OT (now really): Re: Re: Re: [Mplayer-users] Gibraltar

Christoph Lampert lampert at math.chalmers.se
Tue Aug 28 19:25:00 CEST 2001

>> Btw. have they finally come up with an option to switch off runtime-checking 
>> (array boundaries etc.) in Java? People demanded that for a long time 
>> (High-Performance-Java), but I never heard of it actually been accepted. 
>It's done - in M$ java engine :)
>So running a buggy java applet in MSIE or windoze netscape can crash the
>whole house (with all windows broken).

Sh*t, actually I like that. Of course, there has to be a switch for it, 
but once my code runs without broken pointers (haha, I mean, without arrays out 
of bounds), there should not be a check for it every time I access memery. 
Fine with me.

>> And: No native libs! That's cheating. Even I can write a Java program 
>> that calls a native lib called "mplayer" with the correct arguments...
>Of course. I meant a lib only to do mmx/sse accelerated idct/yuv stuff.
>But they must have java implementation too! (it's just an option to
>boost speed)
>Why are no MMX-like instructions in java-asm instruction set?
>They should introduce SIMD extensions for java...

How about inline assembler? That would do the trick :-)
Btw. is there "inline java bytecode"? That would be cool, too...


Dipl. math. Christoph Lampert (complex analysis, integral formulae)
Email: gruel at gmx.de                |     Email: lampert at math.chalmers.se

Mplayer-users mailing list
Mplayer-users at lists.sourceforge.net

More information about the MPlayer-users mailing list