| | | | | | |

Ryzen 5 3550H vs Ryzen 9 3900


Description
The 3550H is based on Zen+ architecture while the 3900 is based on Zen 2.

Using the multithread performance as a reference, the 3550H gets a score of 178.3 k points while the 3900 gets 687.5 k points.

Summarizing, the 3900 is 3.9 times faster than the 3550H. To get a proper comparison between both models, take a look to the data shown below.

Specs
CPUID
810f81
870f10
Core
Picasso
Matisse
Architecture
Base frecuency
2.1 GHz
3.1 GHz
Boost frecuency
3.7 GHz
4.3 GHz
Socket
BGA-FP5
AM4
Cores/Threads
4/8
12/24
TDP
35 W
65 W
Cache L1 (d+i)
4x64+4x32 kB
12x32+12x32 kB
Cache L2
4x512 kB
12x512 kB
Cache L3
4096 kB
4x16384 kB
Date
January 2019
September 2019
Mean monothread perf.
45.07k points
74.97k points
Mean multithread perf.
178.26k 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
3550H
3900
Test#1 (Integers)
3.63k
4.39k (x1.21)
Test#2 (FP)
16.69k
16.99k (x1.02)
Test#3 (Generic, ZIP)
4.92k
7.74k (x1.57)
Test#1 (Memory)
5k
24.57k (x4.91)
TOTAL
30.24k
53.69k (x1.78)

Multithread

3550H

3900
Test#1 (Integers)
13.62k
48.45k (x3.56)
Test#2 (FP)
65.71k
230.81k (x3.51)
Test#3 (Generic, ZIP)
23.7k
122.82k (x5.18)
Test#1 (Memory)
5.64k
54.73k (x9.71)
TOTAL
108.67k
456.8k (x4.2)

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
3550H
3900
Test#1 (Integers)
13.51k
16.85k (x1.25)
Test#2 (FP)
22.34k
26.03k (x1.17)
Test#3 (Generic, ZIP)
4.98k
9.54k (x1.91)
Test#1 (Memory)
4.24k
22.55k (x5.32)
TOTAL
45.07k
74.97k (x1.66)

Multithread

3550H

3900
Test#1 (Integers)
52.21k
229.41k (x4.39)
Test#2 (FP)
95.66k
292.81k (x3.06)
Test#3 (Generic, ZIP)
25.54k
128.48k (x5.03)
Test#1 (Memory)
4.85k
36.8k (x7.59)
TOTAL
178.26k
687.5k (x3.86)

Performance/W
3550H
3900
Test#1 (Integers)
1492 points/W
3529 points/W
Test#2 (FP)
2733 points/W
4505 points/W
Test#3 (Generic, ZIP)
730 points/W
1977 points/W
Test#1 (Memory)
139 points/W
566 points/W
TOTAL
5093 points/W
10577 points/W

Performance/GHz
3550H
3900
Test#1 (Integers)
3652 points/GHz
3920 points/GHz
Test#2 (FP)
6038 points/GHz
6054 points/GHz
Test#3 (Generic, ZIP)
1347 points/GHz
2218 points/GHz
Test#1 (Memory)
1145 points/GHz
5243 points/GHz
TOTAL
12182 points/GHz
17435 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