| | | | | | |

A9 9420 vs Xeon E5-2670 v3


Description
The 9420 is based on Excavator architecture while the E5-2670 v3 is based on Haswell.

Using the multithread performance as a reference, the 9420 gets a score of 46.3 k points while the E5-2670 v3 gets 452.9 k points.

Summarizing, the E5-2670 v3 is 9.8 times faster than the 9420. To get a proper comparison between both models, take a look to the data shown below.

Specs
CPUID
670f00
306f2
Core
Stoney Ridge
Haswell-EP
Architecture
Base frecuency
3 GHz
2.3 GHz
Boost frecuency
3.6 GHz
3.1 GHz
Socket
Micro-BGA
LGA 2011-3
Cores/Threads
2/2
12/24
TDP
15 W
120 W
Cache L1 (d+i)
96+2x32 kB
12x32+12x32 kB
Cache L2
1024 kB
12x256 kB
Cache L3
0 kB
30720 kB
Date
April 2017
September 2014
Mean monothread perf.
33.09k points
36.41k points
Mean multithread perf.
46.33k points
452.9k 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
E5-2670 v3
Test#1 (Integers)
2.56k
3.52k (x1.37)
Test#2 (FP)
9.97k
8.68k (x0.87)
Test#3 (Generic, ZIP)
2.96k
2.83k (x0.96)
Test#1 (Memory)
2.12k
3.13k (x1.47)
TOTAL
17.61k
18.17k (x1.03)

Multithread

9420

E5-2670 v3
Test#1 (Integers)
3.88k
32.7k (x8.42)
Test#2 (FP)
9.26k
105.18k (x11.36)
Test#3 (Generic, ZIP)
3.9k
35.18k (x9.01)
Test#1 (Memory)
1.74k
9.54k (x5.48)
TOTAL
18.79k
182.59k (x9.72)

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
E5-2670 v3
Test#1 (Integers)
8.41k
11.08k (x1.32)
Test#2 (FP)
11.48k
9.57k (x0.83)
Test#3 (Generic, ZIP)
3.29k
3.18k (x0.97)
Test#1 (Memory)
2.24k
3.28k (x1.47)
TOTAL
25.41k
27.11k (x1.07)

Multithread

9420

E5-2670 v3
Test#1 (Integers)
11.25k
139.07k (x12.36)
Test#2 (FP)
13.69k
133.47k (x9.75)
Test#3 (Generic, ZIP)
2.09k
46.6k (x22.34)
Test#1 (Memory)
1.87k
6.65k (x3.56)
TOTAL
28.89k
325.78k (x11.28)

AVX optimized benchmark
The benchmark in mode II (AVX) is optimized to used 256 bits registers beside the first version of the Advanced Vector Extensions (AVX). The first AVX compatible CPU was released in 2011.
Monothread
9420
E5-2670 v3
Test#1 (Integers)
7.55k
11.18k (x1.48)
Test#2 (FP)
13.9k
10.06k (x0.72)
Test#3 (Generic, ZIP)
3.11k
3.21k (x1.03)
Test#1 (Memory)
2.18k
3.12k (x1.43)
TOTAL
26.74k
27.57k (x1.03)

Multithread

9420

E5-2670 v3
Test#1 (Integers)
9.09k
140.98k (x15.51)
Test#2 (FP)
12.25k
145.37k (x11.87)
Test#3 (Generic, ZIP)
3k
45.62k (x15.19)
Test#1 (Memory)
1.81k
6.75k (x3.74)
TOTAL
26.15k
338.73k (x12.95)

AVX2 optimized benchmark
The benchmark in mode III (AVX2), like AVX1, is optimized to used 256 bits registers beside the second version of the Advanced Vector Extensions (AVX). The first AVX2 compatible CPU was released in 2013.
Monothread
9420
E5-2670 v3
Test#1 (Integers)
11.74k
18.62k (x1.59)
Test#2 (FP)
16.05k
10.49k (x0.65)
Test#3 (Generic, ZIP)
3.26k
4.04k (x1.24)
Test#1 (Memory)
2.05k
3.26k (x1.59)
TOTAL
33.09k
36.41k (x1.1)

Multithread

9420

E5-2670 v3
Test#1 (Integers)
18.57k
236.25k (x12.72)
Test#2 (FP)
20.11k
151.05k (x7.51)
Test#3 (Generic, ZIP)
5.27k
58.9k (x11.17)
Test#1 (Memory)
2.38k
6.7k (x2.82)
TOTAL
46.33k
452.9k (x9.77)

Performance/W
9420
E5-2670 v3
Test#1 (Integers)
1238 points/W
1969 points/W
Test#2 (FP)
1340 points/W
1259 points/W
Test#3 (Generic, ZIP)
352 points/W
491 points/W
Test#1 (Memory)
159 points/W
56 points/W
TOTAL
3089 points/W
3774 points/W

Performance/GHz
9420
E5-2670 v3
Test#1 (Integers)
3260 points/GHz
6006 points/GHz
Test#2 (FP)
4458 points/GHz
3385 points/GHz
Test#3 (Generic, ZIP)
905 points/GHz
1303 points/GHz
Test#1 (Memory)
570 points/GHz
1052 points/GHz
TOTAL
9192 points/GHz
11745 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