home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Aminet 50
/
Aminet 50 (2002)(GTI - Schatztruhe)[!][Aug 2002].iso
/
Aminet
/
util
/
boot
/
Blazewcp.lha
/
wpatest.guide
(
.txt
)
< prev
next >
Wrap
Amigaguide Document
|
2000-07-21
|
2KB
|
31 lines
You'll have to forgive this crude guide but there's not much to say about
WPATest. WPATest test the worst case (small and completely non-aligned) and
the best case (large and 32bit aligned) for WritePixelLine8(), WritePixelArray8()
and on OS 3.1 or above WriteChunkyPixels(). WPATest is quite simple to use, but please
read this whole document before using it. WPATest wasn't mean to be shiped to
the general public, which is why I'm supprised that Stephen Brookes allowed me to
include it, thanks Stephen. The only major bug that I know of in WPATest is if you
try and run it on a screen smaller the 640 x 400 it'll crash. This is due to the fact
that it needs a big screen for its window, but doesn't check to see how big the
screen is when it opens its window. I'm sure there are more tho. Aside from that
WPATest is really simple to use. I'll explaine everything below, it's only 4 buttons.
WPA8/WPL8/WCP : This is a cycling button that allows you to pick WritePixelArray8()
WritePixelLine8() or WriteChunkyPixels() to run testes on.
Layers: This test how well the selceted function handels layered operations.
BlazeWCP passes this function.
Masking: This tests whether the selected function supports rastports with masked planes.
BlazeWCP passes this function.
> There is a pause after hitting the Speed button before the test is visably run <
> and is normal. <
Speed: This should be pretty obvious. This runs a small, non-aligned (worst case) test
and a large 32bit aligned (best case) test on the selected function and reports
back the number of thousands of pixels a second that were plotted. It also gives a guess
as to how much data had to be moved and reports this in thousands of bytes a second and how
efficient the routines were at converting to chiipram as compared to just writing to chip.
The % number it returns is kindof acurate but not completely since it compares the speed of
the function reading fastram...converting...writing to chipram to just writing to chipram.
The test screen we use for our test is a 640 x 512 8bit hi-res interlaced PAL screen. WPATest will
also tell if the tested function reports back the wrong numbers of pixels plotted. PatchWPA8 is
the only one I know of that reports this wrong.
BlazeWCP REALLY passes this function ;)