MediaTek Dimensity 920
VS
Qualcomm Snapdragon 870
MediaTek Dimensity 920
VS
Qualcomm Snapdragon 870

Which to select

It is time to pick the winner. What is the difference between MediaTek Dimensity 920 vs Qualcomm Snapdragon 870? What CPU is more powerful? It is quite easy to determine – look at comparison table. The processor with more cores/ threads and also with higher frequency is the absolute winner!

CPU Cores and Base Frequency

Who will win between MediaTek Dimensity 920 vs  Qualcomm Snapdragon 870. The general performance of a CPU can easily be determined based on the number of its cores and the thread count, as well as the base frequency and Turbo frequency. The more GHz and cores a CPU has, the better. Please note that high technical specs require using a powerful cooling system.

2.00 GHz
Frequency
1.80 GHz
8
CPU Cores
8
2.50 GHz
Turbo (1 Core)
3.20 GHz
No
Hyperthreading
No
No
Overclocking
No
2.00 GHz
Turbo (8 Cores)
2.42 GHz
hybrid (big.LITTLE)
Core architecture
hybrid (Prime / big.LITTLE)
2x Cortex-A78
A core
1x Cortex-A77
6x Cortex-A55
B core
3x Cortex-A77
--
C core
4x Cortex-A55

CPU generation and family

MediaTek Dimensity 920
Name
Qualcomm Snapdragon 870
Mobile
Segment
Mobile
MediaTek Dimensity 900/920
CPU group
Qualcomm Snapdragon 865/870
Mediatek Dimensity
Family
Qualcomm Snapdragon
2
Generation
4
--
Predecessor
--
--
Successor
--

Internal Graphics

Some manufacturers complement their CPUs with graphic chips, such a solution being especially popular in laptops. The higher the clock frequency of a GPU is and the bigger its memory, the better. Find a winner - MediaTek Dimensity 920 vs Qualcomm Snapdragon 870. 

ARM Mali-G68 MP4
GPU name
Qualcomm Adreno 650
GPU frequency
0.25 GHz
No turbo
GPU (Turbo)
0.67 GHz
Vallhall 2
Generation
6
12
DirectX Version
12.0
4
Execution units
2
64
Shader
512
1
Max. displays
1
6 nm
Technology
7 nm
Q2/2020
Release date
Q4/2019
--
Max. GPU Memory
--

Hardware codec support

Here we deal with specs that are used by some CPU manufacturers. These numbers are mainly technical and can be neglected for the purpose of the comparison analysis.

Decode / Encode
h264
Decode / Encode
Decode / Encode
JPEG
Decode / Encode
Decode / Encode
VP8
Decode / Encode
Decode / Encode
VP9
Decode / Encode
Decode / Encode
VC-1
Decode
Decode / Encode
AVC
Decode
Decode / Encode
h265 / HEVC (8 bit)
Decode / Encode
Decode / Encode
h265 / HEVC (10 bit)
Decode / Encode
Decode
AV1
No

Memory & PCIe

These are memory standards supported by CPUs. The higher such standards, the better a CPU’s performance is.

no data
Memory type
LPDDR4X-4266 LPDDR5-5500
no data
Max. Memory
16 GB
no data
Memory channels
4
no data
ECC
No

Encryption

Data encryption support

No
AES-NI
No

Memory & AMP; PCIe

LPDDR4X LPDDR5
Memory type
no data
16 GB
Max. Memory
no data
No
ECC
no data
4
Memory channels
no data

Thermal Management

The thermal design power (TDP), sometimes called thermal design point, is the maximum amount of heat generated by a computer chip or component (often a CPU, GPU or system on a chip) that the cooling system in a computer is designed to dissipate under any workload.

--
Tjunction max.
--
--
TDP up
--
--
TDP down
--
10 W
TDP (PL1)
10 W
--
TDP (PL2)
--

Technical details

8
CPU Threads
8
2.00 MB
L3-Cache
3.00 MB
6 nm
Technology
7 nm
Cortex-A78 / Cortex-A55
Architecture
Kryo 585
None
Virtualization
None
N/A
Socket
N/A
Q3/2021
Release date
Q2/2021
ARMv8-A64 (64 bit)
Instruction set (ISA)
x86-64 (64 bit)
--
L2-Cache
2.00 MB
--
Part Number
SM8250-AC

Devices using this processor

You probably know already what devices use CPUs. These can be a desktop or a laptop.

Unknown
Used in
Unknown

Compatibility

Technologies and extensions

Virtualization technologies

Memory specs

Peripherals

AnTuTu 8 benchmark

Phones with best AnTuTu scores. These cell phones are the ones that have obtained the highest score in the well-known AnTuTu ranking

iGPU - FP32 Performance (Single-precision GFLOPS)

This test serves for determining the performance of integrated graphics in Intel and AMD processors. The result is the estimated computing power in the Single-Precision FP32 mode