Picture of the NVIDIA AD107 Silicon Powering the GeForce RTX 4060

The NVIDIA “Ada Lovelace” graphics architecture-based “AD107” is anticipated to be the smallest client GPU. Rumor has it that the next GeForce RTX 4060 (non-Ti) will enable all of the silicon’s shaders. One of the earliest images of the “AD107” in use was obtained by MEGAsizeGPU, showing a little fiberglass substrate and a die that is obviously smaller than the “AD106” powering the RTX 4060 Ti.

3,072 CUDA cores spread across 24 SM (streaming multiprocessors) are anticipated to be present on the “AD107” silicon. Similar to the “AD106,” the GPU has a 128-bit wide memory interface. NVIDIA is anticipated to use standard 17 Gbps GDDR6 memory, translating to a memory bandwidth of 272 GB/s. The RTX 4060 is anticipated to come with memory that is 8 GB in size as standard. With “Ada,” NVIDIA rebalanced the memory sub-system, placing more of an emphasis on on-die caches. The “AD107” includes a 24 MB L2 cache. Similar to the larger “AD106,” the smaller “AD107” has a PCI-Express 4.0 x8 host interface. Power has to be its strongest suit, as even the RTX 4060 at its highest setting is only rated for 119 W of TGP.

About Mohammed Abdulrauf

لدي اهتمام وخبرة بعدة مجالات ابرزها المونتاج وكتابة المراجعات والتصوير والالعاب والرياضة
احب التقنية والكمبيوتر وتركيبه وتطويره واحاول تطوير نفسي في هذه المجالات

About author

Mohammed Abdulrauf

لدي اهتمام وخبرة بعدة مجالات ابرزها المونتاج وكتابة المراجعات والتصوير والالعاب والرياضة احب التقنية والكمبيوتر وتركيبه وتطويره واحاول تطوير نفسي في هذه المجالات