Messages

New daily.

Should fix the keyboard and the drawing problem in LITN (see bugtracker).

Edit: I am aware of the main menu text in The Infinity String not being drawn .
Edit2: Yeah, looks like that was just a problem with the translation file. Nothing to see here.

It's really quite weird.

Indeed because the games you list work for me in the latest version from http://jjs.at/daily (but they should also in R7). Are you sure they are correctly placed on the memory stick? The game files have to be in a direct subfolder of the engine (e.g. in ms0:/PSP/Game/AGS/My game/).

Gemini Rue

At the moment you will not be able to do the fighting and box pushing minigames I think. You could use the Windows binary from my daily build site and play through those sections on your PC because it should be savegame compatible though.

Thanks guys! 

Looks like the PlayBook can only run the Java code, but not native ARM instructions. Because the whole AGS engine is in native code, this is of course bad news. I guess nothing except for reimplementing the whole engine in Java could fix this .

David_OSU, thanks for the info! The problem is that the compiler generated code for neon, but to target the Tegra CPU it has to be set to vfpv3-d16. Fortunately this was easy to fix, a problem in the optimized assembler code would have been very hard to resolve for me.

The PlayBook is a Blackberry device, right? I have no clue how the development environment for that looks, but it is certainly not a simple recompile. This would be an entirely different port and I don't think it will happen tbh (at least not from me).

Ok, so you mean the current development version in the SVN. All changes were only to the editor as far as I can tell, so no problem there. Also all ports are built from the same engine source, so they should be identical in compatibility.

It could always run games from 2.6 to 3.2.1.

Uswin, this is good news indeed!

Regarding the sound stuttering and slow fading, did you try one of the hardware renderers (in the options menu)? It sounds like updating the screen and decompressing the audio is simply too much work for your CPU.

And thanks for the confirmation and logs.


Also: New daily. Should fix the mouse issues.


EDIT:
Regarding the theora crash: Uswin or David_OSU, does anyone of you have a rooted device? If so, can you post the output of the command "cat /proc/cpuinfo" in a terminal emulator? I guess you can also open the file in a text editor/viewer. Somehow I suspect that the Tegra 2 lies about its available features.

EDIT 2:
Maybe this new daily fixes the theora problem on Tegra CPUs: http://jjs.at/daily/AGS_2012-01-24_19-04-16.apk
But it is a stab in the dark. I will post the source later because my local repository got out of sync with the online one and I have to fiddle around with it some more to make it usable again :.

New daily: http://jjs.at/daily

- Fixed font outline drawing
- acsetup.cfg file is now always ignored
- Translation and "smooth character scaling" is now configurable through the UI
- Prevented accidental mouse movements while performing the two finger rightclick
- Same source now compiles for PSP, Android, Windows and Linux
- Exit menu only shown when holding the back button for a bit and releasing, but ESC key functionality not yet working


e: Oops. Didn't realize I would doublepost, sorry.

Uswin, you write that the Theora crash is gone on your Sony Ericcsson Xperia Neo V. But I remember that the original crash occurred on a LG P990 Optimus 2X. Can you confirm that the crash is also gone on that device? The logcat I have shows a SIGILL exception which means that an illegal instruction was fed to the processor.

The "framedrop" option cannot really have influence on the crash btw (at least not in a way that I can imagine).

There does seem to be a problem with theora video playback on nVidia Tegra devices (like yours and uswin's). The crashes in the Sierra remakes occur most probably when the intro logo video is supposed to be played. I guess I have to find a way to make the use of advanced CPU feature optional.

I just tried The Infinity String and I can confirm that the space key is not properly detected. In general keyboard input via hardware keyboard or via software (press the MENU button) should work.

Thanks for the reports!

I will make it an option.

Also I will get rid of reading acsetup.cfg. The only two relevant options are the translation and "smooth scaled sprites", which I will include into the option menu.

How about a plugin to enable the players to configure the android.cfg in-game?

You know what? This is genius. I had thought a bit about how to make Android specific commands available to the game author without fiddling with the editor and breaking things and this never occurred to me for some reason. But it is perfect for e.g. selecting between a "touch mode" without a cursor and the way it is now.

As for custom APKs, I am not sure how this has to be done at the moment. Ideally you would be able to pack every game file into the APK and it would then be installed into the default application directory indicated by the namespace instead of /sdcard/ags/. Building the APK would require the Android SDK though I guess.

[edit: I forgot that you have to sign the APK too, d'oh. So obviously you will want your own key and such.]

I have to disclose here that I am not an Android god in that this is my first Android app and I have never published anything on the market. So if anyone has experience there, I am listening  .


Uswin, I tried to reproduce the crash and I can't unfortunately. I tried it on a HTC Desire S where everything ran smoothly. I will send you a version that logs more debug output if that is ok for you?


Trapezoid, I will check that template out.

I thought about adding a "skip cutscene" button to the exit menu if a cutscene is currently playing.

It would also be possible to have the exit menu only appear if the button is held for, say, 2 seconds. Otherwise it would act as the ESC key.


Does the ScummVM behaviour not make it difficult to pixel hunt? I actually cannot really comment on this because ScummVM runs poorly on my HTC Dream. The mouse is very hard to control and jumps around a lot because the CPU is not very fast and seems to get pegged by the input events.


Regarding the Android store: I surely would make this available once it stabilizes a bit more. But I would only want to do this if it is alright with CJ and the people here. Especially the usage of the namespace com.bigbluecup.android should be discussed. I think it is appropriate and good for future updates but it might make it look too "official"? I would like to hear opinions on that.

New daily available: http://jjs.at/daily

It features ScummVM-style right clicking (hold down one finger and tap with the another). The reason I hadn't implemented this before is that I was on Android 1.6 when I started the port. That version does neither support ScummVM nor multitouch .

The most important thing should be refactoring the current source.

- moving code out of headers (acroom.h!)
- splitting ac.cpp into several files
- renaming variables in a consistent manner
- possibly removing global variables
- enforcing one consisted coding and naming style
- identifying obvious "dirty hacks" and implementing them the right way

This is not so easy because the code is so intertwined and because of the generous use of global variables a lot of functions are not stateless.

edit: Well, basically what monkey_05_06 said.

Alright, I decided to do builds of the Windows engine too. Check it out here: http://jjs.at/daily . It should fix the sound skipping.

In my ports I solved the sound stuttering by introducing a new thread that handles the sound updates. The only problem is that it sometimes throws off lipsyncing by a bit. Btw, the source in my repository now compiles for Windows too, so you guys can check it out if you want.

Imho libdumb is not a problem at all. It is very easy to compile the library for different platforms because dumb itself doesn't have any dependencies (except for allegro for aldumb, but that is a given). The build system it uses is also just a simple makefile and no involved automake or cmake procedure. What I am saying is that I had no trouble compiling it for PSP and Android.

As far as plugins are concerned: I don't see how this could be handled better than it already is. Again, there was no trouble porting it to Android (using mostly the same plugin system as for the Linux port) and even to the PSP.
The three plugins I offer with the ports (AGSBlend by Calin Leafshade and my recreations of ags_snowrain and flashlight) are written so that they work on Android, PSP and Windows with a simple recompile.

edit: Concerning AGS on OSX, doesn't that already work? http://www.adventuregamestudio.co.uk/yabb/index.php?topic=43383.msg578827#msg578827

For Abuse you need the "old" data files: http://abuse.zoy.org/attachment/wiki/download/abuse-data-2.00.tar.gz

There is always an access to the memory stick when a sound is played even if it is present in the cache. I don't really know what is read from the MS, but the sound is taken from the cache. The videos look pretty much like what I get on my PSPs. Probably the slight delay is simply from setting up and playing the sound. So it might be unavoidable.