NVIDIA GeForce RTX 4070 Variant Could be Refreshed With AD103 GPU

[erek]

Interesting, 80 ray tracing acceleration elements included

“According to TPU's GPU database the NVIDIA AD103: "uses the Ada Lovelace architecture and is made using a 5 nm production process at TSMC. With a die size of 379 mm² and a transistor count of 45,900 million it is a large chip. AD103 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-ray tracing, variable-rate shading and more, in upcoming video games. It features 10240 shading units, 320 texture mapping units and 112 ROPs. Also included are 320 tensor cores which help improve the speed of machine learning applications. The GPU also contains 80 ray tracing acceleration cores."

Source: https://www.techpowerup.com/308390/...nt-could-be-refreshed-with-ad103-gpu#comments

 

[Lakados]

Interesting, 80 ray tracing acceleration elements included

“According to TPU's GPU database the NVIDIA AD103: "uses the Ada Lovelace architecture and is made using a 5 nm production process at TSMC. With a die size of 379 mm² and a transistor count of 45,900 million it is a large chip. AD103 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-ray tracing, variable-rate shading and more, in upcoming video games. It features 10240 shading units, 320 texture mapping units and 112 ROPs. Also included are 320 tensor cores which help improve the speed of machine learning applications. The GPU also contains 80 ray tracing acceleration cores."

View attachment 569318
Source: https://www.techpowerup.com/308390/...nt-could-be-refreshed-with-ad103-gpu#comments

Probably why Nvidia was so butthurt over MSI using the "SUPER" name in the lineup.

 

[ZeroBarrier]

Probably why Nvidia was so butthurt over MSI using the "SUPER" name in the lineup.

Well, I would imagine that Nvidia actually has a legal claim to Nvidia GPUs with a Super moniker.

 

[Lakados]

Well, I would imagine that Nvidia actually has a legal claim to Nvidia GPUs with a Super moniker.

Yeah but they moved really fast on that, abnormally so even for Nvidia.

 

[GoldenTiger]

Yeah but they moved really fast on that, abnormally so even for Nvidia.

Maybe they got pointers from Nintendo?

 

[Gideon]

Probably why Nvidia was so butthurt over MSI using the "SUPER" name in the lineup.

Well most just refer to them as a bad deal right now, guess they wanted to wait until they could be a Super bad deal instead. Just not seeing how releasing more over priced cards with slightly better performance is going to change anything in the current market.

 

[ProfessorUtopia]

This strikes me as unlikely. A 4070 refresh with higher core counts than the 4080? Sure, they could kneecap it as a 12GB card and hold the clocks back. But unless it's basically the same price as a 4070 Ti, reviewers will tear them a new one, given the present state of things.

 

[GoodBoy]

The article was reading that core count from the AD103 full chip specs, not what would end up in a 4070. The 4070 would be a cut down chip, likely a way to salvage chips not good enough to be 4080's.