| | | | | | |

Core i7-10870H vs Ryzen 9 3900


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

Using the multithread performance as a reference, the i7-10870H gets a score of 487 k points while the 3900 gets 687.5 k points.

Summarizing, the 3900 is 1.4 times faster than the i7-10870H. To get a proper comparison between both models, take a look to the data shown below.

Specs
CPUID
a0652
870f10
Core
Comet Lake-H
Matisse
Architecture
Base frecuency
2.2 GHz
3.1 GHz
Boost frecuency
5 GHz
4.3 GHz
Socket
FC-BGA 1440
AM4
Cores/Threads
8/16
12/24
TDP
45 W
65 W
Cache L1 (d+i)
8x32+8x32 kB
12x32+12x32 kB
Cache L2
8x256 kB
12x512 kB
Cache L3
16384 kB
4x16384 kB
Date
September 2020
September 2019
Mean monothread perf.
79.2k points
74.97k points
Mean multithread perf.
487.05k points
687.5k 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
i7-10870H
3900
Test#1 (Integers)
4.77k
4.39k (x0.92)
Test#2 (FP)
19.81k
16.99k (x0.86)
Test#3 (Generic, ZIP)
6.29k
7.74k (x1.23)
Test#1 (Memory)
14.03k
24.57k (x1.75)
TOTAL
44.9k
53.69k (x1.2)

Multithread

i7-10870H

3900
Test#1 (Integers)
31.21k
48.45k (x1.55)
Test#2 (FP)
151.01k
230.81k (x1.53)
Test#3 (Generic, ZIP)
50.13k
122.82k (x2.45)
Test#1 (Memory)
8.37k
54.73k (x6.54)
TOTAL
240.71k
456.8k (x1.9)

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
i7-10870H
3900
Test#1 (Integers)
17.17k
16.59k (x0.97)
Test#2 (FP)
25.26k
24.55k (x0.97)
Test#3 (Generic, ZIP)
6.7k
8.55k (x1.27)
Test#1 (Memory)
13.81k
26.28k (x1.9)
TOTAL
62.95k
75.97k (x1.21)

Multithread

i7-10870H

3900
Test#1 (Integers)
116.68k
181.33k (x1.55)
Test#2 (FP)
187.28k
261.01k (x1.39)
Test#3 (Generic, ZIP)
51.52k
126.62k (x2.46)
Test#1 (Memory)
8.39k
66.64k (x7.94)
TOTAL
363.87k
635.6k (x1.75)

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
i7-10870H
3900
Test#1 (Integers)
16.14k
16.16k (x1)
Test#2 (FP)
24.74k
23.75k (x0.96)
Test#3 (Generic, ZIP)
5.98k
9.32k (x1.56)
Test#1 (Memory)
12.7k
23.17k (x1.83)
TOTAL
59.55k
72.4k (x1.22)

Multithread

i7-10870H

3900
Test#1 (Integers)
117.03k
174.73k (x1.49)
Test#2 (FP)
199.88k
275.63k (x1.38)
Test#3 (Generic, ZIP)
50.64k
125.65k (x2.48)
Test#1 (Memory)
8.23k
44.17k (x5.37)
TOTAL
375.77k
620.18k (x1.65)

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
i7-10870H
3900
Test#1 (Integers)
30.38k
16.85k (x0.55)
Test#2 (FP)
27.95k
26.03k (x0.93)
Test#3 (Generic, ZIP)
6.51k
9.54k (x1.47)
Test#1 (Memory)
14.37k
22.55k (x1.57)
TOTAL
79.2k
74.97k (x0.95)

Multithread

i7-10870H

3900
Test#1 (Integers)
216.18k
229.41k (x1.06)
Test#2 (FP)
210.82k
292.81k (x1.39)
Test#3 (Generic, ZIP)
51.57k
128.48k (x2.49)
Test#1 (Memory)
8.47k
36.8k (x4.35)
TOTAL
487.05k
687.5k (x1.41)

Performance/W
i7-10870H
3900
Test#1 (Integers)
4804 points/W
3529 points/W
Test#2 (FP)
4685 points/W
4505 points/W
Test#3 (Generic, ZIP)
1146 points/W
1977 points/W
Test#1 (Memory)
188 points/W
566 points/W
TOTAL
10823 points/W
10577 points/W

Performance/GHz
i7-10870H
3900
Test#1 (Integers)
6075 points/GHz
3920 points/GHz
Test#2 (FP)
5589 points/GHz
6054 points/GHz
Test#3 (Generic, ZIP)
1302 points/GHz
2218 points/GHz
Test#1 (Memory)
2873 points/GHz
5243 points/GHz
TOTAL
15840 points/GHz
17435 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