| | | | | | |

Ryzen 5 2400G vs 3600


Description
The 2400G is based on Zen architecture while the 3600 is based on Zen 2.

Using the multithread performance as a reference, the 2400G gets a score of 198.3 k points while the 3600 gets 348.4 k points.

Summarizing, the 3600 is 1.8 times faster than the 2400G. To get a proper comparison between both models, take a look to the data shown below.

Specs
CPUID
810f10
870f10
Core
Raven Ridge
Matisse
Architecture
Base frecuency
3.6 GHz
3.6 GHz
Boost frecuency
3.9 GHz
4.2 GHz
Socket
AM4
AM4
Cores/Threads
4/8
6/12
TDP
65 W
65 W
Cache L1 (d+i)
4x64+4x32 kB
6x32+6x32 kB
Cache L2
4x512 kB
6x512 kB
Cache L3
4096 kB
32768 kB
Date
January 2018
July 2019
Mean monothread perf.
47.96k points
70.55k points
Mean multithread perf.
198.27k points
348.35k 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
2400G
3600
Test#1 (Integers)
3.8k
4.2k (x1.11)
Test#2 (FP)
17.38k
17.36k (x1)
Test#3 (Generic, ZIP)
5.17k
7.5k (x1.45)
Test#1 (Memory)
3.16k
23.79k (x7.52)
TOTAL
29.52k
52.85k (x1.79)

Multithread

2400G

3600
Test#1 (Integers)
15.28k
22.11k (x1.45)
Test#2 (FP)
76.44k
108.04k (x1.41)
Test#3 (Generic, ZIP)
27.48k
56.81k (x2.07)
Test#1 (Memory)
3.02k
36.39k (x12.07)
TOTAL
122.2k
223.36k (x1.83)

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
2400G
3600
Test#1 (Integers)
13.95k
15.38k (x1.1)
Test#2 (FP)
20.7k
21.94k (x1.06)
Test#3 (Generic, ZIP)
5.33k
7.97k (x1.5)
Test#1 (Memory)
3.03k
24.76k (x8.18)
TOTAL
43.01k
70.05k (x1.63)

Multithread

2400G

3600
Test#1 (Integers)
58.56k
96.91k (x1.65)
Test#2 (FP)
92.95k
135.12k (x1.45)
Test#3 (Generic, ZIP)
28.55k
70.39k (x2.47)
Test#1 (Memory)
2.98k
56.43k (x18.97)
TOTAL
183.04k
358.85k (x1.96)

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
2400G
3600
Test#1 (Integers)
13.04k
15.04k (x1.15)
Test#2 (FP)
22.35k
20.95k (x0.94)
Test#3 (Generic, ZIP)
5.37k
8.57k (x1.6)
Test#1 (Memory)
3.4k
21.71k (x6.39)
TOTAL
44.15k
66.27k (x1.5)

Multithread

2400G

3600
Test#1 (Integers)
56.49k
94.36k (x1.67)
Test#2 (FP)
100.74k
141.73k (x1.41)
Test#3 (Generic, ZIP)
28.21k
69.38k (x2.46)
Test#1 (Memory)
3.04k
32.72k (x10.76)
TOTAL
188.49k
338.19k (x1.79)

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
2400G
3600
Test#1 (Integers)
14.2k
16.04k (x1.13)
Test#2 (FP)
23.23k
24.47k (x1.05)
Test#3 (Generic, ZIP)
5.35k
8.38k (x1.57)
Test#1 (Memory)
5.18k
21.66k (x4.19)
TOTAL
47.96k
70.55k (x1.47)

Multithread

2400G

3600
Test#1 (Integers)
58.24k
113.63k (x1.95)
Test#2 (FP)
105.72k
143.55k (x1.36)
Test#3 (Generic, ZIP)
28.73k
63.84k (x2.22)
Test#1 (Memory)
5.59k
27.34k (x4.89)
TOTAL
198.27k
348.35k (x1.76)

Performance/W
2400G
3600
Test#1 (Integers)
896 points/W
1748 points/W
Test#2 (FP)
1626 points/W
2208 points/W
Test#3 (Generic, ZIP)
442 points/W
982 points/W
Test#1 (Memory)
86 points/W
421 points/W
TOTAL
3050 points/W
5359 points/W

Performance/GHz
2400G
3600
Test#1 (Integers)
3641 points/GHz
3819 points/GHz
Test#2 (FP)
5957 points/GHz
5825 points/GHz
Test#3 (Generic, ZIP)
1372 points/GHz
1995 points/GHz
Test#1 (Memory)
1327 points/GHz
5158 points/GHz
TOTAL
12298 points/GHz
16797 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