|
Crysis: Warhead:
By now, everybody who is anybody knows about
Crysis, and its sequel, Crysis: Warhead. So why don't we skip
the pleasantries and get right to the torture.
Both Single GPU and Dual GPU setups ran through
the same sequence, but in order to keep things organized, I
separated out the data. Each run consisted of multiple
resolutions and AA settings, so that we could get a good feel as to
how well the cards scaled. All other settings were maxed out.
Click the pictures for a larger version. First up, the single
GPUs:
Single GPU:
Benchmark Data:
|
4870 Toxic |
4890 |
|
Min |
Max |
Avg |
Min |
Max |
Avg |
| 1280, 0x |
20.61 |
38.07 |
30.76 |
25.34 |
42 |
33.95 |
| 1280, 2x |
14.95 |
33.17 |
27.08 |
21.23 |
35.89 |
29.53 |
| 1280, 4x |
15.78 |
33.78 |
26.79 |
19.17 |
37.83 |
29.44 |
| 1280, 8x |
13.98 |
31.77 |
24.86 |
11.53 |
35.28 |
26.94 |
| 1680, 0x |
17.69 |
30.05 |
24.9 |
19.86 |
33.99 |
27.2 |
| 1680, 2x |
11.94 |
25.61 |
21.56 |
16.63 |
28.78 |
23.55 |
| 1680, 4x |
14.7 |
26.93 |
21.78 |
15.85 |
29.62 |
23.55 |
| 1680, 8x |
12.47 |
25.52 |
20.14 |
12.86 |
28.03 |
21.71 |
| 1920, 0x |
14.69 |
26.06 |
20.92 |
15.9 |
27.66 |
22.76 |
| 1920, 2x |
11.99 |
21.88 |
18.05 |
13.74 |
25 |
19.68 |
| 1920, 4x |
11.77 |
22.3 |
18.22 |
12.5 |
24.62 |
19.76 |
| 1920, 8x |
9.47 |
21.57 |
16.75 |
10.34 |
23.43 |
17.93 |
| 2560, 0x |
7.62 |
16.15 |
13.06 |
9.21 |
17.39 |
14.24 |
| 2560, 2x |
6.51 |
14.32 |
11.2 |
5.83 |
16.89 |
12.12 |
| 2560, 4x |
4.81 |
13.9 |
10.85 |
5.04 |
15.05 |
11.7 |
| 2560, 8x |
1.29 |
6.81 |
3.95 |
0.69 |
6.06 |
3.52 |
As we can see from the above data, the sweet spot
for these cards is in the 1680x1050 resolution. The 4890
consistently bests the 4870 Toxic by roughly 8%.
Dual GPU:
Benchmark Data:
|
4870X2 |
4890 Crossfire |
|
Min |
Max |
Avg |
Min |
Max |
Avg |
| 1280, 0x |
20.81 |
46.71 |
35.92 |
27.41 |
59.22 |
45.94 |
| 1280, 2x |
19.76 |
43.33 |
32.92 |
25.3 |
53.5 |
41.89 |
| 1280, 4x |
19.87 |
43.26 |
33.06 |
25.68 |
54.3 |
41.96 |
| 1280, 8x |
18.96 |
41.9 |
31.07 |
24.83 |
52.06 |
39.78 |
| 1680, 0x |
17.76 |
40.19 |
31.4 |
22.71 |
50.56 |
39.66 |
| 1680, 2x |
15.63 |
35.84 |
28.49 |
21.56 |
44.68 |
35.42 |
| 1680, 4x |
15.54 |
36.26 |
28.5 |
21.48 |
45.28 |
35.62 |
| 1680, 8x |
13.88 |
34 |
26.32 |
20.18 |
44.32 |
33.59 |
| 1920, 0x |
14.65 |
34.65 |
27.83 |
19.93 |
43.05 |
34.09 |
| 1920, 2x |
14.53 |
31.15 |
25.17 |
18.75 |
38.67 |
30.75 |
| 1920, 4x |
14.58 |
31.74 |
24.97 |
17.86 |
39.55 |
30.84 |
| 1920, 8x |
13.61 |
29.08 |
23.18 |
17.76 |
38.32 |
28.85 |
| 2560, 0x |
11.35 |
25.23 |
20.02 |
13.37 |
30.05 |
23.52 |
| 2560, 2x |
9.37 |
20.93 |
17.43 |
12.65 |
25.76 |
20.67 |
| 2560, 4x |
5.33 |
21.3 |
16.68 |
6.63 |
25.98 |
19.85 |
| 2560, 8x |
0.68 |
8.64 |
5.04 |
0.83 |
9.28 |
5.27 |
This was a much bigger surprise. The 4890
Crossfire pretty well obliterated the 4870X2, for the most part
gaining 17-28%. Playability was maintained all the way through
the 1920x1200 resolution. I am impressed.
|
