| | | | | | |

Ryzen 9 3900X vs Xeon E5-2697 v2


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

Using the multithread performance as a reference, the 3900X gets a score of 695.7 k points while the E5-2697 v2 gets 633.1 k points.

Summarizing, the 3900X is 1.1 times faster than the E5-2697 v2 . To get a proper comparison between both models, take a look to the data shown below.

Specs
CPUID
870f10
306e4
Core
Matisse
Ivy Bridge-EP
Architecture
Base frecuency
3.8 GHz
2.7 GHz
Boost frecuency
4.6 GHz
3.5 GHz
Socket
AM4
LGA 2011
Cores/Threads
12/24
12 /24
TDP
105 W
130 W
Cache L1 (d+i)
12x32+12x32 kB
12x32+12x32 kB
Cache L2
12x512 kB
12x256 kB
Cache L3
4x16384 kB
30720 kB
Date
July 2019
September 2013
Mean monothread perf.
72.51k points
27.92k points
Mean multithread perf.
756.3k 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
3900X
E5-2697 v2
Test#1 (Integers)
4.46k
3.44k (x0.77)
Test#2 (FP)
18.46k
9.54k (x0.52)
Test#3 (Generic, ZIP)
8.05k
4.21k (x0.52)
Test#1 (Memory)
23.71k
3.94k (x0.17)
TOTAL
54.68k
21.13k (x0.39)

Multithread

3900X

E5-2697 v2
Test#1 (Integers)
52.65k
73.17k (x1.39)
Test#2 (FP)
265.41k
230.83k (x0.87)
Test#3 (Generic, ZIP)
140.78k
107.14k (x0.76)
Test#1 (Memory)
46.91k
7.99k (x0.17)
TOTAL
505.76k
419.12k (x0.83)

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
3900X
E5-2697 v2
Test#1 (Integers)
17.35k
10.64k (x0.61)
Test#2 (FP)
25.87k
9.53k (x0.37)
Test#3 (Generic, ZIP)
8.72k
4.12k (x0.47)
Test#1 (Memory)
27.86k
3.62k (x0.13)
TOTAL
79.8k
27.91k (x0.35)

Multithread

3900X

E5-2697 v2
Test#1 (Integers)
201.57k
252.69k (x1.25)
Test#2 (FP)
284.41k
238.68k (x0.84)
Test#3 (Generic, ZIP)
145.16k
115.5k (x0.8)
Test#1 (Memory)
100.56k
10.98k (x0.11)
TOTAL
731.7k
617.84k (x0.84)

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
3900X
E5-2697 v2
Test#1 (Integers)
16.95k
10.53k (x0.62)
Test#2 (FP)
25.37k
9.98k (x0.39)
Test#3 (Generic, ZIP)
9.23k
4k (x0.43)
Test#1 (Memory)
24.83k
3.4k (x0.14)
TOTAL
76.38k
27.92k (x0.37)

Multithread

3900X

E5-2697 v2
Test#1 (Integers)
196.31k
247.05k (x1.26)
Test#2 (FP)
290.5k
264.43k (x0.91)
Test#3 (Generic, ZIP)
143.95k
111.01k (x0.77)
Test#1 (Memory)
64.92k
10.66k (x0.16)
TOTAL
695.69k
633.14k (x0.91)

Performance/W
3900X
E5-2697 v2
Test#1 (Integers)
1870 points/W
1900 points/W
Test#2 (FP)
2767 points/W
2034 points/W
Test#3 (Generic, ZIP)
1371 points/W
854 points/W
Test#1 (Memory)
618 points/W
82 points/W
TOTAL
6626 points/W
4870 points/W

Performance/GHz
3900X
E5-2697 v2
Test#1 (Integers)
3686 points/GHz
3007 points/GHz
Test#2 (FP)
5515 points/GHz
2852 points/GHz
Test#3 (Generic, ZIP)
2007 points/GHz
1144 points/GHz
Test#1 (Memory)
5398 points/GHz
973 points/GHz
TOTAL
16605 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