well, I’m not sure what’s wrong, but all I get is a white applet once it has inited.
Are there any special requirements to run the applet?
Regards,
ribot.
well, I’m not sure what’s wrong, but all I get is a white applet once it has inited.
Are there any special requirements to run the applet?
Regards,
ribot.
On linux, I get appletviewer running OK, and this output:
qfp.Qfp:Hi
qfp.Qfp:start()
Copyright by Kai Wilckens
but then after a long wait where the applet claims to already have been initialized I eventually get:
? w=15640 h=68
kdk.Txt after abc MEM free:1.85 used:11.77 634%
Engine. NO PATHDATA
Engine Init in 33.63
java.lang.ClassCastException
at kdk.w.(Unknown Source)
at kdk.x.k(Unknown Source)
at kdk.dope.a.setSize(Unknown Source)
at kdk.dope.a.(Unknown Source)
at kdk.dope.a.(Unknown Source)
at kdk.dope.b.(Unknown Source)
at kdk.d.a(Unknown Source)
at qfp.Qfp.run(Unknown Source)
at java.lang.Thread.run(Thread.java:534)
qfp.Qfp:bye
which to me looks like…another broken obfuscator! Death to all obfuscators!
(so I reckoin the applet tag is fine, it’s just the code that is broken ;))
my tags were ok!
I put an unobfuscated version here
http://www.gutenachtkopf.de/savemodeforsafarionmacusers/
hope it helps
and replaced the Evil Polygon by the Evil Obfuscator
Still the CCE, but at least you should be able to tell where now ;D
Engine. NO PATHDATA
java.lang.ClassCastException
at kdk.ImageProdBuffered.(ImageProdBuffered.java:18)
at kdk.ImageProd.getImageProd(ImageProd.java:16)
at kdk.dope.Cannabis.setSize(Cannabis.java:246)
at kdk.dope.Cannabis.(Cannabis.java:33)
at kdk.dope.Cannabis.(Cannabis.java:45)
at kdk.dope.CannabisKwt.(CannabisKwt.java:13)
at kdk.Engine.init(Engine.java:139)
at qfp.Qfp.run(Qfp.java:181)
at java.lang.Thread.run(Thread.java:534)
qfp.Qfp:bye
PS you really need to put up some kind of progress indicator in your startup. For about 30 seconds after the applet has loaded it appears to do absolutely nothing, and the vast majority of people are likely to assume it’s broken if nothing happens once it reaches “applet started” which is the point at which your “qfp.Qfp:start()” message appears
(remember that almost no-one running an applet will have the console up as well, so they won’t see even this).
Hey there, well, as swpalmer indicated playing on OS X is almost impossible - the graphics just don’t work (I’ll try and post a screenshot later today). Anyway, I tried the game on my PC platform and was mildly surprised - whilst playing it I felt some of those xenon 2 feelings come back (just wish there was some of that bitmap brothers music in the background!). Played most levels and very much like the two -way scrolling idea.
It would be really nice if a webstart version was made of this game!
Keep up the good work!!!
java: as much platform independent as Donkey Kong?
the broken graphics on OS X:
please have a look at my reply #7
I think I just have to relace this:
image=gs[0].getDefaultConfiguration().createCompatibleImage(w, h);
because of the evil, evil
java.lang.ClassCastException: sun.awt.image.DataBufferNative
u are the experts!
hm… on my pc there is a startup time of about 42 ms after the applet has loaded (less than 400K). Must be OS dependent.
[quote] whilst playing it I felt some of those xenon 2 feelings come back (just wish there was some of that bitmap brothers music in the background!).
[/quote]
yeahhh!!! that’s what I want to hear! When I’ll have completed the game (in about 3 years), I ask Bomb the Bass for sound support
What is this webstart? Bla, could you please point me to a tutorial? Perhaps a tutorial by Kev?
I’ll do that soon, tried it last week but failed… (stupid me)
When it’s webstartable, I’ll have to put in features like jogl & fullscreen (would be really nice).
By now, 50% of the time is spent on System.arraycopy(), really uncool.
At the moment it won’t be impressive as an application, but as applet it is (I think)
per pixel rendering rules!
http://www.ribot.co.uk/images/screen_shot_xenon.gif
hmm,
OS X 10.2.4, Java 1.4.2_04 (Update 1 Developer Preview 2), ATI 9600
hello again
here is a webstart version
http://www.gutenachtkopf.de/webstarttest2/
would like to know if the webstart link works and if the mac-graphics bug ist still there
(leveleditor is still empty)
I put in the xenon2 megablast soundtrack from Bomb the Bass as a midi file.
I’ve been looking for a contact email of him to ask for permission to use it, but only found an address in Holand or so.
does anybody know about copyright issues concerning midi files (without vocals)?
Playing nice on Linux Red Hat 9 with 32fps
-Athlon XP 2000+
-256MB RAM
The only problem (in the web start version) was that it froze when I hit Esc (had to kill the process)
sorry, I didnt capture an exception. Now you should see a status message that says what went wrong (audioDeviceNotAvailable for example)
I think it is strange to play a game in windowed mode, and the window has no border
As for the ‘Esc’ problem, it is now gone. Now it shows the menu and the menu works fine.
Wow, looks great. The only problems:
on linux, appears in a small 640x480 window in top left of screen. A fullscreen window would have been nicer (my desktop res is 1400x1050)
generally very smooth but also slow - it auto disabled all options and yet it was still low fps. Playable, but you felt that you were missing out a bit. This is on a 1Ghz + geforce2, so it ought to be fine ?
Death to all obfuscators!
there is one main reason why its so slow:
but you can reduce the game size, so even Onyx with 500 mHz should reach high fps. A problem at such a small game size is that the gui components in the options screen will overlap.
And the collission detection won’t work for objects that have a pixel size<1
on my celeron laptop (1.7 ghz, 256 ddrram, NO 3D!) I get around 40 fps, on my desktop athlon 2400+ 512sdram+gf2 a litte less fps. (both xp)
So this game is for the mobile businessman with a fast cpu and no 3d card ;D
I think this way ist not the worst, this game will still run in 20 years on the next-generation cpus, I don’t think Doom3 will do, due to compatibility problems.
Is someone out there with a monster cpu (3gHz?) and no 3d card?
That’s a fine approach, it’s just that I’d expect it to be easily twice as fast on the quoted hardware. Are you using all the available tricks for accelerating 2D blitting etc in java?
no tricks
I write eveverything into a int[] buffer with my own routines for scaling, rotation, transparency,… it’s very flexible.
The only possible hw accelleration is when the int[] is drawn as a BufferdImage. When I draw nothing but the background at 800*600 I get max 60 fps. :-/
voila the cpu usage
you see, the pure blitting takes 20%, equal to the time System.arracopy (23.4% and 20.27% ) needs to copy 800*600 ints.
I’d REALLY like to hear that there is a faster way to get ints[] to the screen. I want to stick to ints[], because I love them and its easy and quick to alter the
Working well on OS X now (Java 1.4.2_05 Update 1) - around 40 fps on my 1Hz Powerbook.
But there doesn’t seem to be any way to exit the app?
WOW! Awesome
I can’t come up with any bugs or things that could get improved, since the only problem/feature(?) the game has is that it’s slow, and since you’r doing the graphics in your “special” way, there’s maybe nothing to do about it…
I got 40-120 fps (sound on, 800x600 I think) on WinXP, p4 2,4ghz, 256 ram, gf4.
[quote]no tricks
When I draw nothing but the background at 800*600 I get max 60 fps. :-/
[/quote]
:o
I’m pretty sure you can go much, much faster than that by being a bit more “accelerated-java2d-friendly” - but you’d have to dig around the J2D forum or ask someone like CrystalSquid or Abuse who might know. Personally, it’s been so long since I did pixel-array based java graphics I can’t reallly remember what you can and can’t do - and I was doing it before the BufferedImage etc were even invented, so I’m not much use to you
[quote]the memory bug vanished for some reason.
[/quote]
Nope, tis still there. I just played the obfuscator level, completed it, then tried to play the test level and it crashed the JVM with “OutOfMemoryError”.
You have a memory leak :).
Go get a profiler and find out where you’re leaking memory.
Yeah, it’s OS dependent I think. At least, I just tried on windows and it came up straight away (although this was the fullscreen webstart version I tried this time - nice).