Home Tech According to rumors using the TSMC 4N process, NVIDIA Ada Lovelace processors have a crossover advantage over AMD RDNA 3.

According to rumors using the TSMC 4N process, NVIDIA Ada Lovelace processors have a crossover advantage over AMD RDNA 3.

0
According to rumors using the TSMC 4N process, NVIDIA Ada Lovelace processors have a crossover advantage over AMD RDNA 3.

According to rumors using the TSMC 4N process, NVIDIA Ada Lovelace processors have a crossover advantage over AMD RDNA 3.



By NVIDIA Etta Lovelace handlers The next-generation GeForce RTX40 gaming graphics cards will have an edge over AMD’s RDNA3, as previously announced. Moore’s law is dead.

NVIDIA Ada Lovelace GPUs use the TSMC 4N processing terminal, which offers little advantage over AMD’s RDNA 3 GPUs.

From what we know so far, it’s Nvidia Anticipated Use the TSMC 5nm processor terminal for Ada Loveless GPUs running next-generation gaming graphics cards known as the GeForce RTX40 Series. According to a recent rumor, NVIDIA Ada Lovelace GPUs will be based on the TSMC 4N process terminal.

The AMD Ryzen 7000 processors and AM5 operating system only support DDR5 memory and come with EXPO ‘Memory Profile’ technology.

Yes, the same TSMC 4N process node just runs a coil Huber GPUs To the HPC market for data centers. As far as we know about the TSNC 4N process terminal, this is an update of the 5nm process (not to be confused with the 4nm / N4, which is a completely different terminal). The TSMC 4N process node is specifically designed for Nvidia and offers various enhancements that allow for improved power efficiency, performance and smaller density enhancements compared to the 5nm vanilla TSNC node.

The reasons why NVIDIA chose TSMC’s 4N as a candidate for the next generation of Matrix GPUs are clear. The following cards There will really be greed for power NVIDIA and company will work with the 4N process node to improve it as much as possible. AMD, on the other hand, uses a combination of TSMC It was cut with 5 nm and 6 nm process Based on upcoming MCM and monolithic GPUs RDNA Graphics Engineering3 Although they do not bring the improvements provided by 4N, they do offer MCM approach, which is expected to be very effective.

So at the end of the day, Nvidia gets a better edge, while AMD offers a better design approach. At the end of the day, these may not be important for end users who want to play their games only on the best hardware (graphics cards).

NVIDIA CUDA GPU (Rumors):

GPU TU102 GA102 M102
Flag SKU RTX 2080 Ti RTX 3090 Ti RTX 4090?
Architectural Engineering Touring Ampere Ada Lovelace
Treatment TSMC 12nm NFF Samsung 8nm TSMC 4N?
The amount of death 754 mm 628 mm ~ 600 mm
Graphics Processing Groups (GPC) 6 7 12
Textile Processing Tools (TPC) 36 42 72
Multiprocessor flow (SM) 72 84 144
CUDA cores 4608 10752 18432
L2. Temporary storage 6 MB 6 MB 96 MB
Theoretical TFLOPs 16 TFLOPs 40 TFLOPs ~ 90 TFLOPs?
Memory type GDDR6 GDDR6X GDDR6X
Memory capacity 11GB (2080D) 24 GB (3090 Ti) 24 GB (4090?)
Memory speed 14 Gbps 21 Gbps 24 Gbps?
Memory bandwidth 616 GB / sec 1,008 GB / sec 1152 GB / V?
Memory bus 384 bit 384 bit 384 bit
PCIe interface PCIe General 3.0 PCIe General 4.0 PCIe General 4.0
DGP 250 watts 350 watts 600 watts?
Release September 2018 September 20 2H 2022 (TBC)

See also  How to see the specifications of Apple Conference 2022 and expected iPhone | Technology

LEAVE A REPLY

Please enter your comment!
Please enter your name here