 |
Untitled Document
Unreal Tournament 2003
I will use the Botmatch tests here, as they are much more indicative of real
game performance. I am using the Citadel botmatch benchmark, because it is by
far the most strenuous UT2003 benchmark, scoring average frame rates at least
20fps below the scores the other benchmarks give. The first table contains the
average frame rate with all the detail settings set to the lowest settings,
in 16 bit colour, and the second table has the results with all the detail settings
at maximum, in 32 bit colour.
To make the tables easier to read, all results under 50 fps are shown in red,
to represent an unsatisfactory score, and scores between 50 and 80 fps are shown
in orange, and any score over 80fps is shown in green. The margin of error is
only around .3 FPS.
| UT2003 Citadel Botmatch Low Detail
16 bit Colour |
| |
Sparkle 5900 |
Gainward 5900 |
Sparkle 5600 |
Terratec 5200 |
Terratec 5600 |
Sapphire 9600 pro |
Asus 9600 XT |
Asus 9800 XT |
| 320x240 |
91
fps |
91
fps |
91
fps |
91
fps |
91
fps |
91
fps |
91
fps |
91
fps |
| 512x384 |
91
fps |
91
fps |
91
fps |
88
fps |
91
fps |
91
fps |
91
fps |
91
fps |
| 640x480 |
90
fps |
91
fps |
90
fps |
83
fps |
91
fps |
91
fps |
91
fps |
91
fps |
| 800x600 |
90
fps |
91
fps |
90
fps |
72
fps |
90
fps |
91
fps |
91
fps |
91
fps |
| 1024x768 |
91
fps |
90
fps |
84
fps |
53
fps |
84
fps |
90
fps |
91
fps |
91
fps |
| 1152x864 |
90
fps |
91
fps |
78
fps |
45
fps |
77
fps |
89
fps |
90
fps |
92
fps |
| 1280x960 |
90
fps |
90
fps |
71
fps |
38
fps |
70
fps |
86
fps |
89
fps |
92
fps |
| 1280x1024 |
92
fps |
92
fps |
62
fps |
32
fps |
61
fps |
81
fps |
86
fps |
93
fps |
| 1600x1200 |
88
fps |
87
fps |
73
fps |
26
fps |
51
fps |
68
fps |
77
fps |
90
fps |
| UT2003 Citadel Botmatch High Detail
32 bit Colour |
| |
Sparkle 5900 |
Gainward 5900 |
Sparkle 5600 |
Terratec 5200 |
Terratec 5600 |
Sapphire 9600 pro |
Asus 9600 XT |
Asus 9800 XT |
| 320x240 |
57
fps |
57
fps |
57fps |
57
fps |
57
fps |
57
fps |
57
fps |
57
fps |
| 512x384 |
57
fps |
57
fps |
57
fps |
53
fps |
57
fps |
57
fps |
57
fps |
57
fps |
| 640x480 |
57
fps |
56
fps |
57
fps |
48
fps |
56
fps |
57
fps |
57
fps |
57
fps |
| 800x600 |
57
fps |
57
fps |
55
fps |
38
fps |
55
fps |
57
fps |
57
fps |
57
fps |
| 1024x768 |
57
fps |
56
fps |
48
fps |
26
fps |
47
fps |
56
fps |
56
fps |
56
fps |
| 1152x864 |
56
fps |
56
fps |
42
fps |
22
fps |
41
fps |
55 fps |
56
fps |
57
fps |
| 1280x960 |
56
fps |
56
fps |
37
fps |
18
fps |
35
fps |
52
fps |
55
fps |
56
fps |
| 1280x1024 |
57
fps |
57
fps |
33
fps |
17
fps |
32
fps |
51
fps |
55
fps |
59
fps |
| 1600x1200 |
54
fps |
54
fps |
25
fps |
12
fps |
24
fps |
37
fps |
42
fps |
56
fps |
And now putting these scores into perspective, here are the results from the
Anubis botmatch, a rather less taxing UT2003 benchmark.
| UT2003 Anubis Botmatch Low Detail
16 bit Colour |
| |
Sparkle 5900 |
Gainward 5900 |
Sparkle 5600 |
Terratec 5200 |
Terratec 5600 |
Sapphire 9600 pro |
Asus 9600 XT |
Asus 9800 XT |
| 320x240 |
117
fps |
118
fps |
118
fps |
117
fps |
118
fps |
118
fps |
118
fps |
119
fps |
| 512x384 |
117
fps |
118
fps |
117
fps |
116
fps |
118
fps |
118
fps |
118
fps |
119
fps |
| 640x480 |
117
fps |
118
fps |
117
fps |
113
fps |
116
fps |
117
fps |
118
fps |
119
fps |
| 800x600 |
117
fps |
117
fps |
117
fps |
104
fps |
117
fps |
117
fps |
118
fps |
119
fps |
| 1024x768 |
117
fps |
117
fps |
113
fps |
83
fps |
112
fps |
117
fps |
118
fps |
119
fps |
| 1152x864 |
117
fps |
117
fps |
107
fps |
71
fps |
106
fps |
117
fps |
118
fps |
118
fps |
| 1280x960 |
117
fps |
116
fps |
98
fps |
60
fps |
96
fps |
113
fps |
117
fps |
118
fps |
| 1280x1024 |
117
fps |
117
fps |
96
fps |
57
fps |
94
fps |
112
fps |
118
fps |
119
fps |
| 1600x1200 |
117
fps |
116
fps |
52
fps |
41
fps |
70
fps |
83
fps |
97
fps |
118
fps |
|
UT2003 Anubis Botmatch High Detail
32 bit Colour |
| |
Sparkle 5900 |
Gainward 5900 |
Sparkle 5600 |
Terratec 5200 |
Terratec 5600 |
Sapphire 9600 pro |
Asus 9600 XT |
Asus 9800 XT |
| 320x240 |
91 fps |
91 fps |
91 fps |
91 fps |
91 fps |
91 fps |
91 fps |
91 fps |
| 512x384 |
91 fps |
91 fps |
90 fps |
87 fps |
90 fps |
91 fps |
91 fps |
91 fps |
| 640x480 |
91 fps |
91 fps |
91 fps |
80 fps |
90 fps |
91 fps |
91 fps |
91 fps |
| 800x600 |
92 fps |
91 fps |
89 fps |
66 fps |
89 fps |
91 fps |
90 fps |
91 fps |
| 1024x768 |
91 fps |
91 fps |
78 fps |
47 fps |
75 fps |
90 fps |
90 fps |
91 fps |
| 1152x864 |
91 fps |
91 fps |
68 fps |
39 fps |
64 fps |
87 fps |
90 fps |
92 fps |
| 1280x960 |
90 fps |
91 fps |
58 fps |
32 fps |
55 fps |
79 fps |
86 fps |
91 fps |
| 1280x1024 |
85 fps |
85 fps |
52 fps |
29 fps |
49 fps |
71 fps |
78 fps |
86 fps |
| 1600x1200 |
88 fps |
87 fps |
40 fps |
21 fps |
38 fps |
53fps |
61 fps |
90 fps |
As you can see, each benchmark appears to have an effective maximum frame rate,
which is most likely set by the limitations of the computer itself, as the botmatch
benchmarks involve running the game AI as well as just rendering the picture.
This makes these results a better indication of actual game frame rates, even
if they do not shed any light on the small differences in performance between
the faster cards. As the almost unchanging results from the FX 5900 cards and
the 9800 XT show, these high end cards are overkill for Unreal Tournament 2003.
<< Back | 1 | | 2 | | 3 | | 4 | | 5 | | 6 | | 7 | | 8 | | 9 | | 10 | Next >>
|
|
