| | | | | | |

Xeon E5-2620 v4 vs E5-2697 v2


Description
The E5-2620 v4 is based on Broadwell architecture while the E5-2697 v2 is based on Ivy Bridge.

Using the multithread performance as a reference, the E5-2620 v4 gets a score of 147.8 k points while the E5-2697 v2 gets 633.1 k points.

Summarizing, the E5-2697 v2 is 4.3 times faster than the E5-2620 v4. To get a proper comparison between both models, take a look to the data shown below.

Specs
CPUID
406f1
306e4
Core
Broadwell-EP
Ivy Bridge-EP
Architecture
Base frecuency
2.1 GHz
2.7 GHz
Boost frecuency
3 GHz
3.5 GHz
Socket
Socket 2011-3
LGA 2011
Cores/Threads
8/16
12 /24
TDP
85 W
130 W
Cache L1 (d+i)
8x32+8x32 kB
12x32+12x32 kB
Cache L2
8x256 kB
12x256 kB
Cache L3
20480 kB
30720 kB
Date
March 2016
September 2013
Mean monothread perf.
29.39k points
27.92k points
Mean multithread perf.
237.83k points
633.14k 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
E5-2620 v4
E5-2697 v2
Test#1 (Integers)
2.6k
3.44k (x1.32)
Test#2 (FP)
7.55k
9.54k (x1.26)
Test#3 (Generic, ZIP)
2.61k
4.21k (x1.61)
Test#1 (Memory)
2.03k
3.94k (x1.94)
TOTAL
14.79k
21.13k (x1.43)

Multithread

E5-2620 v4

E5-2697 v2
Test#1 (Integers)
17.41k
73.17k (x4.2)
Test#2 (FP)
75.46k
230.83k (x3.06)
Test#3 (Generic, ZIP)
26.25k
107.14k (x4.08)
Test#1 (Memory)
4.65k
7.99k (x1.72)
TOTAL
123.78k
419.12k (x3.39)

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
E5-2620 v4
E5-2697 v2
Test#1 (Integers)
8.22k
10.64k (x1.3)
Test#2 (FP)
12.06k
9.53k (x0.79)
Test#3 (Generic, ZIP)
4.01k
4.12k (x1.03)
Test#1 (Memory)
2.89k
3.62k (x1.25)
TOTAL
27.18k
27.91k (x1.03)

Multithread

E5-2620 v4

E5-2697 v2
Test#1 (Integers)
61.34k
252.69k (x4.12)
Test#2 (FP)
90.21k
238.68k (x2.65)
Test#3 (Generic, ZIP)
26.2k
115.5k (x4.41)
Test#1 (Memory)
4.62k
10.98k (x2.37)
TOTAL
182.37k
617.84k (x3.39)

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
E5-2620 v4
E5-2697 v2
Test#1 (Integers)
6.04k
10.53k (x1.74)
Test#2 (FP)
10.18k
9.98k (x0.98)
Test#3 (Generic, ZIP)
2.74k
4k (x1.46)
Test#1 (Memory)
2.41k
3.4k (x1.42)
TOTAL
21.37k
27.92k (x1.31)

Multithread

E5-2620 v4

E5-2697 v2
Test#1 (Integers)
47.51k
247.05k (x5.2)
Test#2 (FP)
75.12k
264.43k (x3.52)
Test#3 (Generic, ZIP)
19.72k
111.01k (x5.63)
Test#1 (Memory)
5.45k
10.66k (x1.95)
TOTAL
147.8k
633.14k (x4.28)

Performance/W
E5-2620 v4
E5-2697 v2
Test#1 (Integers)
559 points/W
1900 points/W
Test#2 (FP)
884 points/W
2034 points/W
Test#3 (Generic, ZIP)
232 points/W
854 points/W
Test#1 (Memory)
64 points/W
82 points/W
TOTAL
1739 points/W
4870 points/W

Performance/GHz
E5-2620 v4
E5-2697 v2
Test#1 (Integers)
2013 points/GHz
3007 points/GHz
Test#2 (FP)
3393 points/GHz
2852 points/GHz
Test#3 (Generic, ZIP)
915 points/GHz
1144 points/GHz
Test#1 (Memory)
802 points/GHz
973 points/GHz
TOTAL
7123 points/GHz
7976 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