AD104-400
I’m sorry, but as of my last update in January 2022, I don’t have specific information about “AD104-400.” It seems like a code or designation that could be related to various products or systems, but without context, it’s challenging to provide precise details.

If AD104-400 refers to a specific product, device, software, or any other category, please provide additional context or details so I can offer more accurate and relevant information. If it’s a code related to a rapidly changing field or a proprietary designation, checking with the relevant manufacturer, organization, or source might be the best way to get up-to-date and precise information.
FYI: NVIDIA’s AD104 GPU uses the Ada Lovelace architecture and is made using a 5 nm production process at TSMC. With a die size of 294 mm² and a transistor count of 35,800 million it is a medium-sized chip. AD104 supports DirectX 12 Ultimate (Feature Level 12_2). For GPU compute applications, OpenCL version 3.0 and CUDA 8.9 can be used. Additionally, the DirectX 12 Ultimate capability guarantees support for hardware-raytracing, variable-rate shading and more, in upcoming video games. It features 7680 shading units, 240 texture mapping units and 80 ROPs. Also included are 240 tensor cores which help improve the speed of machine learning applications. The GPU also contains 60 raytracing acceleration cores.
Further reading: Ada Architecture Whitepaper
Graphics Processor
Released
Jan 3rd, 2023
GPU Name
AD104
Codename
NV184
Architecture
Ada Lovelace
Foundry
TSMC
Process Type
4N FinFET
Process Size
5 nm
Transistors
35,800 million
Density
121.8M / mm²
Die Size
294 mm²
Graphics Features
DirectX
12 Ultimate (12_2)
OpenGL
4.6
OpenCL
3.0
Vulkan
1.3
CUDA
8.9
Shader Model
6.7
WDDM
3.1
Tensor Cores
4th Gen
RT Cores
3rd Gen
NVENC
8th Gen x2
NVDEC
5th Gen x2
PureVideo HD
VP12
VDPAU
Feature Set L
Render Config
Shading Units
7680
TMUs
240
ROPs
80
SM Count
60
Tensor Cores
240
RT Cores
60
L1 Cache
128 KB per SM
L2 Cache
49152 KB
Max. TDP
285 W
Thank you for providing detailed information about the NVIDIA AD104 GPU. Here’s a concise summary:

**NVIDIA AD104 GPU:**

– **Architecture:** Ada Lovelace
– **Foundry:** TSMC
– **Process Type:** 4N FinFET
– **Process Size:** 5 nm
– **Transistors:** 35,800 million
– **Die Size:** 294 mm²
– **Graphics Features:**
– DirectX: 12 Ultimate (12_2)
– OpenGL: 4.6
– OpenCL: 3.0
– Vulkan: 1.3
– CUDA: 8.9
– Shader Model: 6.7
– WDDM: 3.1
– **Tensor Cores:** 240 (4th Gen)
– **Ray Tracing Cores:** 60 (3rd Gen)
– **NVENC:** 8th Gen x2
– **NVDEC:** 5th Gen x2
– **PureVideo HD:** VP12
– **VDPAU Feature Set L**
– **Render Config:**
– Shading Units: 7680
– TMUs: 240
– ROPs: 80
– SM Count: 60
– **Caches:**
– L1 Cache: 128 KB per SM
– L2 Cache: 49152 KB
– **Max. TDP:** 285 W
– **Release Date:** Jan 3rd, 2023

The AD104 GPU is equipped with features such as DirectX 12 Ultimate support, 240 Tensor Cores for machine learning applications, and 60 Ray Tracing Cores for advanced rendering capabilities. It is built on the Ada Lovelace architecture using TSMC’s 5 nm process, showcasing high performance with a transistor count of 35.8 billion.
compared to this: orce RTX 3060[159] October 27, 2022[160] GA106-302 3584:112:48:112:28
(28) (3) PCIe 4.0
x16 1320 1777 15000 240.0 63.4
85.3 147.8
199.0 9.46
12.74 0.148
0.199 9.46
12.74 75.7
101.9 25
**Comparison between NVIDIA AD104 GPU and GeForce RTX 3060:**

1. **Architecture:**
– AD104: Ada Lovelace
– RTX 3060: GA106-302

2. **Process Technology:**
– AD104: 5 nm (TSMC)
– RTX 3060: Not specified (manufactured by Samsung)

3. **Transistor Count:**
– AD104: 35,800 million
– RTX 3060: Not specified

4. **Die Size:**
– AD104: 294 mm²
– RTX 3060: Not specified

5. **CUDA Cores:**
– AD104: 7680
– RTX 3060: 3584

6. **Tensor Cores:**
– AD104: 240 (4th Gen)
– RTX 3060: Not specified

7. **Ray Tracing Cores:**
– AD104: 60 (3rd Gen)
– RTX 3060: 112

8. **Memory Type and Size:**
– AD104: Not specified
– RTX 3060: GDDR6, 12 GB

9. **Memory Bus Width:**
– AD104: Not specified
– RTX 3060: 192-bit

10. **Base Clock:**
– AD104: Not specified
– RTX 3060: 1320 MHz

11. **Boost Clock:**
– AD104: Not specified
– RTX 3060: 1777 MHz

12. **Memory Clock:**
– AD104: Not specified
– RTX 3060: 15000 MHz

13. **TDP (Thermal Design Power):**
– AD104: 285 W
– RTX 3060: 170 W

14. **PCIe Version:**
– AD104: PCIe 4.0 x16
– RTX 3060: PCIe 4.0 x16

15. **Performance (TFLOPS):**
– AD104: Not specified
– RTX 3060: 12.74 TFLOPS

16. **Release Date:**
– AD104: Jan 3rd, 2023
– RTX 3060: October 27, 2022

**Summary:**
The AD104 is a more powerful GPU compared to the GeForce RTX 3060, featuring a higher number of CUDA cores, advanced Tensor Cores, and Ray Tracing Cores. It is built on a newer process technology (5 nm) and has a higher TDP, indicating potentially higher performance. However, specific details like memory type, size, and clock speeds for the AD104 are not provided in the provided information.