| | | | | | |

Ryzen 9 3900X vs Ryzen 7 2700X


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

Using the multithread performance as a reference, the 3900X gets a score of 756.3 k points while the 2700X gets 431.1 k points.

Summarizing, the 3900X is 1.8 times faster than the 2700X. To get a proper comparison between both models, take a look to the data shown below.

Specs
CPUID
870f10
800f82
Core
Matisse
Pinnacle Ridge
Architecture
Base frecuency
3.8 GHz
3.7 GHz
Boost frecuency
4.6 GHz
4.3 GHz
Socket
AM4
AM4
Cores/Threads
12/24
8/16
TDP
105 W
105 W
Cache L1 (d+i)
12x32+12x32 kB
8x64+8x32 kB
Cache L2
12x512 kB
8x512 kB
Cache L3
4x16384 kB
16384 kB
Date
July 2019
April 2018
Mean monothread perf.
72.51k points
64.83k points
Mean multithread perf.
756.3k points
431.14k 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
3900X
2700X
Test#1 (Integers)
4.46k
4.25k (x0.95)
Test#2 (FP)
18.46k
19.72k (x1.07)
Test#3 (Generic, ZIP)
8.05k
5.84k (x0.73)
Test#1 (Memory)
23.71k
21.36k (x0.9)
TOTAL
54.68k
51.17k (x0.94)

Multithread

3900X

2700X
Test#1 (Integers)
52.65k
35.04k (x0.67)
Test#2 (FP)
265.41k
181.42k (x0.68)
Test#3 (Generic, ZIP)
140.78k
64.86k (x0.46)
Test#1 (Memory)
46.91k
8.21k (x0.17)
TOTAL
505.76k
289.54k (x0.57)

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
3900X
2700X
Test#1 (Integers)
17.35k
15.22k (x0.88)
Test#2 (FP)
25.87k
24.12k (x0.93)
Test#3 (Generic, ZIP)
8.72k
5.87k (x0.67)
Test#1 (Memory)
27.86k
21k (x0.75)
TOTAL
79.8k
66.21k (x0.83)

Multithread

3900X

2700X
Test#1 (Integers)
201.57k
126.81k (x0.63)
Test#2 (FP)
284.41k
229.86k (x0.81)
Test#3 (Generic, ZIP)
145.16k
61.74k (x0.43)
Test#1 (Memory)
100.56k
9.77k (x0.1)
TOTAL
731.7k
428.19k (x0.59)

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
3900X
2700X
Test#1 (Integers)
16.95k
14.44k (x0.85)
Test#2 (FP)
25.37k
24.92k (x0.98)
Test#3 (Generic, ZIP)
9.23k
5.8k (x0.63)
Test#1 (Memory)
24.83k
19.07k (x0.77)
TOTAL
76.38k
64.24k (x0.84)

Multithread

3900X

2700X
Test#1 (Integers)
196.31k
122.01k (x0.62)
Test#2 (FP)
290.5k
220.34k (x0.76)
Test#3 (Generic, ZIP)
143.95k
59.89k (x0.42)
Test#1 (Memory)
64.92k
9.9k (x0.15)
TOTAL
695.69k
412.13k (x0.59)

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
3900X
2700X
Test#1 (Integers)
16.73k
15.34k (x0.92)
Test#2 (FP)
24.74k
25.75k (x1.04)
Test#3 (Generic, ZIP)
8.82k
5.79k (x0.66)
Test#1 (Memory)
22.22k
17.95k (x0.81)
TOTAL
72.51k
64.83k (x0.89)

Multithread

3900X

2700X
Test#1 (Integers)
248.24k
123.07k (x0.5)
Test#2 (FP)
309.66k
239.3k (x0.77)
Test#3 (Generic, ZIP)
140.01k
60.6k (x0.43)
Test#1 (Memory)
58.39k
8.18k (x0.14)
TOTAL
756.3k
431.14k (x0.57)

Performance/W
3900X
2700X
Test#1 (Integers)
2364 points/W
1172 points/W
Test#2 (FP)
2949 points/W
2279 points/W
Test#3 (Generic, ZIP)
1333 points/W
577 points/W
Test#1 (Memory)
556 points/W
78 points/W
TOTAL
7203 points/W
4106 points/W

Performance/GHz
3900X
2700X
Test#1 (Integers)
3638 points/GHz
3568 points/GHz
Test#2 (FP)
5377 points/GHz
5987 points/GHz
Test#3 (Generic, ZIP)
1917 points/GHz
1347 points/GHz
Test#1 (Memory)
4831 points/GHz
4175 points/GHz
TOTAL
15763 points/GHz
15078 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