Supplementry Notes for SpeedStr v.5.0
---------------------------------------------------------------

The legal stuff...

This software and the accompanying files are sold "as is" and 
without warranties as to performance of merchantability or any 
other warranties whether expressed or implied. Because of the 
various hardware and software environments into which SpeedStr 
may be put, no warranty of fitness for a particular purpose is 
offered.

Good data processing procedure dictates that any program be 
thoroughly tested with non-critical data before relying on it.  
The user must assume the entire risk of using the program. Any 
liability of the seller will be limited exclusively to product 
replacement or refund of purchase price.

Note that SpeedStr is free, so don't expect much in terms of 
"money back" guarantees.

Glad that's over...

Unfortunately the original documentation for SpeedStr arrived 
in the original Japanese, or at least as best as I could tell. 
Fortunately I had enough e-mail contact with the authours to 
figure it out.

Requirements for the test program:

þ 530,000 bytes free (DOS 5.0 or above, obviously)

þ SVGA video, 640x480-256 capability (they went "pretty" on 
  this one)

The benchtest program is easily launched... simply:

drive:\path> sped <enter>

That's all you need to do. Once launched you will be presented 
with a nicely designed screen. On the far left there is a 
textual information window. Just above it is an "off" button. 
To the center there is a mouse "well", and flanking that two 
rather elaborate looking buttons. Yes, you will need a mouse to 
run this thing. At the bottom of these items is a display 
screen.

The "buttons" all appear to be activated by a single left mouse 
button click; the right mouse button doesn't do a thing (what? 
designed on a Mac?). The originators went to great pains to try 
and convert the original Japanese text to a suitable English. I 
didn't have too many problems figuring it out.

A single mouse click on the off button and you exit. A single 
click on the sphere within a cage and you run the tests for a 
single session. A click on the sphere with the ring (sort of a 
Saturn?) allows you to run the tests for as many times as you 
wish, up to a total of nine.

Ah, yes the tests. It would appear that there are a total of five:

þ full rectangles, all 256 colours of the availalble palette

þ vertical lines, full palette sweep

þ random rectangles (random colours and sizes, it appears)

þ random single pixel lines (random colours and locations)

þ a weird pulsing thing (hard drive activity here... as well as 
  graphics' pumping?)

After all of the tests are done you are presented with the 
Zen-Beta Turn measurement of your system's potential; the 
benchmark result, if you will.

Although not presented I did find out that the maximum score is 
99. After 25 tests I personally averaged around 79 with my 
computer. According to my contacts in Japan this is a 
reasonably fast "score", with the best machines available there 
averaging in the high 80's and low 90's.

The test is remarkably clean in execution and seems to be very 
consistent.

My Japanese pals confided in me that inconsistent scores (high 
variability as an example) is a clear example of video card 
"cheating".

I can neither confirm nor deny this, but it is an interesting 
test none the less. I suggest you sit down some night and give 
it a try.

David Anjo
david.anjo@canrem.com