Intel Xeon D-2795NT vs Google Tensor

Last updated:

CPU comparison with benchmarks

-VS-

CPU lineage

Intel Xeon D-2795NT or Intel Xeon D-2795NT – which processor offers superior performance? In this comparison, we examine disparities and assess which of these two CPUs outperforms the other. We delve into technical specifications and benchmark outcomes.
The Intel Xeon D-2795NT features 20 processor cores and has the capability to manage 40 threads concurrently.
It was released in Q1/2022 and belongs to the 4 generation of the Intel Xeon D series.
To use the Intel Xeon D-2795NT, you'll need a motherboard with a BGA 2579 socket.
The Google Tensor features 8 processor cores and has the capability to manage 8 threads concurrently.
It was released in Q4/2021 and belongs to the 1 generation of the Google Tensor series.
Intel Xeon D-2700 Group Google Tensor
4 Generation 1
Intel Xeon D-2795NT Name Google Tensor
Desktop / Server Segment Mobile
Intel Xeon D Family Google Tensor
 
 

CPU Cores and Base Frequency

The Intel Xeon D-2795NT has 20 CPU cores and can calculate 40 threads in parallel.
The clock frequency of the Intel Xeon D-2795NT is 2.0 GHz
and turbo frequency for one core is 3.1 GHz.
The number of CPU cores greatly affects the speed of the processor and is an important performance indicator.
The Google Tensor has 8 CPU cores and can calculate 8 threads in parallel.
The clock frequency of the A-Core is 1.8 GHz.
None None 1.8 GHz
3.1 GHz Turbo Frequency (1 core) None
None None 2.8 GHz
normal Core architecture hybrid (Prime / big.LITTLE)
2.0 GHz Frequency None
20 CPU Cores 8
No Overclocking No
None None 2x Cortex-X1
None None 4x Cortex-A55
Yes Hyperthreading No
None None 2x Cortex-A76
20x Sunny Cove Cores None
40 Threads 8
 
 

Internal Graphics

The Intel Xeon D-2795NT does not have integrated graphics.
The Google Tensor has integrated graphics, called iGPU for short.
Specifically, the Google Tensor uses the ARM Mali-G78 MP20, which has 320 texture shaders
and 20 execution units.
The iGPU uses the system's main memory as graphics memory and sits on the processor's die.
-- GPU (Turbo) --
-- Technology 5 nm
-- Direct X 12
0 bytes Max. GPU Memory 0 bytes
-- Execution units 20
-- Release date Q4/2021
GPU name ARM Mali-G78 MP20
-- Shaders 320
-- GPU frequency 0.76 GHz
-- Generation Vallhall 2
-- Max. displays --
 
 

Artificial Intelligence and Machine Learning

-- AI specifications Google Edge TPU @ 1.6 TOPS
-- AI hardware Google Tensor AI
 
 

Hardware codec support

A photo or video codec that is accelerated in hardware can greatly accelerate the working speed of a processor and extend the battery life of notebooks or smartphones when playing videos.
-- h265 / HEVC (8 bit) Decode / Encode
-- VP9 Decode / Encode
-- VP8 Decode / Encode
-- VC-1 Decode / Encode
-- JPEG Decode / Encode
-- h264 Decode / Encode
-- h265 / HEVC (10 bit) Decode / Encode
-- AVC Decode / Encode
-- AV1 Decode
 
 

Memory & PCIe

The processor supports a maximum memory capacity of 1.0 TB distributed across 4 memory channels. It offers a peak memory bandwidth of 93.8 GB/s. Both the type and quantity of memory can have a substantial impact on the overall system performance.
The processor supports a maximum memory capacity of 12.0 GB distributed across 2 memory channels. It offers a peak memory bandwidth of 53.0 GB/s. Both the type and quantity of memory can have a substantial impact on the overall system performance.
DDR4-2933 Memory type LPDDR5-5500
93.8 GB/s Bandwidth 53.0 GB/s
Yes AES-NI No
Yes ECC No
pci PCIe pci
1.0 TB Max. Memory 12.0 GB
4 Memory channels 2
 
 

Thermal Management

The processor has a thermal design power (TDP) of 110 W watts.
TDP indicates the cooling solution needed to effectively manage the processor's heat. It generally provides an approximate indication of the actual power consumption of the CPU itself.
The processor has a thermal design power (TDP) of 10 W watts.
110 W TDP (PL1 / PBP) 10 W
-- Tjunction max --
 
 

Technical details

The Intel Xeon D-2795NT is manufactured using a 10 nm process.
A smaller manufacturing process indicates a more contemporary and energy-efficient CPU.
A substantial cache can significantly enhance the processor's performance, particularly in scenarios like gaming.
The Google Tensor is manufactured using a 5 nm process.
VT-x, VT-x EPT, VT-d Virtualization None
10 nm Technology 5 nm
x86-64 (64 bit) Instruction set (ISA) ARMv8-A64 (64 bit)
SSE4.1, SSE4.2, AVX2, AVX-512 ISA extensions
-- Part Number --
Monolithic Chip design --
Q1/2022 Release date Q4/2021
2165 $ Release price --
30.0 MB L2-Cache 8.0 MB
Technical data sheet Documents Technical data sheet
0 bytes L3-Cache 0 bytes
BGA 2579 Socket
Windows 10, Linux Operating systems Android
Ice Lake Architecture G1