NVIDIA Confirms the specs of its AD102, AD103 and AD104 GPUs

The die sizes and transistor counts for the AD102, AD103, and AD104 GPUs were fully disclosed by NVIDIA. In the next weeks, all three will premiere. The specifics on the flagship AD102 GPU, which is designed for the RTX 4090 graphics card, had already been published by NVIDIA, but the information on AD104 and AD103 was still lacking.

  • AD102: 608 mm² die, 76.3B transistors
  • AD103: 378.6 mm² die, 45.9B transistors
  • AD104: 294.5 mm² die, 35.8B transistors

This means that each of the three xtor devices has a density greater than 121M per square mm (it is actually identical for AD103 and AD104). Additionally, Ampere GA102 GPU flagship has 7.5 billion fewer transistors than AD104, which has 35.8 billion transistors (28.3B). In comparison, GA102 is more than two times bigger than AD104.

The number of Render Output Units (ROP) on NVIDIA Ada GPUs is substantially more than that of the predecessor, reaching 192 ROPs for AD102. While AD104 had 80 ROPs, the AD103 GPU has the same number as GA102 (112). Rasterization performance should be enhanced by increased ROP count.

In order to make way for other logical blocks, NVIDIA removed NVLink from the design, among other adjustments.

The L2 cache, however, has grown greatly at the same period. The specific sizes for each SKU, AD102 96MB, AD103 64MB, and AD104 48MB, have now been confirmed by NVIDIA. Both RTX 4080 versions are confirmed to have fully unlocked L2 cache on their respective GPUs; the 4080 16GB has 64MB, while the 4080 12GB has 48MB.

In order to make way for other logical blocks, NVIDIA removed NVLink from the design, among other adjustments. The L2 cache, however, has grown greatly at the same period. The specific sizes for each SKU, AD102 96MB, AD103 64MB, and AD104 48MB, have now been confirmed by NVIDIA. Both RTX 4080 versions are confirmed to have fully unlocked L2 cache on their respective GPUs; the 4080 16GB has 64MB, while the 4080 12GB has 48MB.

Additionally, according to HKEPC, NVIDIA clarified what TSMC 4N—not to be confused with N4—really implies. This process is a die downsize of the TSMC 5N process, although the architecture is still 5 nm. This “clarification” only has one flaw: NVIDIA gives false information on the 4nm process.

Also Read:

LEAVE A REPLY

Please enter your comment!
Please enter your name here

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

More like this

AMD & Intel Gain GPU Market Share in Korea as NVIDIA Struggles with Availability

AMD & Intel Gain GPU Market Share in Korea...

The GPU landscape is shifting in 2025, and for once, it's not NVIDIA dominating the charts. AMD...
NVIDIA GeForce RTX 5050, 5060, and 5070 GPUs Spotted: Affordable Powerhouses Coming Soon?

NVIDIA GeForce RTX 5050, 5060, and 5070 GPUs Spotted:...

The GPU market is heating up once again as NVIDIA's next-generation GeForce RTX 50 series, including the...
NVIDIA GeForce RTX 5070: A New Contender in the GPU Market, Launching Alongside AMD's RX 9070 Series

NVIDIA GeForce RTX 5070: Launching Alongside AMD’s RX 9070...

The GPU market is about to heat up. As of early February 2025, new rumors surrounding NVIDIA’s...
NVIDIA RTX 50 Series Laptop GPUs

NVIDIA RTX 50 Series Laptop GPUs Delayed Until March...

NVIDIA RTX 50 Series : NVIDIA’s highly anticipated GeForce RTX 50 series laptop GPUs are reportedly facing...
NVIDIA

NVIDIA & MediaTek Eye AI-Powered Smartphone & PC Markets

NVIDIA : NVIDIA and MediaTek are rumored to be collaborating on a new AI-powered smartphone chip,...

LATEST NEWS

P90 – Tune Blaster in Free Fire: Your Ultimate Guide to Snagging This Epic Skin

P90 - Tune Blaster in Free Fire! Attention, Free Fire warriors! Are you ready to take your P90 game to the next level? Brace yourselves,...

Kareena Kapoor Khan Monochrome Magic: Elevating Workwear to New Heights

Kareena Kapoor Khan Fashion Updates! Hey there, fashion enthusiasts! Ready to spice up your work wardrobe? Well, buckle up because Kareena Kapoor Khan just dropped...

Meta $200 Billion AI Gamble: Inside the Tech Giant’s Massive Data Center Plans

Meta’s $200 Billion AI Gamble! Hey there, tech enthusiasts and social media aficionados! Buckle up, because we’re about to dive into a story that’s hotter...

Disney Villains Take Over Fortnite: How to Snag Captain Hook, Maleficent, and Cruella De Vil Skins!

Disney Villains Take Over Fortnite! Attention, Fortnite fanatics and Disney diehards! The worlds of pixie dust and battle royale have collided in the most deliciously...

Featured