Here is Kurnal
Thanks for the Die given by the British Shorthair
Let's talk briefly about the Dimensity 9200
First of all, the Die we received is from the Vivo X90
which is a mass-produced version
Its TopDiemark is MT6985, the previous generation Dimensity 9000 was MT6983
Insulting MTK here
Diemark#
This is the Diemark of MTK
24FEB2022
AHJ11296B
The first line is Time
The second line is an unknown rule number
Time is Day Month Year
It can be inferred that it was produced on February 24, 2022
It was released on February 12, 2022
Dieshot#
Let's take a look at the Dieshot
The Diesize is 11.34x10.63mm, which is exactly 120.5
Known TSMC N4 is 146mtr (6T)
There is a difference of 5e
And because of the 9-year compulsory education, the chickens and rabbits are in the same cage
Known 6T density is 146MTR, what about 7.5T?
Calculate the usage of the 7.5T library in D9200
The answer is around 25%-20%
So let's do a graph extraction
CPU#
So this generation of Dimensity uses a 1+3+4 three-cluster architecture
This is its CPU Cluster
The layout is basically the same
What's special is that the sandwich structure in D9K has become a structure similar to "n"
The CPU L3 has become a CPU wrapping around the L3
This may have some improvements in cache access latency
And the layout of the CPU cores has changed from lying down to standing up, reducing the overall width of the CPU Cluster
In terms of L3 Cache
From the previous inability to distinguish between L3 tag logic and irregular cache shapes
It has evolved into a regular-shaped cache with obvious tag logic, and its cache design is more like SDM's design,
In terms of microarchitecture, it has changed from the previous generation's
X2+A710+A510c
to
X3+A715+A510c
The manufacturing process is still TSMC N4, and it is speculated that a density library is used
After all, compared to the core of competitive products of the same generation
Then there is the issue of Core shot
In fact, you can see that the cores of each company are different
Take the commonly used A510 as an example
The L2 Cache, FE, and shared FP below are all different
This is enough to prove that although each company uses Arm's microarchitecture, they have made changes instead of staying the same
The so-called Kyro also proves that it is not purely a public version, and the specific changes are not clear
In terms of SLC
There is no difference from the previous generation, still 2x3M 6M SLC, only the tag logic has changed
GPU#
In terms of GPU, this time it has been upgraded from the previous generation's Mali G710 MP10 to Mali-Immortalis-G715
Although they belong to the same Valhall architecture, there are additions
Among them, Ray Tracing, which is the RTU module,
is placed in the shader core
It only occupies 4% of the shader core area
but achieves a 300% improvement in gaming performance (according to the ppt)
Similarly, in its core configuration
There is no change in G715i compared to G715
Although the G710 doubled the number of single-core compute units compared to the previous generation G78
But in G715, the number of single-core compute units has doubled again
Each core has 128 FMA
Can complete 256 FP32 operations/clock
While the pixel and texture capabilities remain unchanged.
In its texturing unit, there is no change compared to G710
In the G710 architecture
There is no change in its ISA configuration
No changes were found in the GPU Cache in the Dieshot, it is determined to be 3MiB
APU#
In fact, MTK used a large area to write this APU
The previous generation was 4+2
This generation is also 4+2
There is not enough information to see the difference
Only a slight change can be seen in the big core, and the general core has basically not changed
From the official website of MTK:
The sixth-generation MediaTek APU has upgraded the shared memory engine
Allowing the APU's VPU, DMA, and DLA processors to achieve higher computational efficiency.
The new generation network architecture search technology brings better performance and power consumption for machine learning (ML) applications.
Thanks to the eXtreme power-saving mode and APU hardware upgrades
Compared to the fifth-generation APU (APU 590)
The AI performance of APU 690 has increased by up to 35%
AI video super-resolution (AI-SR) energy efficiency has increased by 45%
AI noise reduction (AI-NR) energy efficiency has increased by 30%.
I'm not talented enough to annotate
Modem#
There have actually been some changes in the Modem section
But the GPU's Core is still learning and hasn't learned how to draw the layout yet
So let's just take a look
That's it
Here is Kurnal
Please indicate the source as Kurnal when spreading
Just a side note (Easter egg)
Actually, we also got a media sample of D9200
After X-ray, it was found to be pure plastic without any Si elements
Hilarious
That's it