| | | | | | |

A9 9420 vs Xeon E5472


Description
The 9420 is based on Excavator architecture while the E5472 is based on Core.

Using the multithread performance as a reference, the 9420 gets a score of 28.9 k points while the E5472 gets 48.6 k points.

Summarizing, the E5472 is 1.7 times faster than the 9420. To get a proper comparison between both models, take a look to the data shown below.

Specs
CPUID
670f00
10676
Core
Stoney Ridge
Harpertown
Architecture
Base frecuency
3 GHz
3 GHz
Boost frecuency
3.6 GHz
3 GHz
Socket
Micro-BGA
LGA-771
Cores/Threads
2/2
4/4
TDP
15 W
80 W
Cache L1 (d+i)
96+2x32 kB
4x32+4x32 kB
Cache L2
1024 kB
2x6144 kB
Cache L3
0 kB
kB
Date
April 2017
November 2007
Mean monothread perf.
33.09k points
16.62k points
Mean multithread perf.
46.33k points
48.62k points

Non-optimized benchmark
The benchmark in Mode 0 (FPU) measures cpu performance with non-optimized software. It uses the basic µinstructions from the i386 architecture with the i387 floating point unit. This mode is compatible with all CPUs so it's practical to compare very different CPUs
Monothread
9420
E5472
Test#1 (Integers)
2.56k
2.29k (x0.89)
Test#2 (FP)
9.97k
5.85k (x0.59)
Test#3 (Generic, ZIP)
2.96k
2.44k (x0.83)
Test#1 (Memory)
2.12k
4.81k (x2.27)
TOTAL
17.61k
15.39k (x0.87)

Multithread

9420

E5472
Test#1 (Integers)
3.88k
9.06k (x2.33)
Test#2 (FP)
9.26k
23.09k (x2.49)
Test#3 (Generic, ZIP)
3.9k
9.7k (x2.48)
Test#1 (Memory)
1.74k
2.53k (x1.45)
TOTAL
18.79k
44.38k (x2.36)

SSE3 optimized benchmark
The benchmark in mode I (SSE) is optimized for the use of SIMD instructions with 128 bits register and the SSE set up to version 3. Nearly every modern CPU has support for this mode.
Monothread
9420
E5472
Test#1 (Integers)
8.41k
2.88k (x0.34)
Test#2 (FP)
11.48k
6.36k (x0.55)
Test#3 (Generic, ZIP)
3.29k
2.55k (x0.78)
Test#1 (Memory)
2.24k
4.83k (x2.16)
TOTAL
25.41k
16.62k (x0.65)

Multithread

9420

E5472
Test#1 (Integers)
11.25k
11.19k (x0.99)
Test#2 (FP)
13.69k
24.85k (x1.82)
Test#3 (Generic, ZIP)
2.09k
10.04k (x4.81)
Test#1 (Memory)
1.87k
2.54k (x1.36)
TOTAL
28.89k
48.62k (x1.68)

Performance/W
9420
E5472
Test#1 (Integers)
750 points/W
140 points/W
Test#2 (FP)
912 points/W
311 points/W
Test#3 (Generic, ZIP)
139 points/W
125 points/W
Test#1 (Memory)
125 points/W
32 points/W
TOTAL
1926 points/W
608 points/W

Performance/GHz
9420
E5472
Test#1 (Integers)
2336 points/GHz
961 points/GHz
Test#2 (FP)
3188 points/GHz
2119 points/GHz
Test#3 (Generic, ZIP)
913 points/GHz
850 points/GHz
Test#1 (Memory)
621 points/GHz
1610 points/GHz
TOTAL
7058 points/GHz
5539 points/GHz

Monothread performance graph
Monothread performance graphics gives the performance vs time. They are useful to measure the time it takes to the CPU to reach the maximum performance.

Usually, CPU's performance will be steady during these tests but if it has a slow frequency strategy, the first samples will show a lower score.


Test#1 (Integers) [points vs time]

grafica bm.hardlimit.com


Test#2 (FP) [points vs time]

grafica bm.hardlimit.com


Test#3 (Generic, ZIP) [points vs time]

grafica bm.hardlimit.com


Test#1 (Memory) [points vs time]

grafica bm.hardlimit.com

Multithread performance graph
Multithread graphs measure the performance against a heavy load during certain time.

If CPU's TDP doesn't limit the frequency and the machine is properly cooled, performance should remain steady vs time. Otherwise, the performance score will oscillate or decrease over time.


Test#1 (Integers) [points vs time]

grafica bm.hardlimit.com


Test#2 (FP) [points vs time]

grafica bm.hardlimit.com


Test#3 (Generic, ZIP) [points vs time]

grafica bm.hardlimit.com


Test#1 (Memory) [points vs time]

grafica bm.hardlimit.com

Hardlimit Benchmark Central - Ver. 3.11.4