworks/base.git;a=blob;f=core/java/android/view/ViewGroup.java
I really think that thi
> There are a wide range of libraries available that will let you
> collect stack traces (Flurry, DroidDrop, etc.) or LogCat output
> (Android Log Collector). Integrate one of these into your application
> in the next update, so you can get more diagnostic information.
I use stack trace collector,
Sorry for bumping but I'd really appreciate some help.
--
Regards,
Bart Janusz
On 17 Wrz, 06:48, a1 wrote:
> Hi,
>
> Does someone have or have access to this phone? After last update of
> my game I two comments and a email from a user stating that there are
> some freezes
Hi,
Does someone have or have access to this phone? After last update of
my game I two comments and a email from a user stating that there are
some freezes and crashes on this very handset, but without logcat it's
not much I can do. If someone has myTouch Slide and would like to help
please send m
> Does anyone has a simple example of to use LocalActivityManager or
> something else to meet my requirements.
void createInnerActivity(ViewGroup container, Class activityClass)
{
if (container.getChildCount() != 0) {
container.removeViewAt(0);
}
final Intent i
> Because you can't put an Activity inside a View - that's simply not
> supported and makes no sense.
You can and moreover it makes a lot of sense, TabHost in fact relies
on this.
>
> > see:
> > developer.android.com/reference/android/app/LocalActivityManager.html
>
> And ... ?
Subclass Activity
> But I'm not sure that native code is any harder to patch, and there
> are still identifiable syscalls or calls back up to java for i/o to
> show where it tries to accomplish verification.
First of all it's much harder to bypass especially if you are dealing
with optimized code and you will have
, eg. alloc some object etc
return performValidityCheck()
}
sure entry point is in bytecode but removing entire call will cause
crash.
--
Bart Janusz (Beepstreet)
>
> Only obfuscation will help to some extend, making it harder to find
> what to patch, IMHO of course :)
>
> Hugo
>
&
ake it
> harder to find the point to patch in the decompiled code.
> They will always be able to pirate your app, but it's not as easy as
> copying the apk and putting it up on a website anymore. So in that
> sense it helps.
>
> Hugo
>
> On Aug 24, 6:48 am, a
It's really simple, if you use stock implementation of LVL your app
will be cracked in no time, otherwise (if you make the code
complicated enough and then obfuscate) it will take at least week or
two to crack it. If you use native code and put most of the antypiracy
code there you will be good for
Hi,
Do you use default policy implementation provided with LVL
(ServerManagedPolicy)? With 10k user base I received about 10 "can't
validate" reports, and rollback LVL.
--
Bart Janusz (Beepstreet)
On 8 Sie, 11:37, Pent wrote:
> LVL has been in use for around 48 hours with my modest several
> th
> I don't know how google could push paid apps but is it a coincidence
> the ratio of paid to free is massively higher on apple?
Because developers from only 9 countries can sell apps?
> I personally wouldn't mind a moderated market place. I know apple aren't
> popular
> here but no one can deny
> Reading this thread, I understood that one spammer already has 3 or 4
> accounts on the Market. This means he already paid 100 dollars to publish
> apps.
>
> Developing an app costs much more that the 10 $ of publishing fee you
> propose. Even the simplest would require at least a couple of hours
On 6 Sie, 16:25, Alessandro Pellizzari wrote:
> Il Fri, 06 Aug 2010 09:05:32 -0500, TreKing ha scritto:
>
> > Actually, I think it's a great idea. What hurts legitimate developers
> > (and users, and the Android platform in general) is this rampant,
> > uncontrolled garbage in the Market. Start
This will not change anything, the biggest problem with spam is that,
well, they spam all categories, so adding additional category is
pointless. I think that there should be a publisher fee for each app,
this will stop spam for sure.
--
Bart Janusz (Beepstreet)
On 6 Sie, 07:42, metal mikey wrot
On 1 Sie, 21:12, sblantipodi wrote:
> have you tryed using a non authorized gmail account from emulator?
> it returns always licensed... have you noticed the same?
I do not test on emulator, since my games are GL based emulator is
unusable (also it's slower than actual device), but from what I re
Hi everybody,
I've just pushed emergency fix after last update which disabled LVL
check, 3h after update I received about 5 emails and 4 1* comments due
to licence verification failure on legit copies of my game.
I've used slightly modified ServerManagedPolicy (with minimum cache
time set to 1 we
Proguard make it a bit harder to RE, but it'll still possible (and the
APKtool gives you possibility to debug which is a really powerful RE
tool), moreover you cannot use all of proguard optimization because
you will not be able to convert classes to dex, in fact you can only
use shrink and agressi
On May 6, 9:18 pm, strazzere wrote:
> As you could also note - there is nothing in Android-Market-Api's
> license against this type of use.
Yes there is, google APIs services in point 5.3 states that you are
not allowed to use undocumented APIs:
"5.3 You agree not to access (or attempt to acce
It uses this project: http://code.google.com/p/android-market-api/,
you can do same, just note that in proto purchased field is missing,
but you can simply extend App message in market.proto, purchased field
has id 34.
--
Regards,
Bart Janusz (Beepstreet)
On May 6, 8:22 am, Edward Falk wrote:
>
On May 5, 8:09 pm, dadical wrote:
> Hey Tim.
>
> You're correct that validating purchase with the market is a key piece
> of our solution. Figuring out how exactly to do that using Google's
> binary market protocol in an efficient way (try doing everything that
> AAL does in a 35 KB library) wa
Running calculations in separate thread will only make things worse,
but as I understand you are performing all math in Java I think using
native code here can dramatically boost performance
--
Regards,
Bart Janusz
On Apr 22, 5:15 am, BobG wrote:
> I have this audio analyzer app that fills a b
Check this bug
http://code.google.com/p/android/issues/detail?id=2828&q=glsurfaceview&colspec=ID%20Type%20Status%20Owner%20Summary%20Stars
- as you'll see current implementation allows only one glsurfaceview
instance per process
On Mar 18, 8:36 am, Tim Liu wrote:
> Hi,
>
> I'm having a problem w
I'm sure bunch of divisions has nothing to do with your performance
problems. If you call glColor such frequently you are also calling
draw functions frequently (possibly with different textures) pushing
geometry in small chunks will kill you framerate (not to mention
pushing a lot of textures) try
Usually stream returned by http connection will support mark (see
InputStream#markSupported), but it's safer to wrap returned stream
with BufferedInputStream since size of buffer is unknown on original
stream (and it may not be large enough to keep all necessary data).
With stream which support mar
On 29 Gru, 17:49, Mika wrote:
> OH sorry guys, my bad, I wasn't specific enough.
> I'm TRYING To shift myself into fixed point(FP) world and trying to
> get rid of the long calculations :) So.. does the topic make more
> sense now? Sorry for being such a numnum and not realising that FP =
> fixe
Well, yes, you can simply shift to fixed since t is difference which I
should be around 20.
But what's the point? I'm using floating point exclusively for
calculations, and really do
not have problems with performance. If you do not perform heavy
calculations it doesn't matter,
and if you do interp
gt; I bet if you looked at the EGL_CONFIG_CAVEAT attribute, the RGBA
> configs will be EGL_NON_CONFORMANT_CONFIG. Passing
> EGL_CONFIG_CAVEAT=EGL_NONE as your desired configuration should filter
> out the RGBA configs assuming my assertion is true.
>
> Ernest Woo
> Woo Games
>
On 9 Lis, 10:13, "Michael Leahy (Tactel US)"
wrote:
> Greets,
>
> As with a lot of folks doing OpenGL ES dev it's really a test on
> device situation. I'm really trying to get out the Typhon API (http://
> typhon.egrsoftware.com/) which is an open source platform for real
> time / game developme
Congratulations! Your application 'Blockx 3D Pro (ADC2)' was selected
by Android users as one of the top 20 in the Casual category!
Well, good news... bad is that I know that this version doesn't work
on droid.
--
Bart (arcone1) Janusz
--
You received this message because you are subscribed to
I think you should first take a look at this page:
http://www.google.com/support/androidmarket/bin/answer.py?answer=150324,
as you see Argentina is not supported merchant location.
--
Regards,
Bart Janusz
On 28 Paź, 23:02, shadow_of__soul wrote:
> Hi,
>
> i'm interested to develop games for the
I dig a bit more to find reason of this behavior and in case someone
will have similar problem I'd like to summarize my findings:
This is original allocator dump:
> D/Blockx.GameController(900): Surface created
> E/SurfaceFlinger(577): not enough memory for layer bitmap size=1642496
> (w=480,
> 10-24 12:42:49.860: DEBUG/Blockx.GameController(900): Surface created
> 10-24 12:42:49.890: ERROR/SurfaceFlinger(577): not enough memory for
> layer bitmap size=1642496 (w=480, h=854, stride=480, format=1)
> 10-24 12:42:49.890: DEBUG/MemoryDealer(577): LayerBitmap (0x2e9350,
> size=8388608)
>
I'm testing my application on WVGA854 emulator. I use two activities
with GL surface one is a menu and second is game itself. While
launching game in logcat there is a message about out of memory error
and application is killed:
10-24 12:42:49.860: DEBUG/Blockx.GameController(900): Surface create
> - Has anyone else run into this issue?
Judging on the stacktraces I'll say you run into this issue:
http://code.google.com/p/android/issues/detail?id=2828.
--
Bart (arcone1) Janusz
--~--~-~--~~~---~--~~
You received this message because you are subscribed to
87 matches
Mail list logo
