| | | | | | |

Ryzen 9 3900 vs Core i5-12500H


Description
The 3900 is based on Zen 2 architecture while the i5-12500H is based on Alder Lake.

Using the multithread performance as a reference, the 3900 gets a score of 687.5 k points while the i5-12500H gets 225.7 k points.

Summarizing, the 3900 is 3 times faster than the i5-12500H. To get a proper comparison between both models, take a look to the data shown below.

Specs
CPUID
870f10
906a3
Core
Matisse
Alder Lake-H
Architecture
Base frecuency
3.1 GHz
2.5 GHz
Boost frecuency
4.3 GHz
4.5 GHz
Socket
AM4
BGA 1744
Cores/Threads
12/24
12/16
TDP
65 W
45 W
Cache L1 (d+i)
12x32+12x32 kB
4x32/8X64+4x48/8X32 kB
Cache L2
12x512 kB
4x1280/2x2048 kB
Cache L3
4x16384 kB
18432 kB
Date
September 2019
February 2022
Mean monothread perf.
74.97k points
58.37k points
Mean multithread perf.
687.5k points
225.65k 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
i5-12500H
Test#1 (Integers)
4.39k
6.93k (x1.58)
Test#2 (FP)
16.99k
18.76k (x1.1)
Test#3 (Generic, ZIP)
7.74k
9.46k (x1.22)
Test#1 (Memory)
24.57k
8.17k (x0.33)
TOTAL
53.69k
43.33k (x0.81)

Multithread

3900

i5-12500H
Test#1 (Integers)
48.45k
18.04k (x0.37)
Test#2 (FP)
230.81k
66.78k (x0.29)
Test#3 (Generic, ZIP)
122.82k
37.55k (x0.31)
Test#1 (Memory)
54.73k
6.08k (x0.11)
TOTAL
456.8k
128.45k (x0.28)

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
i5-12500H
Test#1 (Integers)
16.59k
21.93k (x1.32)
Test#2 (FP)
24.55k
18.79k (x0.77)
Test#3 (Generic, ZIP)
8.55k
10.5k (x1.23)
Test#1 (Memory)
26.28k
8.11k (x0.31)
TOTAL
75.97k
59.33k (x0.78)

Multithread

3900

i5-12500H
Test#1 (Integers)
181.33k
61.07k (x0.34)
Test#2 (FP)
261.01k
86.08k (x0.33)
Test#3 (Generic, ZIP)
126.62k
41.08k (x0.32)
Test#1 (Memory)
66.64k
6.02k (x0.09)
TOTAL
635.6k
194.25k (x0.31)

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
i5-12500H
Test#1 (Integers)
16.16k
21.11k (x1.31)
Test#2 (FP)
23.75k
19.61k (x0.83)
Test#3 (Generic, ZIP)
9.32k
9.53k (x1.02)
Test#1 (Memory)
23.17k
7.72k (x0.33)
TOTAL
72.4k
57.97k (x0.8)

Multithread

3900

i5-12500H
Test#1 (Integers)
174.73k
61.44k (x0.35)
Test#2 (FP)
275.63k
93.66k (x0.34)
Test#3 (Generic, ZIP)
125.65k
40.09k (x0.32)
Test#1 (Memory)
44.17k
6.01k (x0.14)
TOTAL
620.18k
201.2k (x0.32)

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
i5-12500H
Test#1 (Integers)
16.85k
22.61k (x1.34)
Test#2 (FP)
26.03k
17.73k (x0.68)
Test#3 (Generic, ZIP)
9.54k
9.63k (x1.01)
Test#1 (Memory)
22.55k
8.4k (x0.37)
TOTAL
74.97k
58.37k (x0.78)

Multithread

3900

i5-12500H
Test#1 (Integers)
229.41k
85.49k (x0.37)
Test#2 (FP)
292.81k
94.93k (x0.32)
Test#3 (Generic, ZIP)
128.48k
39.15k (x0.3)
Test#1 (Memory)
36.8k
6.08k (x0.17)
TOTAL
687.5k
225.65k (x0.33)

Performance/W
3900
i5-12500H
Test#1 (Integers)
3529 points/W
1900 points/W
Test#2 (FP)
4505 points/W
2110 points/W
Test#3 (Generic, ZIP)
1977 points/W
870 points/W
Test#1 (Memory)
566 points/W
135 points/W
TOTAL
10577 points/W
5014 points/W

Performance/GHz
3900
i5-12500H
Test#1 (Integers)
3920 points/GHz
5025 points/GHz
Test#2 (FP)
6054 points/GHz
3939 points/GHz
Test#3 (Generic, ZIP)
2218 points/GHz
2139 points/GHz
Test#1 (Memory)
5243 points/GHz
1867 points/GHz
TOTAL
17435 points/GHz
12971 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