I'm with you on the free press. It's the newspapers I can't stand - Tom Stoppard
When we got the cards we plugged them in well know A8R-MVP, RD480 based motherboard. At this time it is the best RD480 motherboard around even some of the readers will disagree. Well I have to say that they didn't try the other solutions around which were even more troublesome for us. Asus RD480 based board is not perfect but it would run things nicely.
Asus EAX 1600XT Silent card comes in a nice box and first thing you will notice is that it has rather stylish
black heat pipe block over the GPU. There are two pipes leading from the GPU block to the back of the card where you
have a rather big radiator that has a task to dissipate the heath. Don't remember that Asus brands on front and the
back of the cards. It does a great job but the cards get rather hot. We would sacrifice that for the sounds of silence,
no question about it.
The card is powered with 256 MB of DDR 3 memory with 128 bit memory interface and knowing RV530 chip it has four pipelines but twelve pixel Shaders and it is fast enough to beat Geforce 6600 GT card but will lose from a 6800 GS. You will find a nice manual in the box, a driver CD, Asus DVD software and Xpand Rally game. Asus is also supplying a dongle with S video and composite ins and outs.
We were testing a single EAX1600XT Silent 587/1386 MHz versus two of those cards in Crossfire mode. We used
RD480 based A8R-MVP board.
Just before we were about to post this article we decided to check the Theo's claims that X1600 cards should run
faster on a new RD580 chipset. Since we had a chance to test A8R-32 Deluxe motherboard based on RD580 chipset we
decided to give it a try. He was right, those cards will run faster on the new RD580 boards.
Benchmarketing:
We used :
Asus EAX1600XT Silent 587/1386 MHz
Asus A8R-MPV Crossfire RD480 based motherboard
Asus A8R-32 MVP Deluxe RD580 based motherboard
Athlon FX 57 at 2.8 GHZ
2x1024 MB OCZ gold edition PC 4000 memory
Western Digital 8MB 160GB SATA drive
Akasa AK128 EVO Athlon 64 cooler
Akasa 650W multi rail PSU
OCZ Power Stream 600W PSU
You may notice that we used two power supply units but there is a big well kept secret behind the single rail versus dual rail PSU's and we will write about it soon enough. Let me just say that its not power supply unit manufacturers fault it's rather graphic and CPU manufacturers fault that those boards wont be stabile with ATX 2.0 or later multi rail PSUs.
| Doom 3 |
1024x768
|
1280x1024
|
1600x1200
|
2048x1536
|
| EAX1600XT Silent 587/1386 MHz A8R-MVP |
56.6
|
39.2
|
28.9
|
17.2
|
| EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
62.3
|
43.7
|
32.5
|
20.0
|
| Crossfire EAX1600XT Silent 587/1386 MHz A8R-MVP |
83.0
|
58.9
|
42.9
|
27.0
|
| Crossfire EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
90.3
|
69.5
|
51.1
|
32.3
|
| Doom 3 High Quality FSAA 4X + Aniso 8X |
1024x768
|
1280x1024
|
1600x1200
|
2048x1536
|
| EAX1600XT Silent 587/1386 MHz A8R-MVP |
38.9
|
27.9
|
19.0
|
9.0
|
| EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
46.1
|
32.0
|
21.3
|
9.6
|
| Crossfire EAX1600XT Silent 587/1386 MHz |
69.6
|
47.4
|
31.9
|
8.3
|
| Crossfire EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
77.2
|
54.3
|
36.3
|
8.8
|
| Far Cry |
1024x768
|
1280x960
|
1600x1200
|
2048x1536
|
| EAX1600XT Silent 587/1386 MHz A8R-MVP |
80.24
|
62.02
|
45.53
|
24.52
|
| EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
76.21
|
65.47
|
48.53
|
24.69
|
| Crossfire EAX1600XT Silent 587/1386 MHz |
82.87
|
68.84
|
52.73
|
31.25
|
| Crossfire EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
76.04
|
74.74
|
71.79
|
34.88
|
| Far Cry High Quality FSAA 4X + Aniso 8X |
1024x768
|
1280x1024
|
1600x1200
|
2048x1536
|
| EAX1600XT Silent 587/1386 MHz A8R-MVP |
50.60
|
37.29
|
25.49
|
8.45
|
| EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
53.36
|
39.52
|
26.45
|
10.42
|
| Crossfire EAX1600XT Silent 587/1386 MHz |
67.67
|
50.63
|
36.24
|
12.31
|
| Crossfire EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
73.36
|
56.71
|
41.18
|
17.84
|
| Half Life 2 |
1024x768
|
1280x960
|
1600x1200
|
2048x1536
|
| EAX1600XT Silent 587/1386 MHz A8R-MVP |
67.85
|
51.66
|
43.41
|
31.38
|
| EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
0
|
0
|
0
|
0
|
| Crossfire EAX1600XT Silent 587/1386 MHz |
70.84
|
58.46
|
48.58
|
35.37
|
| Crossfire EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
0
|
0
|
0
|
0
|
| Half Life 2 High Quality FSAA 4X + Aniso 8X |
1024x768
|
1280x1024
|
1600x1200
|
2048x1536
|
| EAX1600XT Silent 587/1386 MHz A8R-MVP |
46.53
|
30.26
|
24.63
|
15.55
|
| EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
0
|
0
|
0
|
0
|
| Crossfire EAX1600XT Silent 587/1386 MHz |
59.54
|
39.43
|
32.28
|
21.17
|
| Crossfire EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
0
|
0
|
0
|
0
|
| Quake 4 |
1024x768
|
1280x1024
|
1600x1200
|
2048x1536
|
| EAX1600XT Silent 587/1386 MHz A8R-MVP |
81.5
|
57.1
|
42.8
|
26.3
|
| EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
99.0
|
70.3
|
53.4
|
34.2
|
| Crossfire EAX1600XT Silent 587/1386 MHz |
113.2
|
78.5
|
58.7
|
37.8
|
| Crossfire EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
126.0
|
101.6
|
71.0
|
49.5
|
| Quake 4 High Quality FSAA 4X + Aniso 8X |
1024x768
|
1280x1024
|
1600x1200
|
2048x1536
|
| EAX1600XT Silent 587/1386 MHz A8R-MVP |
62.5
|
44.5
|
30.7
|
15.2
|
| EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
77.2
|
54.9
|
36.7
|
18.0
|
| Crossfire EAX1600XT Silent 587/1386 MHz |
97.5
|
68.1
|
47.2
|
25.5
|
| Crossfire EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
115.8
|
84.5
|
57.9
|
30.3
|
| FEAR |
1024x768
|
1280x1024
|
1600x1200
|
2048x1536
|
| EAX1600XT Silent 587/1386 MHz A8R-MVP |
43
|
30
|
20
|
12
|
| EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
53
|
38
|
26
|
14
|
| Crossfire EAX1600XT Silent 587/1386 MHz |
64
|
43
|
29
|
16
|
| Crossfire EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
72
|
48
|
32
|
19
|
| FEAR High Quality FSAA 4X + Aniso 8X |
1024x768
|
1280x1024
|
1600x1200
|
2048x1536
|
| EAX1600XT Silent 587/1386 MHz A8R-MVP |
23
|
17
|
11
|
5
|
| EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
37
|
27
|
17
|
7
|
| Crossfire EAX1600XT Silent 587/1386 MHz |
40
|
26
|
21
|
8
|
| Crossfire EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
43
|
29
|
26
|
14
|
| Battlefield 2 |
1024x768
|
1280x1024
|
1600x1200
|
2048x1536
|
| EAX1600XT Silent 587/1386 MHz A8R-MVP |
62.73
|
45.57
|
34.10
|
22.58
|
| EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
66.53
|
47.95
|
40.99
|
24.69
|
| Crossfire EAX1600XT Silent 587/1386 MHz |
72.71
|
54.46
|
42.07
|
24.65
|
| Crossfire EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
81.34
|
64.09
|
50.41
|
34.79
|
| Battlefield 2 High Quality FSAA 4X + Aniso 8X |
1024x768
|
1280x1024
|
1600x1200
|
2048x1536
|
| EAX1600XT Silent 587/1386 MHz A8R-MVP |
44.91
|
34.05
|
27.40
|
18.24
|
| EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
50.38
|
35.90
|
29.01
|
20.12
|
| Crossfire EAX1600XT Silent 587/1386 MHz |
62.45
|
47.25
|
37.45
|
22.96
|
| Crossfire EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
69.92
|
54.58
|
44.60
|
25.49
|
| Composite Figures 3Dmark 03 |
3DMark 03
|
Game2
|
Game3
|
Game3
|
| EAX1600XT Silent 587/1386 MHz A8R-MVP |
9331
|
66.5
|
55.2
|
62.9
|
| EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
9465
|
66.8
|
56.7
|
64.2
|
| Crossfire EAX1600XT Silent 587/1386 MHz |
11610
|
89.1
|
76.1
|
83.2
|
| Crossfire EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
13706
|
103.3
|
86.6
|
97.1
|
| Composite Figures 3Dmark 03 |
Single Texturinng
|
Multi Textur.
|
Vertex Shader
|
Vertex Shader
|
| EAX1600XT Silent 587/1386 MHz A8R-MVP |
1868.2
|
2310.9
|
39.4
|
65.5
|
| EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
1868.4
|
2314.3
|
41.3
|
72.6
|
| Crossfire EAX1600XT Silent 587/1386 MHz |
3177.1
|
3770.5
|
60.9
|
98.7
|
| Crossfire EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
3482.6
|
4205.6
|
67.3
|
114.8
|
| Composite Figures 3Dmark 05 |
3DMark 05
|
Game1
|
Game2
|
Game2
|
| EAX1600XT Silent 587/1386 MHz A8R-MVP |
5108
|
23.9
|
14.6
|
24.5
|
| EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
5458
|
25.1
|
15.6
|
26.6
|
| Crossfire EAX1600XT Silent 587/1386 MHz |
8777
|
40.4
|
26.8
|
40.0
|
| Crossfire EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
9484
|
42.1
|
28.6
|
45.3
|
| Composite Figures 3Dmark 05 |
Single Texturinng
|
Multi Textur.
|
Pixel Shader
|
VS/VS
|
| EAX1600XT Silent 587/1386 MHz A8R-MVP |
2028.7
|
2351.6
|
113.4
|
73.3/47.0
|
| EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
2061.6
|
2351.6
|
118.7
|
74.2/47.1
|
| Crossfire EAX1600XT Silent 587/1386 MHz |
3241.2
|
3806.4
|
148.0
|
123.7/87.4
|
| Crossfire EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
3807.8
|
4266.8
|
186.6
|
127.7/91.8
|
| Composite Figures 3Dmark 06 |
3DMark 06
|
|||
| EAX1600XT Silent 587/1386 MHz A8R-MVP |
2210
|
|||
| EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
2336
|
|||
| Crossfire EAX1600XT Silent 587/1386 MHz |
3741
|
|||
| Crossfire EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
3933
|
|||
| SM2.0 Test |
Score
|
GT1
|
GT2
|
|
| EAX1600XT Silent 587/1386 MHz A8R-MVP |
857
|
6.693
|
7.580
|
|
| EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
905
|
7.092
|
7.990
|
|
| Crossfire EAX1600XT Silent 587/1386 MHz |
1625
|
12.707
|
14.369
|
|
| Crossfire EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
1698
|
13.270
|
15.035
|
|
| HDR/SM3.0 Test |
Score
|
HDR1
|
HDR2
|
|
| EAX1600XT Silent 587/1386 MHz A8R-MVP |
855
|
7.493
|
9.614
|
|
| EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
920
|
8.284
|
10.116
|
|
| Crossfire EAX1600XT Silent 587/1386 MHz |
1625
|
14.577
|
17.937
|
|
| Crossfire EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
1721
|
15.263
|
19.151
|
|
| CPU Test |
Score
|
CPU1
|
CPU2
|
|
| EAX1600XT Silent 587/1386 MHz A8R-MVP |
1085
|
0.342
|
0.550
|
|
| EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
1083
|
0.342
|
0.550
|
|
| Crossfire EAX1600XT Silent 587/1386 MHz |
1033
|
0.334
|
0.511
|
|
| Crossfire EAX1600XT Silent 587/1386 MHz A8R32-MVP Deluxe |
1083
|
0.342
|
0.550
|
We don't need to tell you that two cards will end up faster than a single card. But we have to say that in Doom 3 a single card on A8R-32 MVP Deluxe runs three to five frames faster. The same applies for RD580 crossfire as it will outperform the old motherboard by as much as 10 frames at certain resolutions. This definitely is a nice performance increase.
Once you turn the FSAA the performance gap becomes even bigger. A single card on RD580 will run up to eight frames faster while in Crossfire, RD580 runs up to eight frames faster. The Far Cry is on the RD580 side with an exception that this board will lose in 1024x768 scores. In all the other scores RD580 will be at least three frames faster while at 1600x1200 it manages to be astonishing 19 frames faster. The same performance ratio occurs when you turn FSAA and Aniso on. The RD580 is simply faster.
Half Life 2 was the hard one. We could not test this one at RD580 board, it would just act weird. We are working to find out what is going on with this test but even if we wanted to play Half Life 2 or Counter Strike source on the machine the game would stop every few seconds and then will happily continue. This really mess your FPS scores and we decided no to write the scores down. We suspect it is the power issue but we are not sure now.
Quake 4 again likes RD580, It runs 18 frames faster with a single card at 10x7 or 13 frames faster in Crossfire
and the Crossfire can show the difference of the astonishing 22 frames at 1280x1024 or 12 frames at 1600x1200 or
2048x1536. Even the single card scores significant 10 FPS faster in most of the cases. Quake 4 FSAA and Aniso on will
confirm the previous scores; it will just run significantly faster in every single case.
FEAR also likes Asus A8R-32 MVP Deluxe motherboard. A single card scores up to 10 frames faster while the Crossfire mode scores up to 8 frames. RD580 is constantly faster than RD480. Once you turn FSAA and 16X Anisotropic filtering the scores gets closer but you still see that RD580 is clearly faster. It is also faster in every single scenario in Battlefield 2.
Conclusion
We like the fact that mainstream crossfire setup can be completely silent. Asus made a great motherboard based on
RD580 chipset while it certainly won out hearts with two X1600XT cards with heath pipes. Asus A8R-32 MVP Deluxe works
well and has some high end features. We tested this board with two X1800 and X1900 cards in Crossfire as well, and we
will tell you all about the boards feature in that separate part.
We were shocked to learn that RD580 will end up faster significantly than RD480 as once we moved from Nforce 4 SLI to Nforce 4 SLI 32 PCIe lanes we only got some marginal performance increase. I guess ATI did a great job with RD580 and it works much better than RD480 and partners like Asus decided to do a high end board based on the design. ATI's Northbridge features 36 lanes while Nvidia has 20 lanes in Northbridge and twenty in Southbridge.
ASUS Extreme AX1600XT SILENT/TVD, Radeon X1600 XT, 256MB GDDR3 cards will cost your around 199 or roughly £149 and is a great way to make your Crossfire system silent. Akasa EVO 120 cooler helped us out in that task and for roughly 600 for two graphic cards and a motherboard you got a very decent performance rig. It won't have any power or stability issues. Let me remind you that a single X1900 XTX card will cost roughly the same and the performance will be very close, plus you need to spend additional 200 for the motherboard. The choice is always yours but we like this silent way of doing a dual card Crossfire. µ
Tested by: Sanjin Rados and Fuad Abazovic