| | | | | | |

Ryzen 9 3900 vs Core i3-10105


Description
The 3900 is based on Zen 2 architecture while the i3-10105 is based on Comet Lake.

Using the multithread performance as a reference, the 3900 gets a score of 687.5 k points while the i3-10105 gets 253.9 k points.

Summarizing, the 3900 is 2.7 times faster than the i3-10105. To get a proper comparison between both models, take a look to the data shown below.

Specs
CPUID
870f10
a0653
Core
Matisse
Comet Lake-S
Architecture
Base frecuency
3.1 GHz
3.7 GHz
Boost frecuency
4.3 GHz
4.4 GHz
Socket
AM4
LGA 1200
Cores/Threads
12/24
4/8
TDP
65 W
65 W
Cache L1 (d+i)
12x32+12x32 kB
4x32+4x32 kB
Cache L2
12x512 kB
4x256 kB
Cache L3
4x16384 kB
6144 kB
Date
September 2019
March 2021
Mean monothread perf.
74.97k points
68.92k points
Mean multithread perf.
687.5k points
253.93k 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
i3-10105
Test#1 (Integers)
4.39k
4.5k (x1.03)
Test#2 (FP)
16.99k
18.39k (x1.08)
Test#3 (Generic, ZIP)
7.74k
5.78k (x0.75)
Test#1 (Memory)
24.57k
8.3k (x0.34)
TOTAL
53.69k
36.97k (x0.69)

Multithread

3900

i3-10105
Test#1 (Integers)
48.45k
16.13k (x0.33)
Test#2 (FP)
230.81k
75.62k (x0.33)
Test#3 (Generic, ZIP)
122.82k
27.88k (x0.23)
Test#1 (Memory)
54.73k
3.57k (x0.07)
TOTAL
456.8k
123.2k (x0.27)

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
i3-10105
Test#1 (Integers)
16.59k
16.19k (x0.98)
Test#2 (FP)
24.55k
23.07k (x0.94)
Test#3 (Generic, ZIP)
8.55k
6.04k (x0.71)
Test#1 (Memory)
26.28k
8.46k (x0.32)
TOTAL
75.97k
53.77k (x0.71)

Multithread

3900

i3-10105
Test#1 (Integers)
181.33k
57.59k (x0.32)
Test#2 (FP)
261.01k
100.63k (x0.39)
Test#3 (Generic, ZIP)
126.62k
28.59k (x0.23)
Test#1 (Memory)
66.64k
3.56k (x0.05)
TOTAL
635.6k
190.38k (x0.3)

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
i3-10105
Test#1 (Integers)
16.16k
16.11k (x1)
Test#2 (FP)
23.75k
24.12k (x1.02)
Test#3 (Generic, ZIP)
9.32k
5.71k (x0.61)
Test#1 (Memory)
23.17k
7.89k (x0.34)
TOTAL
72.4k
53.83k (x0.74)

Multithread

3900

i3-10105
Test#1 (Integers)
174.73k
58.25k (x0.33)
Test#2 (FP)
275.63k
105.73k (x0.38)
Test#3 (Generic, ZIP)
125.65k
27.8k (x0.22)
Test#1 (Memory)
44.17k
3.51k (x0.08)
TOTAL
620.18k
195.29k (x0.31)

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
i3-10105
Test#1 (Integers)
16.85k
28.9k (x1.71)
Test#2 (FP)
26.03k
25.85k (x0.99)
Test#3 (Generic, ZIP)
9.54k
5.88k (x0.62)
Test#1 (Memory)
22.55k
8.29k (x0.37)
TOTAL
74.97k
68.92k (x0.92)

Multithread

3900

i3-10105
Test#1 (Integers)
229.41k
111.69k (x0.49)
Test#2 (FP)
292.81k
110.23k (x0.38)
Test#3 (Generic, ZIP)
128.48k
28.45k (x0.22)
Test#1 (Memory)
36.8k
3.56k (x0.1)
TOTAL
687.5k
253.93k (x0.37)

Performance/W
3900
i3-10105
Test#1 (Integers)
3529 points/W
1718 points/W
Test#2 (FP)
4505 points/W
1696 points/W
Test#3 (Generic, ZIP)
1977 points/W
438 points/W
Test#1 (Memory)
566 points/W
55 points/W
TOTAL
10577 points/W
3907 points/W

Performance/GHz
3900
i3-10105
Test#1 (Integers)
3920 points/GHz
6568 points/GHz
Test#2 (FP)
6054 points/GHz
5876 points/GHz
Test#3 (Generic, ZIP)
2218 points/GHz
1337 points/GHz
Test#1 (Memory)
5243 points/GHz
1883 points/GHz
TOTAL
17435 points/GHz
15664 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