Single card @ 90 MHz (default)
results
|
640x480
|
800x600
|
GLQuake
Demo2 |
103.1 |
66.4 |
Quake II
Demo1 |
88.5 |
60.1 |
Quake II
Crusher |
44.3 |
36.6 |
Unreal timedemo
.2 |
49.81 |
27.31 |
Metabyte decided to include
an overclocking utility with the card. Without
any cooling the card reached 100 MHz without problems
in all the games.
Single card @ 100 MHz
(overclocked) results
|
640x480
|
800x600
|
GLQuake
Demo2 |
111.3 |
72.8 |
Quake II
Demo1 |
90.1 |
65.5 |
Quake II
Crusher |
45.0 |
38.1 |
Unreal timedemo
.2 |
51.56 |
29.82 |
STB Black Magic 3D @ 90
MHz vs. Wicked3D Voodoo2 @ 90 MHz
STB
Black Magic 3D |
Wicked
3D Voodoo2 |
Dark
Blue |
Red |
GLQuake Demo2 - 640x480
Quake II Demo1 - 640x480
Quake II Crusher -
640x480
Unreal - 640x480
|
GLQuake Demo2 - 800x600
Quake II Demo1 - 800x600
Quake II Crusher -
800x600
Unreal - 800x600
|
When these two cards are
placed against each other, the results show almost
nothing. Since the cards use different types of
memory (as mentioned earlier: 90 MHz EDO ram on
the Black Magic, 100 MHz EDO ram on the Wicked3D)
it seems pointless to run them at the same speed.
Without cooling, the STB
Black Magic could not run Unreal long enough to
do a timedemo when being overclocked. Since the
Wicked3D did it with no problems, it can be compared
@ 100 MHz to the Black Magic @ 90 MHz. In the
custom of Tweak3D, here are the results of that
test.
STB Black Magic 3D @ 90
MHz vs. Wicked3D Voodoo2 @ 100 MHz
STB
Black Magic 3D |
Wicked
3D Voodoo2 |
Dark
Blue |
Red |
GLQuake Demo2 - 640x480
Quake II Demo1 - 640x480
Quake II Crusher -
640x480
Unreal - 640x480
|
GLQuake Demo2 - 800x600
Quake II Demo1 - 800x600
Quake II Crusher -
800x600
Unreal - 800x600
|
With the Wicked3D Voodoo2
running at 100 MHz, it beats the STB card by a
slight margin. If you take into account that the
card can be overclocked easier, the card is
faster than the others.
Conclusion