| | | | | | |

Xeon E5645 vs E5-2620 v4


Description
The E5645 is based on Westmere architecture while the E5-2620 v4 is based on Broadwell.

Using the multithread performance as a reference, the E5645 gets a score of 77 k points while the E5-2620 v4 gets 182.4 k points.

Summarizing, the E5-2620 v4 is 2.4 times faster than the E5645 . To get a proper comparison between both models, take a look to the data shown below.

Specs
CPUID
206c2
406f1
Core
Westmere-EP
Broadwell-EP
Architecture
Base frecuency
2.4 GHz
2.1 GHz
Boost frecuency
2.8 GHz
3 GHz
Socket
LGA 1366
Socket 2011-3
Cores/Threads
6 /6
8/16
TDP
80 W
85 W
Cache L1 (d+i)
64 kB
8x32+8x32 kB
Cache L2
256 kB
8x256 kB
Cache L3
12288 kB
20480 kB
Date
March 2010
March 2016
Mean monothread perf.
15.64k points
29.39k points
Mean multithread perf.
76.96k points
237.83k 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
E5645
E5-2620 v4
Test#1 (Integers)
1.78k
2.6k (x1.46)
Test#2 (FP)
3.91k
7.55k (x1.93)
Test#3 (Generic, ZIP)
2.52k
2.61k (x1.03)
Test#1 (Memory)
2.42k
2.03k (x0.84)
TOTAL
10.64k
14.79k (x1.39)

Multithread

E5645

E5-2620 v4
Test#1 (Integers)
9.67k
17.41k (x1.8)
Test#2 (FP)
22.92k
75.46k (x3.29)
Test#3 (Generic, ZIP)
14.31k
26.25k (x1.83)
Test#1 (Memory)
3.81k
4.65k (x1.22)
TOTAL
50.7k
123.78k (x2.44)

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
E5645
E5-2620 v4
Test#1 (Integers)
6.47k
8.22k (x1.27)
Test#2 (FP)
4.4k
12.06k (x2.74)
Test#3 (Generic, ZIP)
2.43k
4.01k (x1.65)
Test#1 (Memory)
2.33k
2.89k (x1.24)
TOTAL
15.64k
27.18k (x1.74)

Multithread

E5645

E5-2620 v4
Test#1 (Integers)
34.43k
61.34k (x1.78)
Test#2 (FP)
24.23k
90.21k (x3.72)
Test#3 (Generic, ZIP)
14.48k
26.2k (x1.81)
Test#1 (Memory)
3.81k
4.62k (x1.21)
TOTAL
76.96k
182.37k (x2.37)

Performance/W
E5645
E5-2620 v4
Test#1 (Integers)
430 points/W
722 points/W
Test#2 (FP)
303 points/W
1061 points/W
Test#3 (Generic, ZIP)
181 points/W
308 points/W
Test#1 (Memory)
48 points/W
54 points/W
TOTAL
962 points/W
2146 points/W

Performance/GHz
E5645
E5-2620 v4
Test#1 (Integers)
2311 points/GHz
2739 points/GHz
Test#2 (FP)
1573 points/GHz
4021 points/GHz
Test#3 (Generic, ZIP)
867 points/GHz
1335 points/GHz
Test#1 (Memory)
833 points/GHz
963 points/GHz
TOTAL
5584 points/GHz
9059 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