[Prev][Next][Index][Thread]

Re: Matt, anybody... E/L vs E/D?



>>>>> "Michelle" == Michelle Pankowski <d9060469@helios.usq.edu.au> writes:

    Michelle> Linux 2.0 is linux at its finest, all 32 bit and really
    Michelle> fast drivers, (tested with the fastest X server
    Michelle> available too).  Windows95 is all 16 bit with a .hfv
    Michelle> volume to slow thing down as well...

    Michelle> So why the hell is it so much faster in every damn area,
    Michelle> and would it completely destroy Linux when the VCPU and
    Michelle> a 32 bit port comes on line?

Going through X windows slows down Executor/Linux's graphics
substantially.  It's not really a fair comparison with Win95, because
under Win95 Executor takes over the entire screen.  Once we have a
native Win32 port (where Executor can exist in a window on your
desktop) the comparison will be more fair.

Your numbers showed much slower graphics under svgalib than under
Win95.  They should be about the same unless svgalib doesn't fully
support your video card.  What kind of video card do you have?

    Michelle> Maybe you should write a 16 bit version for Linux to see
    Michelle> if you could make it go faster. (Yes, I know, it was a
    Michelle> joke)

Just to be clear, Executor itself is entirely 32-bit on all platforms.
Win95 and DOS may have 16-bit drivers, but the CPU benchmarks
shouldn't be affected by what kind of drivers your system has (unless
you start paging or something).  The fact that they came out worse
under Linux suggests a multitasking-related artifact, or just random
variation.  Instructing Speedometer to run each test 10 times should
give more reliable numbers.  Also, I suggest you try the CPU
"benchmark suite" if you haven't already; that will average out many
benchmarks for you (the plain "CPU test" benchmark just bubble sorts
an array and isn't a great test).

Thanks for the relative performance numbers...

-Mat


References: