Viewing page 5 of 6 pages.
Previous 1 2 3 4 5 6 Next
Performance
|
Benchmark Comparison
|
|
Test State
|
i7-965
|
i7-870
|
i5-750
|
QX9650
|
Q9550
|
X4 955
|
|
NOD32
|
2:02
|
2:15
|
2:31
|
2:23
|
2:31
|
2:27
|
|
WinRAR
|
2:16
|
2:28
|
2:38
|
2:51
|
2:59
|
3:05
|
|
iTunes
|
3:11
|
3:29
|
3:48
|
3:20
|
3:33
|
4:34
|
|
TMPGEnc
|
2:29
|
3:10
|
3:21
|
3:07
|
3:14
|
2:52
|
|
PCMark05
|
10502
|
8985
|
8390
|
9061
|
8760
|
9004
|
The 2.66 GHz i5-750 slightly outpaced the 2.83 GHz Q9550 in two of our timed
benchmarks, NOD32 and WinRAR. iTunes and TMPGEnc favored the Q9550, similar
margin. PCMark seemed to give the edge to the high-clocked processor, but for
all intents and purposes, the two are more or less equal in performance.
The 2.93 GHz i7-870 and 3.00 GHz QX9650 traded blows in a similar fashion.
The biggest difference between the LGA1156 and LGA775 processors seemed to be
in WinRAR, a very memory-dependent application. The lower latency and high bandwidth
provided by integrated memory controllers helped the Lynnfield CPUs take a large
lead in this particular test.
It would seem that these LGA1156 processors bridge the performance gap between
LGA1366 and LGA775, offering a nice performance boost over Core 2 Quads, but
falling short of the Bloomfield i7's. In our most stressful test, video encoding
with TMPGEnc, the Core i7-965
EE beat the i7-870 by 22% even though its clock speed was only 66 MHz
higher. AMD's second fastest desktop processor, the Phenom
II X4 955 Black Edition, is priced similarly to the i5-750 and depending
on the application, the i5-750 may be slightly faster.
Average/Total Benchmark Power Consumption
|
Average Benchmark Power Consumption
|
|
Benchmark
|
i7-965
|
i7-870
|
i5-750
|
QX9650
|
Q9550
|
X4 955
|
|
NOD32
|
111W
|
76W
|
75W
|
88W
|
84W
|
128W
|
|
WinRAR
|
108W
|
81W
|
79W
|
98W
|
96W
|
128W
|
|
iTunes
|
115W
|
78W
|
75W
|
92W
|
89W
|
137W
|
|
TMPGEnc
|
155W
|
108W
|
109W
|
118W
|
113W
|
167W
|
|
Total Benchmark Power Consumption (Watt-hours)
|
|
Benchmark
|
i7-965
|
i7-870
|
i5-750
|
QX9650
|
Q9550
|
X4 955
|
|
NOD32
|
3.76
|
2.85
|
3.15
|
3.50
|
3.52
|
5.23
|
|
WinRAR
|
4.08
|
3.33
|
3.47
|
4.66
|
4.77
|
6.58
|
|
iTunes
|
6.10
|
4.53
|
4.75
|
5.11
|
5.27
|
10.43
|
|
TMPGEnc
|
6.42
|
5.70
|
6.09
|
6.13
|
6.09
|
7.98
|
The total amount of power used during our timed benchmarks favored the Lynnfields
as well. In watt-hours used, the
i7-870 beat out the QX9650 by 15%, while the i5-750 bested the Q9550 by
11%. Both processors were also more efficient than the i7-965. As for the
X4 955 BE, well the numbers speak for themselves.
Underclocked Average/Total Benchmark Power Consumption
|
Average System Benchmark Power Consumption
|
|
Benchmark
|
i7-870
|
i7-870
|
i5-750
|
i5-750
|
Q9550
|
Q9550S |
|
Stock
|
UV
|
Stock
|
UV
|
|
NOD32
|
76W
|
63W
|
75W
|
71W
|
84W
|
78W
|
|
WinRAR
|
81W
|
71W
|
79W
|
70W
|
96W
|
87W
|
|
iTunes
|
78W
|
75W
|
75W
|
73W
|
89W
|
81W
|
|
TMPGEnc
|
108W
|
91W
|
109W
|
95W
|
113W
|
102W
|
|
i7-870 undervolted to 1.025V stable (~1.1375V stock).
i5-750 undervolted to 1.0625V stable (~1.1625V stock).
|
|
Total System Benchmark Power Consumption (Watt-hours)
|
|
Benchmark
|
i7-870
|
i7-870
|
i5-750
|
i5-750
|
Q9550
|
Q9550S
|
|
Stock
|
UV
|
Stock
|
UV
|
|
NOD32
|
2.85
|
2.36
|
3.15
|
2.98
|
3.52
|
3.27
|
|
WinRAR
|
3.33
|
2.92
|
3.47
|
3.07
|
4.77
|
4.28
|
|
iTunes
|
4.53
|
4.35
|
4.75
|
4.62
|
5.27
|
4.79
|
|
TMPGEnc
|
5.70
|
4.80
|
6.09
|
5.30
|
6.09
|
5.53
|
|
i7-870 undervolted to 1.025V stable (~1.1375V stock).
i5-750 undervolted to 1.0625V stable (~1.1625V stock).
|
Undervolting shaved 12% off the total power used by the i7-870 during our timed
tests and 9% for the i5-750. Being able to run a stressful video encoding application
like TMPGEnc on a quad-core processor drawing less than 100W from the wall is
superb. Our i7-870 sample was the overall winner in efficiency. It had a fairly
low stock operating voltage and it also undervolted further than its lower-clocked
cousin.
Even at stock voltages, both Lynnfields edged out the Q9550S,
which has a TDP rating of 65W.
| Help support this site, buy from one of our affiliate retailers! |
|
