| | | | | | |

Xeon E5-2620 v4 vs E3-1220 v3


Description
The E5-2620 v4 is based on Broadwell architecture while the E3-1220 v3 is based on Haswell.

Using the multithread performance as a reference, the E5-2620 v4 gets a score of 237.8 k points while the E3-1220 v3 gets 140.8 k points.

Summarizing, the E5-2620 v4 is 1.7 times faster than the E3-1220 v3 . To get a proper comparison between both models, take a look to the data shown below.

Specs
CPUID
406f1
306c3
Core
Broadwell-EP
Haswell-WS
Architecture
Base frecuency
2.1 GHz
3.1 GHz
Boost frecuency
3 GHz
3.5 GHz
Socket
Socket 2011-3
LGA 1150
Cores/Threads
8/16
4 /4
TDP
85 W
80 W
Cache L1 (d+i)
8x32+8x32 kB
64 kB
Cache L2
8x256 kB
256 kB
Cache L3
20480 kB
8192 kB
Date
March 2016
June 2013
Mean monothread perf.
29.39k points
39.16k points
Mean multithread perf.
237.83k points
140.76k 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
E3-1220 v3
Test#1 (Integers)
2.6k
3.45k (x1.32)
Test#2 (FP)
7.55k
9.96k (x1.32)
Test#3 (Generic, ZIP)
2.61k
4.26k (x1.63)
Test#1 (Memory)
2.03k
4.27k (x2.11)
TOTAL
14.79k
21.94k (x1.48)

Multithread

E5-2620 v4

E3-1220 v3
Test#1 (Integers)
17.41k
14.01k (x0.8)
Test#2 (FP)
75.46k
37.68k (x0.5)
Test#3 (Generic, ZIP)
26.25k
17.05k (x0.65)
Test#1 (Memory)
4.65k
7.35k (x1.58)
TOTAL
123.78k
76.09k (x0.61)

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
E5-2620 v4
E3-1220 v3
Test#1 (Integers)
12.16k
19.17k (x1.58)
Test#2 (FP)
11.82k
10.55k (x0.89)
Test#3 (Generic, ZIP)
3.02k
4.79k (x1.59)
Test#1 (Memory)
2.4k
4.65k (x1.94)
TOTAL
29.39k
39.16k (x1.33)

Multithread

E5-2620 v4

E3-1220 v3
Test#1 (Integers)
103.77k
73.13k (x0.7)
Test#2 (FP)
102.28k
42.06k (x0.41)
Test#3 (Generic, ZIP)
26.13k
18.03k (x0.69)
Test#1 (Memory)
5.65k
7.54k (x1.34)
TOTAL
237.83k
140.76k (x0.59)

Performance/W
E5-2620 v4
E3-1220 v3
Test#1 (Integers)
1221 points/W
914 points/W
Test#2 (FP)
1203 points/W
526 points/W
Test#3 (Generic, ZIP)
307 points/W
225 points/W
Test#1 (Memory)
66 points/W
94 points/W
TOTAL
2798 points/W
1759 points/W

Performance/GHz
E5-2620 v4
E3-1220 v3
Test#1 (Integers)
4054 points/GHz
5478 points/GHz
Test#2 (FP)
3939 points/GHz
3014 points/GHz
Test#3 (Generic, ZIP)
1005 points/GHz
1368 points/GHz
Test#1 (Memory)
799 points/GHz
1328 points/GHz
TOTAL
9798 points/GHz
11188 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