Executor 2.0/executorv2.0.iso/pc/linux/extra/docs/maillist/text/archive.96/text1131.txt

Text File | 1996-07-25 | 1.6 KB | 30 lines

I was benchmarking Executor versus some real Mac's (Performa 637CD for
instance) the other day, using both VBE 1.2 as my #9 FX Motion 771 (4MB
VRAM) does natively and with UniVBE 5.1 (unregistered so far) for 2.0
compliance, and found that while my graphics numbers improved greatly
under Speedometer 3.23 (they almost doubled in 2-bit mode, and close to it
also in 4-bit & 8-bit) there was nowhere near the speedup this card is
good for elsewhere.  Am I correct to assume that while UniVBE provides
emulation of 2.0 BIOS extensions, it gives no access at all the the
hardware acceleration features of the #9 Video card?  If this is so, would
a Win32 version using DirectDraw when available work with drivers that DO
use all the hardware acceleration available, and would this in turn lead
to even greater graphics performance?

Furthermore (and I know this gets kinda murky) would the lessened time
Executor spends doing graphics overhead lead to increased performance in
other quarters of the emulations, ie enhanced CPU and math throughput
maybe?  Also, the docs, and the mailinglist a while ago, got on the
subject of the rough instruction ratio Executor encounters going from 68k
to x86 code.  I have seen stated that it is pretty consistently 3:1 but
that it was hoped in going to a certain new kind of dynamic compilation we
could get to maybe 2:1 and I was wondering if anything more is known about
such lofty prospects at this time.

Executor is awesome!!  My Mac friend was VERY impressed to see what I can
do without Apple hardware and said SoftWindows on the PPC's he has at work
does nowhere near as good a job!!  Best of luck guys....

G.M.