home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
PC Gamer (Italian) 30
/
PC Gamer IT CD 30 1-2.iso
/
BOOTMARK
/
X
/
XBench.txt
< prev
Wrap
Text File
|
1997-11-14
|
5KB
|
163 lines
!!!!!! IMPORTANT NOTES !!!!!!!
Using X as a benchmark for 3D adapters.
The current demo of X can be run to safe a profile of the number of
frames per second that it was able to archive for every single cutscene
of the demo.
Since the different scenes use different amounts of resources (like
transparent faces, vertices, renderstate changes) the differences
between the certain scenes can give you a very precise impression
of what a 3d accelerator can do.
During a run of the "X-Demo" in benchmark mode, you can not forward
or rewind scenes like you can (with "space" and "B") in normal runs.
You should not stop the program otherwise during a benchmark run, or
you will influence the benchmark results.
Results of the benchmark are of course dependent of the chosen
resolution. X currently defaults to 800*600 for ALL adapters.
This will change for the final game, since many older 3D
accelerators will still deliver very good results with
resolutions like 400*300 or 512*384.
Benchmarks will give different results with the same resolution
in window and fullscreen mode!
To compare different accelerators running in window mode (where
possible) might help reducing the "round off" through frame
synchronization that happens in fullscreen mode.
To be able to not only compare the speed of the different cards
but also the image quality produced, you can turn on the
screenshot capture option. If you have this option turned on
X will try to safe one screenshot per scene of the demo to
the directory "c:\cap". This directory has to exist for capturing
to work!
The benchmarking results will be saved to the file "c:\xperf.txt".
You can use those numbers easily to import them into a spreadsheet
program and generate charts to compare the results.
The performance relevant settings will be printed in this file too.
However just because a feature is turned ON does not necessarily
mean that it is (fully) supported by the hardware. So to make
REALLY sure people can compare the results best you can do is
reprint screenshots (yes we are not unhappy about this fact :)
Do not safe screenshots in the same run you want to do
a reliable benchmark. The saving will influence the speed
of the program.
Also note that the 3D sound effects can slow down the overall
framerate a LOT. So do not compare results with 3D sounds turned
on to results taken without 3d sounds.
Since you can not change the setting of the X Demo during
a benchmark run, you have to start the demo in normal mode
first, change the settings by pressing F11 or F12
(e.g: Turn off 3D sounds) then leave the demo (ESC)
and start it in benchmark mode again (this time NOT ignoring
the settings which were saved in the registry!)
Feel free to print screenshots and results from X in any
publications. But please mention the name of the game and
the egosoft homepage URL with it!
Please write us if you use and like X for the purpose
of testing 3D accelerators and stay tuned for the final
version of the game!
If you wish to make the demo available to other people or
put it on a CD, please contact us!
Keyboard:
"Esc" - leave the demo
"Space" - skip to next scene (not in benchmark)
"b" - go back to last scene (not in benchmark)
"F11" - Graphic and 3D settings
"F12" - Sound settings
For questions or bug reports contact: egoinfo@egosoft.com
---------------------------------------------------------
Sample Output of the X demo on my machine.
---------------------------------------------------------
X benchmark output.
Executable compiled: Fri Nov 14 06:51:18 1997
Visual Quality settings during test:
Screen resolution: 800*600
Transparency through ONE ONE blending! *BEST*
Edgeantialiasing OFF (not supported by most cards anyway)
Filtering ON
Dithering OFF
Other performance relevant settings:
All optimizations of X turned ON
3D Soundeffects were turned on (can considerably slow down execution)
(c)1997 EGOSOFT
For more information http://www.egosoft.com
Scene 00: 80.760fps
Scene 01: 111.578fps
Scene 02: 67.921fps
Scene 03: 43.077fps
Scene 04: 48.785fps
Scene 05: 35.119fps
Scene 06: 43.457fps
Scene 07: 44.395fps
Scene 08: 37.193fps
Scene 09: 111.844fps
Scene 10: 42.791fps
Scene 11: 57.522fps
Scene 12: 44.783fps
Scene 13: 32.398fps
Scene 14: 54.524fps
Scene 15: 55.356fps
Scene 16: 32.280fps
Scene 17: 43.146fps
Scene 18: 47.690fps
Scene 19: 37.828fps
Scene 20: 113.977fps
Scene 21: 46.158fps
Scene 22: 36.157fps
Scene 23: 50.487fps
Scene 24: 33.647fps
Scene 25: 25.145fps
Scene 26: 20.132fps
Scene 27: 57.913fps
Scene 28: 60.064fps
Scene 29: 61.569fps
Scene 30: 80.923fps
Scene 31: 35.516fps
Scene 32: 54.022fps
Scene 33: 42.462fps
Scene 34: 49.385fps
Scene 35: 33.024fps
Scene 36: 109.913fps
Scene 37: 53.642fps
Scene 38: 39.590fps
Scene 39: 61.231fps
Scene 40: 59.816fps
Scene 41: 75.040fps
Scene 42: 68.931fps
Scene 43: 64.874fps
Scene 44: 72.149fps
Scene 45: 58.956fps
Scene 46: 44.896fps
Scene 47: 46.483fps
Overall average framerate: 54.761fps
X and the X demo and benchmark are copyrighted 1997
Ego Software, Herzogenrath, Germany
and
Funsoft, Kaarst, Germany