| | | | | | |

Xeon E5-2697 v2 vs Ryzen 9 3900


Description
The E5-2697 v2 is based on Ivy Bridge architecture while the 3900 is based on Zen 2.

Using the multithread performance as a reference, the E5-2697 v2 gets a score of 633.1 k points while the 3900 gets 620.2 k points.

Summarizing, the E5-2697 v2 is 1 times faster than the 3900. To get a proper comparison between both models, take a look to the data shown below.

Specs
CPUID
306e4
870f10
Core
Ivy Bridge-EP
Matisse
Architecture
Base frecuency
2.7 GHz
3.1 GHz
Boost frecuency
3.5 GHz
4.3 GHz
Socket
LGA 2011
AM4
Cores/Threads
12 /24
12/24
TDP
130 W
65 W
Cache L1 (d+i)
12x32+12x32 kB
12x32+12x32 kB
Cache L2
12x256 kB
12x512 kB
Cache L3
30720 kB
4x16384 kB
Date
September 2013
September 2019
Mean monothread perf.
27.92k points
74.97k points
Mean multithread perf.
633.14k points
687.5k 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
E5-2697 v2
3900
Test#1 (Integers)
3.44k
4.39k (x1.28)
Test#2 (FP)
9.54k
16.99k (x1.78)
Test#3 (Generic, ZIP)
4.21k
7.74k (x1.84)
Test#1 (Memory)
3.94k
24.57k (x6.23)
TOTAL
21.13k
53.69k (x2.54)

Multithread

E5-2697 v2

3900
Test#1 (Integers)
73.17k
48.45k (x0.66)
Test#2 (FP)
230.83k
230.81k (x1)
Test#3 (Generic, ZIP)
107.14k
122.82k (x1.15)
Test#1 (Memory)
7.99k
54.73k (x6.85)
TOTAL
419.12k
456.8k (x1.09)

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
E5-2697 v2
3900
Test#1 (Integers)
10.64k
16.59k (x1.56)
Test#2 (FP)
9.53k
24.55k (x2.58)
Test#3 (Generic, ZIP)
4.12k
8.55k (x2.07)
Test#1 (Memory)
3.62k
26.28k (x7.27)
TOTAL
27.91k
75.97k (x2.72)

Multithread

E5-2697 v2

3900
Test#1 (Integers)
252.69k
181.33k (x0.72)
Test#2 (FP)
238.68k
261.01k (x1.09)
Test#3 (Generic, ZIP)
115.5k
126.62k (x1.1)
Test#1 (Memory)
10.98k
66.64k (x6.07)
TOTAL
617.84k
635.6k (x1.03)

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
E5-2697 v2
3900
Test#1 (Integers)
10.53k
16.16k (x1.54)
Test#2 (FP)
9.98k
23.75k (x2.38)
Test#3 (Generic, ZIP)
4k
9.32k (x2.33)
Test#1 (Memory)
3.4k
23.17k (x6.81)
TOTAL
27.92k
72.4k (x2.59)

Multithread

E5-2697 v2

3900
Test#1 (Integers)
247.05k
174.73k (x0.71)
Test#2 (FP)
264.43k
275.63k (x1.04)
Test#3 (Generic, ZIP)
111.01k
125.65k (x1.13)
Test#1 (Memory)
10.66k
44.17k (x4.14)
TOTAL
633.14k
620.18k (x0.98)

Performance/W
E5-2697 v2
3900
Test#1 (Integers)
1900 points/W
2688 points/W
Test#2 (FP)
2034 points/W
4240 points/W
Test#3 (Generic, ZIP)
854 points/W
1933 points/W
Test#1 (Memory)
82 points/W
680 points/W
TOTAL
4870 points/W
9541 points/W

Performance/GHz
E5-2697 v2
3900
Test#1 (Integers)
3007 points/GHz
3758 points/GHz
Test#2 (FP)
2852 points/GHz
5524 points/GHz
Test#3 (Generic, ZIP)
1144 points/GHz
2167 points/GHz
Test#1 (Memory)
973 points/GHz
5388 points/GHz
TOTAL
7976 points/GHz
16838 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