Ryzen 9 3900 vs 3900X
Description
Both models 3900 and 3900X are based on Zen 2 architecture.
Zen 2 is fabricated on the 7nm process from TSMC and it’s the third generation of Zen CPUs. It comes with 64kB of L1 cache and 512kB of L2 cache per core. Zen 2 CPUs are divided into 3 categories: Matisse (desktop), Rome (Server) and Castle Peak (high-end desktop).
Using the multithread performance as a reference, the 3900 gets a score of 687.5 k points while the 3900X gets 756.3 k points.
Summarizing, the 3900X is 1.1 times faster than the 3900. To get a proper comparison between both models, take a look to the data shown below.
Zen 2 is fabricated on the 7nm process from TSMC and it’s the third generation of Zen CPUs. It comes with 64kB of L1 cache and 512kB of L2 cache per core. Zen 2 CPUs are divided into 3 categories: Matisse (desktop), Rome (Server) and Castle Peak (high-end desktop).
Using the multithread performance as a reference, the 3900 gets a score of 687.5 k points while the 3900X gets 756.3 k points.
Summarizing, the 3900X is 1.1 times faster than the 3900. To get a proper comparison between both models, take a look to the data shown below.
Specs
CPUID
870f10
870f10
Core
Matisse
Matisse
Base frecuency
3.1 GHz
3.8 GHz
Boost frecuency
4.3 GHz
4.6 GHz
Socket
AM4
AM4
Cores/Threads
12/24
12/24
TDP
65 W
105 W
Cache L1 (d+i)
12x32+12x32 kB
12x32+12x32 kB
Cache L2
12x512 kB
12x512 kB
Cache L3
4x16384 kB
4x16384 kB
Date
September 2019
July 2019
Mean monothread perf.
74.97k points
72.51k points
Mean multithread perf.
687.5k points
756.3k 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
3900X
Test#1 (Integers)
4.39k
4.46k (x1.02)
Test#2 (FP)
16.99k
18.46k (x1.09)
Test#3 (Generic, ZIP)
7.74k
8.05k (x1.04)
Test#1 (Memory)
24.57k
23.71k (x0.96)
TOTAL
53.69k
54.68k (x1.02)
Multithread
3900
3900X
Test#1 (Integers)
48.45k
52.65k (x1.09)
Test#2 (FP)
230.81k
265.41k (x1.15)
Test#3 (Generic, ZIP)
122.82k
140.78k (x1.15)
Test#1 (Memory)
54.73k
46.91k (x0.86)
TOTAL
456.8k
505.76k (x1.11)
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
3900X
Test#1 (Integers)
16.59k
17.35k (x1.05)
Test#2 (FP)
24.55k
25.87k (x1.05)
Test#3 (Generic, ZIP)
8.55k
8.72k (x1.02)
Test#1 (Memory)
26.28k
27.86k (x1.06)
TOTAL
75.97k
79.8k (x1.05)
Multithread
3900
3900X
Test#1 (Integers)
181.33k
201.57k (x1.11)
Test#2 (FP)
261.01k
284.41k (x1.09)
Test#3 (Generic, ZIP)
126.62k
145.16k (x1.15)
Test#1 (Memory)
66.64k
100.56k (x1.51)
TOTAL
635.6k
731.7k (x1.15)
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
3900X
Test#1 (Integers)
16.16k
16.95k (x1.05)
Test#2 (FP)
23.75k
25.37k (x1.07)
Test#3 (Generic, ZIP)
9.32k
9.23k (x0.99)
Test#1 (Memory)
23.17k
24.83k (x1.07)
TOTAL
72.4k
76.38k (x1.05)
Multithread
3900
3900X
Test#1 (Integers)
174.73k
196.31k (x1.12)
Test#2 (FP)
275.63k
290.5k (x1.05)
Test#3 (Generic, ZIP)
125.65k
143.95k (x1.15)
Test#1 (Memory)
44.17k
64.92k (x1.47)
TOTAL
620.18k
695.69k (x1.12)
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
3900X
Test#1 (Integers)
16.85k
16.73k (x0.99)
Test#2 (FP)
26.03k
24.74k (x0.95)
Test#3 (Generic, ZIP)
9.54k
8.82k (x0.92)
Test#1 (Memory)
22.55k
22.22k (x0.99)
TOTAL
74.97k
72.51k (x0.97)
Multithread
3900
3900X
Test#1 (Integers)
229.41k
248.24k (x1.08)
Test#2 (FP)
292.81k
309.66k (x1.06)
Test#3 (Generic, ZIP)
128.48k
140.01k (x1.09)
Test#1 (Memory)
36.8k
58.39k (x1.59)
TOTAL
687.5k
756.3k (x1.1)
Performance/W
3900
3900X
Test#1 (Integers)
3529 points/W
2364 points/W
Test#2 (FP)
4505 points/W
2949 points/W
Test#3 (Generic, ZIP)
1977 points/W
1333 points/W
Test#1 (Memory)
566 points/W
556 points/W
TOTAL
10577 points/W
7203 points/W
Performance/GHz
3900
3900X
Test#1 (Integers)
3920 points/GHz
3638 points/GHz
Test#2 (FP)
6054 points/GHz
5377 points/GHz
Test#3 (Generic, ZIP)
2218 points/GHz
1917 points/GHz
Test#1 (Memory)
5243 points/GHz
4831 points/GHz
TOTAL
17435 points/GHz
15763 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