ABCDEGHIJKLMNOPQRSTUVWXYZ
1
項目OverallStandardC4DTracking備註
Kadid works.
2
CPU頻率測試
3
預設1006.099.697.8105.4
4
4.6GHz1002.5100.695.1104.7
5
4.8Ghz1037.3103.2101.6106.9
6
5.0Ghz1067.0106.6104.1109.5
7
5.2Ghz1089.0109.4105.2111.6
8
RAM通道及容量測試 (鎖定2133)
9
1Channel單通16x1=16G815.875.683.991.2
10
2Channel雙通 8x2=16G944.893.992.997.2
11
2Channel雙通 8x4=32G943.092.893.598.1
12
2Channel雙通16x2=32G947.592.993.699.6
13
RAM頻率測試 (鎖定CL16-18-18-38)
14
2133943.392.691.4100.7
15
2400962.593.892.9104.5
16
2666986.597.195.0105.4
17
30001005.099.297.5106.1
18
32001014.3100.796.5107.8
19
35001020.3102.398.3105.2
20
RAM時序測試
21
2666 (CL16-18-18-38)986.597.195.0105.4
22
2666 (CL12-18-18-38)986.097.196.4103.8
23
2666 (CL12-14-14-28)990.898.695.5103.6
24
XMP及內顯測試
25
內顯+off XMP (2133 CL15-15-15-36) 893.878.495.0105.7
26
內顯+on XMP (3200 CL16-18-18-38)923.384.698.0102.1
27
CUDA+off XMP (2133 CL15-15-15-36) 951.593.593.6100.0
28
CUDA+on XMP (3200 CL16-18-18-38)1006.099.697.8105.4
29
極限測試
30
5.2Ghz Auto+3600 16-19-19-381112.3110.9107.9115.2
31
5.2Ghz Auto+3600 18-21-21-431088.3109.2105.5111.4
32
5.2Ghz Auto1081.0107.2106.6111.4
33
Win vs 黑蘋果Hackintosh
34
Win10 Mercury CUDA1006.099.697.8105.4
35
macOS Mercury Metal919.893.492.988.2
36
macOS Mercury OpenCL920.393.593.088.1
37
Win10 software only923.384.698.0102.1
38
macOS software only884.385.294.488.9
39
Puget Systems數據
40
Puget Systems i9-9900K985.398.697.699.3
41
Puget Systems i7-9700K933.595.186.396.9
42
Puget Systems i5-9600K795.883.065.986.4
43
Puget Systems R7 2700X727.380.157.573.2
44
X99 RAM通道及容量測試
45
4Channel 四通32G655.363.167.368.6
46
2Channel 雙通32G649.561.667.269.4
47
2Channel 雙通16G650.362.966.667.7
48
1Channel 單通8G649.365.862.765.4
49
其他
50
AMD R7 1700 (all core3.7Ghz)+32G 2400747.377.769.973.6
51
MacBook Pro (13-inch, Early 2015) NT$40,000up
276.325.718.740.4
52
測試工具 : Puget Systems Adobe After Effects CC Benchmark
53
Puget Systems - After Effects CC 2019 CPU Roundup: Intel vs AMD vs Mac
54
在RAM測試上僅調整主要時序四個數值,其他皆auto,故僅供參考
55
跑分分數皆有一定誤差值
56
57
結論
58
CPU頻率,重要,頻率對效能的提升是肯定的
59
RAM通道,非常重要,雙通道對效能影響大(通俗講就是插兩條RAM比插一條好很多,8Gx2g勝過16Gx1),例外的是我在X99平台上測試容量與通道皆無差異
60
RAM容量,影響小,僅影響預覽長度,就整體來說32G是一個甜蜜點
61
RAM頻率,普通重要,但差距看起來不適合花大錢買,頻率3000和3200是較划算的選擇
62
RAM時序,影響小(不確定),這似乎關係到RAM本身的差異,但保險起見還是跟時脈配合盡可能縮緊時序
63
NVDIA顯卡(CUDA),重要,使用到CUDA加速特效的comp,依比例會有一定的效能提升
64
黑蘋果,大多情況相較Win10會有5~10%的效能損失
65
RAW DATA https://drive.google.com/open?id=1sw9C1oaVxW1vfKSFBKeljJ5HPEx5dKrV
66
67
測試設備
68
CPU:Intel Core i9 9900K
CPU散熱器:快睿 A40 Ultimate
主機板:技嘉 Z390 AORUS MASTER
記憶體:芝奇 G.SKILL 三叉戟16G*2 DDR4-3200 CL16 =32G
系統碟:Plextor M9PeGN 512G /M.2 PCIe
顯卡:AORUS GTX 1080 Ti Xtreme Edition 11G
電源:振華 Leadex Gold 850W
系統:Windows 10 version 1809
69
黑蘋果Hackintosh:
macOS High Sierra 10.13.6(17G65)
NVIDIA CUDA Driver 418.105
NVIDIA Web Driver 387.10.10.10.40.105
EFI: https://github.com/cmer/gigabyte-z390-aorus-master-hackintosh
70
71
CPU:INTEL I7-5820K (OC 4.2Ghz)
CPU散熱器:保銳 ETS-T40-TB
主機板:華碩X99-A
記憶體:美光DDR4 2133 8G*4 =32G
系統碟:威剛SP920 256G SSD SATA
顯卡:AORUS GTX 1080 Ti Xtreme Edition
72
73
After Effects電腦搭配參考指南
74
75
AE Benchmark 分數統計表格
76
→按此填寫 AE Benchmark 分數回報表單←
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100