| | | | | | |

Ryzen 7 2700X vs Ryzen 9 3900


Description
The 2700X is based on Zen+ architecture while the 3900 is based on Zen 2.

Using the multithread performance as a reference, the 2700X gets a score of 431.1 k points while the 3900 gets 687.5 k points.

Summarizing, the 3900 is 1.6 times faster than the 2700X. To get a proper comparison between both models, take a look to the data shown below.

Specs
CPUID
800f82
870f10
Core
Pinnacle Ridge
Matisse
Architecture
Base frecuency
3.7 GHz
3.1 GHz
Boost frecuency
4.3 GHz
4.3 GHz
Socket
AM4
AM4
Cores/Threads
8/16
12/24
TDP
105 W
65 W
Cache L1 (d+i)
8x64+8x32 kB
12x32+12x32 kB
Cache L2
8x512 kB
12x512 kB
Cache L3
16384 kB
4x16384 kB
Date
April 2018
September 2019
Mean monothread perf.
64.83k points
74.97k points
Mean multithread perf.
431.14k 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
2700X
3900
Test#1 (Integers)
4.25k
4.39k (x1.03)
Test#2 (FP)
19.72k
16.99k (x0.86)
Test#3 (Generic, ZIP)
5.84k
7.74k (x1.33)
Test#1 (Memory)
21.36k
24.57k (x1.15)
TOTAL
51.17k
53.69k (x1.05)

Multithread

2700X

3900
Test#1 (Integers)
35.04k
48.45k (x1.38)
Test#2 (FP)
181.42k
230.81k (x1.27)
Test#3 (Generic, ZIP)
64.86k
122.82k (x1.89)
Test#1 (Memory)
8.21k
54.73k (x6.67)
TOTAL
289.54k
456.8k (x1.58)

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
2700X
3900
Test#1 (Integers)
15.22k
16.59k (x1.09)
Test#2 (FP)
24.12k
24.55k (x1.02)
Test#3 (Generic, ZIP)
5.87k
8.55k (x1.46)
Test#1 (Memory)
21k
26.28k (x1.25)
TOTAL
66.21k
75.97k (x1.15)

Multithread

2700X

3900
Test#1 (Integers)
126.81k
181.33k (x1.43)
Test#2 (FP)
229.86k
261.01k (x1.14)
Test#3 (Generic, ZIP)
61.74k
126.62k (x2.05)
Test#1 (Memory)
9.77k
66.64k (x6.82)
TOTAL
428.19k
635.6k (x1.48)

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
2700X
3900
Test#1 (Integers)
14.44k
16.16k (x1.12)
Test#2 (FP)
24.92k
23.75k (x0.95)
Test#3 (Generic, ZIP)
5.8k
9.32k (x1.61)
Test#1 (Memory)
19.07k
23.17k (x1.21)
TOTAL
64.24k
72.4k (x1.13)

Multithread

2700X

3900
Test#1 (Integers)
122.01k
174.73k (x1.43)
Test#2 (FP)
220.34k
275.63k (x1.25)
Test#3 (Generic, ZIP)
59.89k
125.65k (x2.1)
Test#1 (Memory)
9.9k
44.17k (x4.46)
TOTAL
412.13k
620.18k (x1.5)

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
2700X
3900
Test#1 (Integers)
15.34k
16.85k (x1.1)
Test#2 (FP)
25.75k
26.03k (x1.01)
Test#3 (Generic, ZIP)
5.79k
9.54k (x1.65)
Test#1 (Memory)
17.95k
22.55k (x1.26)
TOTAL
64.83k
74.97k (x1.16)

Multithread

2700X

3900
Test#1 (Integers)
123.07k
229.41k (x1.86)
Test#2 (FP)
239.3k
292.81k (x1.22)
Test#3 (Generic, ZIP)
60.6k
128.48k (x2.12)
Test#1 (Memory)
8.18k
36.8k (x4.5)
TOTAL
431.14k
687.5k (x1.59)

Performance/W
2700X
3900
Test#1 (Integers)
1172 points/W
3529 points/W
Test#2 (FP)
2279 points/W
4505 points/W
Test#3 (Generic, ZIP)
577 points/W
1977 points/W
Test#1 (Memory)
78 points/W
566 points/W
TOTAL
4106 points/W
10577 points/W

Performance/GHz
2700X
3900
Test#1 (Integers)
3568 points/GHz
3920 points/GHz
Test#2 (FP)
5987 points/GHz
6054 points/GHz
Test#3 (Generic, ZIP)
1347 points/GHz
2218 points/GHz
Test#1 (Memory)
4175 points/GHz
5243 points/GHz
TOTAL
15078 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