NVIDIA Flagship “Ada Lovelace” GPU Rumored To Feature 18176 Cores, 48 GB Memory, 24 Gbps Speeds & 800W TBP

OSTN Staff

NVIDIA Ada Lovelace 'GeForce RTX 40' GPUs Power Limits Detailed: AD102 at 800W, AD103 at 450W, AD104 at 400W, AD106 at 260W

Details regarding the flagship NVIDIA Ada Lovelace GPU SKU have been reported by Kopite7kimi which makes it seem like the ultimate graphics chip ever made.

NVIDIA Ada Lovelace Flagship GPU Rumored To Feature AD102-450 GPU With 18176 Cores, 48 GB “24 Gbps” GDDR6X Memory & 800W TBP

This isn’t the first time that such a high-end Ada Lovelace GPU SKU has been talked about. The previous rumor, which also came from the same leaker, reported on a Titan-Class graphics card within the Ada Lovelace GPU lineup which would feature some insane specifications. Once again, this isn’t the fully enabled AD102 GPU while the previous variant was mentioned as a 900W TBP SKU with the full 18432 CUDA Cores.

According to the rumored specs, the graphics card will be rocking the Ada Lovelace architecture & feature a slightly cut-down configuration “AD102-450-A1”, rocking 142 SMs (out of 144 SMs) on 18,432 CUDA cores (out of 18432 CUDA cores). Based on a clock speed of around 3 GHz, this graphics card will easily break past the 100 TFLOPs Compute barrier. The graphics card is said to be equipped with 48 GB of GDDR6X memory running across a 384-bit bus interface.

What’s interesting is that NVIDIA will not hold back on the VRAM specs and feature the latest 24 Gbps memory modules, delivering up to 1.152 TB/s of VRAM bandwidth to the GPU. That’s a 14% increase in memory bandwidth compared to the existing RTX 3090 Ti flagship which features 21 Gbps memory dies. The upcoming RTX 4090 is also expected to utilize the same 21 Gbps memory dies with only the flagship ‘Ti’ model getting the 24 Gbps dies.

As for power consumption, the new NVIDIA Flagship AD102 GPU-powered graphics card will be insanity, with almost double the TDP of the RTX 3090 Ti, rated at up to 800W. Considering that a single 16-pin connector can only provide 600 Watts of power, a dual 16-pin connector config will have to be utilized for this monster of a card if it ever becomes a reality. The graphcis card may utilize the PG137-SKU0

Based on the Ampere lineup, we saw that NVIDIA not only didn’t release a Titan graphics card but essentially replaced the Titan series with its BFGPU-class GeForce RTX lineup. The higher capacity cards still launched as the workstation RTX A**** lineup which also got the full GA102 treatment but besides the RTX 3090 Ti and the RTX A6000, there was no Titan class. So will a Titan-class GPU make sense for Ada Lovelace or will this specific SKU end up as the next-gen gaming BFGPU and workstation flagship? Well, we can’t say for sure but one thing is definitely correct such a GPU configuration will indeed be insane in terms of specifications, power draw, and price. The card, if it ever comes to retail, will definitely launch after the RTX 4090 which is expected to be unveiled later this fall.

NVIDIA GeForce RTX 40 Series Graphics Card Lineup (Rumored):

Graphics Card GPU PCB Variant SM Units / Cores Memory / Bus Memory Clock / Bandwidth TGP Power Connectors Launch
NVIDIA Titan A / GeForce RTX 40? AD102-450? PG137-SKU0 142/ 18176? 48 GB / 384-bit 24 Gbps / 1.15 TB/s ~800W 2x 16-pin TBD
NVIDIA GeForce RTX 4090 Ti AD102-350? TBD 144 / 18432? 24 GB / 384-bit 24 Gbps / 1.15 TB/s ~600W 1x 16-pin TBD
NVIDIA GeForce RTX 4090 AD102-300? PG137/139 SKU330 128 / 16384? 24 GB / 384-bit 21 Gbps / 1.00 TB/s ~450W 1x 16-pin Q4 2022
NVIDIA GeForce RTX 4080 AD103-300? PG13*/139 SKU360 80 / 10240? 16 GB / 256-bit 21 Gbps / 672 GB/s ~420W 1x 16-pin Q4 2022
NVIDIA GeForce RTX 4070 AD104-275? PG141-310 SKU341 56 / 7168? 10 GB / 160-bit 18 Gbps / 360 GB/s ~300W 1x 16-pin Q4 2022
NVIDIA GeForce RTX 4060 AD106-***? TBD >36 / 4608? 8 GB / 128-bit TBD ~200W 1 x 16-pin Q1 2023

Which NVIDIA GeForce RTX 40 series graphics card are you looking forward to the most?

The post NVIDIA Flagship “Ada Lovelace” GPU Rumored To Feature 18176 Cores, 48 GB Memory, 24 Gbps Speeds & 800W TBP by Hassan Mujtaba appeared first on Wccftech.

Powered by WPeMatico

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.