| | | | | | |

Ryzen 9 3900X vs Core i7-10870H


Description
The 3900X is based on Zen 2 architecture while the i7-10870H is based on Comet Lake.

Using the multithread performance as a reference, the 3900X gets a score of 756.3 k points while the i7-10870H gets 487 k points.

Summarizing, the 3900X is 1.6 times faster than the i7-10870H. To get a proper comparison between both models, take a look to the data shown below.

Specs
CPUID
870f10
a0652
Core
Matisse
Comet Lake-H
Architecture
Base frecuency
3.8 GHz
2.2 GHz
Boost frecuency
4.6 GHz
5 GHz
Socket
AM4
FC-BGA 1440
Cores/Threads
12/24
8/16
TDP
105 W
45 W
Cache L1 (d+i)
12x32+12x32 kB
8x32+8x32 kB
Cache L2
12x512 kB
8x256 kB
Cache L3
4x16384 kB
16384 kB
Date
July 2019
September 2020
Mean monothread perf.
72.51k points
79.2k points
Mean multithread perf.
756.3k points
487.05k 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
i7-10870H
Test#1 (Integers)
4.46k
4.77k (x1.07)
Test#2 (FP)
18.46k
19.81k (x1.07)
Test#3 (Generic, ZIP)
8.05k
6.29k (x0.78)
Test#1 (Memory)
23.71k
14.03k (x0.59)
TOTAL
54.68k
44.9k (x0.82)

Multithread

3900X

i7-10870H
Test#1 (Integers)
52.65k
31.21k (x0.59)
Test#2 (FP)
265.41k
151.01k (x0.57)
Test#3 (Generic, ZIP)
140.78k
50.13k (x0.36)
Test#1 (Memory)
46.91k
8.37k (x0.18)
TOTAL
505.76k
240.71k (x0.48)

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
i7-10870H
Test#1 (Integers)
17.35k
17.17k (x0.99)
Test#2 (FP)
25.87k
25.26k (x0.98)
Test#3 (Generic, ZIP)
8.72k
6.7k (x0.77)
Test#1 (Memory)
27.86k
13.81k (x0.5)
TOTAL
79.8k
62.95k (x0.79)

Multithread

3900X

i7-10870H
Test#1 (Integers)
201.57k
116.68k (x0.58)
Test#2 (FP)
284.41k
187.28k (x0.66)
Test#3 (Generic, ZIP)
145.16k
51.52k (x0.35)
Test#1 (Memory)
100.56k
8.39k (x0.08)
TOTAL
731.7k
363.87k (x0.5)

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
i7-10870H
Test#1 (Integers)
16.95k
16.14k (x0.95)
Test#2 (FP)
25.37k
24.74k (x0.98)
Test#3 (Generic, ZIP)
9.23k
5.98k (x0.65)
Test#1 (Memory)
24.83k
12.7k (x0.51)
TOTAL
76.38k
59.55k (x0.78)

Multithread

3900X

i7-10870H
Test#1 (Integers)
196.31k
117.03k (x0.6)
Test#2 (FP)
290.5k
199.88k (x0.69)
Test#3 (Generic, ZIP)
143.95k
50.64k (x0.35)
Test#1 (Memory)
64.92k
8.23k (x0.13)
TOTAL
695.69k
375.77k (x0.54)

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
i7-10870H
Test#1 (Integers)
16.73k
30.38k (x1.82)
Test#2 (FP)
24.74k
27.95k (x1.13)
Test#3 (Generic, ZIP)
8.82k
6.51k (x0.74)
Test#1 (Memory)
22.22k
14.37k (x0.65)
TOTAL
72.51k
79.2k (x1.09)

Multithread

3900X

i7-10870H
Test#1 (Integers)
248.24k
216.18k (x0.87)
Test#2 (FP)
309.66k
210.82k (x0.68)
Test#3 (Generic, ZIP)
140.01k
51.57k (x0.37)
Test#1 (Memory)
58.39k
8.47k (x0.14)
TOTAL
756.3k
487.05k (x0.64)

Performance/W
3900X
i7-10870H
Test#1 (Integers)
2364 points/W
4804 points/W
Test#2 (FP)
2949 points/W
4685 points/W
Test#3 (Generic, ZIP)
1333 points/W
1146 points/W
Test#1 (Memory)
556 points/W
188 points/W
TOTAL
7203 points/W
10823 points/W

Performance/GHz
3900X
i7-10870H
Test#1 (Integers)
3638 points/GHz
6075 points/GHz
Test#2 (FP)
5377 points/GHz
5589 points/GHz
Test#3 (Generic, ZIP)
1917 points/GHz
1302 points/GHz
Test#1 (Memory)
4831 points/GHz
2873 points/GHz
TOTAL
15763 points/GHz
15840 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