Ryzen 9 3900 vs Ryzen 5 3500U
Description
The 3900 is based on Zen 2 architecture while the 3500U is based on Zen+.
Using the multithread performance as a reference, the 3900 gets a score of 687.5 k points while the 3500U gets 141 k points.
Summarizing, the 3900 is 4.9 times faster than the 3500U. To get a proper comparison between both models, take a look to the data shown below.
Using the multithread performance as a reference, the 3900 gets a score of 687.5 k points while the 3500U gets 141 k points.
Summarizing, the 3900 is 4.9 times faster than the 3500U. To get a proper comparison between both models, take a look to the data shown below.
Specs
CPUID
870f10
810f81
Core
Matisse
Picasso
Base frecuency
3.1 GHz
2.1 GHz
Boost frecuency
4.3 GHz
3.7 GHz
Socket
AM4
BGA-FP5
Cores/Threads
12/24
4/8
TDP
65 W
15 W
Cache L1 (d+i)
12x32+12x32 kB
4x64+6x32 kB
Cache L2
12x512 kB
4x512 kB
Cache L3
4x16384 kB
4096 kB
Date
September 2019
January 2019
Mean monothread perf.
74.97k points
36.64k points
Mean multithread perf.
687.5k points
140.97k 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
3500U
Test#1 (Integers)
4.39k
3.24k (x0.74)
Test#2 (FP)
16.99k
13.16k (x0.77)
Test#3 (Generic, ZIP)
7.74k
4.41k (x0.57)
Test#1 (Memory)
24.57k
3.32k (x0.14)
TOTAL
53.69k
24.12k (x0.45)
Multithread
3900
3500U
Test#1 (Integers)
48.45k
12.95k (x0.27)
Test#2 (FP)
230.81k
57.84k (x0.25)
Test#3 (Generic, ZIP)
122.82k
21.05k (x0.17)
Test#1 (Memory)
54.73k
3.51k (x0.06)
TOTAL
456.8k
95.36k (x0.21)
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
3900
3500U
Test#1 (Integers)
16.59k
11.98k (x0.72)
Test#2 (FP)
24.55k
15.85k (x0.65)
Test#3 (Generic, ZIP)
8.55k
4.53k (x0.53)
Test#1 (Memory)
26.28k
3.29k (x0.13)
TOTAL
75.97k
35.65k (x0.47)
Multithread
3900
3500U
Test#1 (Integers)
181.33k
48.28k (x0.27)
Test#2 (FP)
261.01k
74.19k (x0.28)
Test#3 (Generic, ZIP)
126.62k
23.29k (x0.18)
Test#1 (Memory)
66.64k
3.47k (x0.05)
TOTAL
635.6k
149.23k (x0.23)
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
3900
3500U
Test#1 (Integers)
16.16k
11.1k (x0.69)
Test#2 (FP)
23.75k
18.97k (x0.8)
Test#3 (Generic, ZIP)
9.32k
3.9k (x0.42)
Test#1 (Memory)
23.17k
3.39k (x0.15)
TOTAL
72.4k
37.37k (x0.52)
Multithread
3900
3500U
Test#1 (Integers)
174.73k
45.74k (x0.26)
Test#2 (FP)
275.63k
75.47k (x0.27)
Test#3 (Generic, ZIP)
125.65k
21.89k (x0.17)
Test#1 (Memory)
44.17k
3.49k (x0.08)
TOTAL
620.18k
146.6k (x0.24)
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
3500U
Test#1 (Integers)
16.85k
10.88k (x0.65)
Test#2 (FP)
26.03k
18.09k (x0.7)
Test#3 (Generic, ZIP)
9.54k
4.29k (x0.45)
Test#1 (Memory)
22.55k
3.36k (x0.15)
TOTAL
74.97k
36.64k (x0.49)
Multithread
3900
3500U
Test#1 (Integers)
229.41k
42.44k (x0.18)
Test#2 (FP)
292.81k
73.82k (x0.25)
Test#3 (Generic, ZIP)
128.48k
20.76k (x0.16)
Test#1 (Memory)
36.8k
3.96k (x0.11)
TOTAL
687.5k
140.97k (x0.21)
Performance/W
3900
3500U
Test#1 (Integers)
3529 points/W
2829 points/W
Test#2 (FP)
4505 points/W
4921 points/W
Test#3 (Generic, ZIP)
1977 points/W
1384 points/W
Test#1 (Memory)
566 points/W
264 points/W
TOTAL
10577 points/W
9398 points/W
Performance/GHz
3900
3500U
Test#1 (Integers)
3920 points/GHz
2942 points/GHz
Test#2 (FP)
6054 points/GHz
4890 points/GHz
Test#3 (Generic, ZIP)
2218 points/GHz
1161 points/GHz
Test#1 (Memory)
5243 points/GHz
909 points/GHz
TOTAL
17435 points/GHz
9901 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.
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]
Test#2 (FP) [points vs time]
Test#3 (Generic, ZIP) [points vs time]
Test#1 (Memory) [points vs time]
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.
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]
Test#2 (FP) [points vs time]
Test#3 (Generic, ZIP) [points vs time]
Test#1 (Memory) [points vs time]
Hardlimit Benchmark Central - Ver. 3.11.4