| | | | | | |

Ryzen 9 3900 vs Xeon Silver 4208


Description
The 3900 is based on Zen 2 architecture while the Silver 4208 is based on Cascade Lake.

Using the multithread performance as a reference, the 3900 gets a score of 687.5 k points while the Silver 4208 gets 311.6 k points.

Summarizing, the 3900 is 2.2 times faster than the Silver 4208. To get a proper comparison between both models, take a look to the data shown below.

Specs
CPUID
870f10
50657
Core
Matisse
Cascade Lake-SP
Architecture
Base frecuency
3.1 GHz
2.1 GHz
Boost frecuency
4.3 GHz
3.2 GHz
Socket
AM4
LGA 3647
Cores/Threads
12/24
8/16
TDP
65 W
85 W
Cache L1 (d+i)
12x32+12x32 kB
8x32+8x32 kB
Cache L2
12x512 kB
8x1024 kB
Cache L3
4x16384 kB
11264 kB
Date
September 2019
April 2019
Mean monothread perf.
74.97k points
38.61k points
Mean multithread perf.
687.5k points
311.57k 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
3900
Silver 4208
Test#1 (Integers)
4.39k
2.59k (x0.59)
Test#2 (FP)
16.99k
9.5k (x0.56)
Test#3 (Generic, ZIP)
7.74k
3.15k (x0.41)
Test#1 (Memory)
24.57k
5.9k (x0.24)
TOTAL
53.69k
21.13k (x0.39)

Multithread

3900

Silver 4208
Test#1 (Integers)
48.45k
21.09k (x0.44)
Test#2 (FP)
230.81k
80.93k (x0.35)
Test#3 (Generic, ZIP)
122.82k
30.76k (x0.25)
Test#1 (Memory)
54.73k
8.51k (x0.16)
TOTAL
456.8k
141.3k (x0.31)

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
3900
Silver 4208
Test#1 (Integers)
16.85k
16.86k (x1)
Test#2 (FP)
26.03k
12.08k (x0.46)
Test#3 (Generic, ZIP)
9.54k
3.37k (x0.35)
Test#1 (Memory)
22.55k
6.29k (x0.28)
TOTAL
74.97k
38.61k (x0.51)

Multithread

3900

Silver 4208
Test#1 (Integers)
229.41k
150.99k (x0.66)
Test#2 (FP)
292.81k
118.54k (x0.4)
Test#3 (Generic, ZIP)
128.48k
33.44k (x0.26)
Test#1 (Memory)
36.8k
8.6k (x0.23)
TOTAL
687.5k
311.57k (x0.45)

Performance/W
3900
Silver 4208
Test#1 (Integers)
3529 points/W
1776 points/W
Test#2 (FP)
4505 points/W
1395 points/W
Test#3 (Generic, ZIP)
1977 points/W
393 points/W
Test#1 (Memory)
566 points/W
101 points/W
TOTAL
10577 points/W
3666 points/W

Performance/GHz
3900
Silver 4208
Test#1 (Integers)
3920 points/GHz
5269 points/GHz
Test#2 (FP)
6054 points/GHz
3776 points/GHz
Test#3 (Generic, ZIP)
2218 points/GHz
1054 points/GHz
Test#1 (Memory)
5243 points/GHz
1966 points/GHz
TOTAL
17435 points/GHz
12065 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