launched a new Nvidia GPU architecture today in his GTC keynote

Jensen Huang, Nvidia CEO, launched a new Nvidia GPU architecture today in his GTC keynote.

Ada Lovelace is named after the mathematician and computer programmer who was widely regarded as the greatest.

[{"selector":"#anim-9b970a32-ca38-4c68-9c4d-13d7c2e45da3","keyframes":[{"transform":"scale(1)","offset":0},{"transform":"scale(1.5)","offset":0.33},{"transform":"scale(0.95)","offset":0.66},{"transform":"scale(1)","offset":1}],"delay":0,"duration":2000,"easing":"ease-in-out","fill":"both","iterations":1}] [{"selector":"#anim-91bf16cd-59bf-45c7-9519-2f29ec501e2f [data-leaf-element=\"true\"]","keyframes":{"transform":["translate3d(-62.56318898136048%, 0, 0)","translate3d(0%, 0, 0)"]},"delay":0,"duration":2000,"easing":"cubic-bezier(.3,0,.55,1)","fill":"both"}]

Two GPUs were also announced by the company, both based on Ada Lovelace architecture

[{"selector":"#anim-1e02f1f0-4f35-47d8-8441-3ba650e41a5d","keyframes":[{"transform":"scale(1)","offset":0},{"transform":"scale(1.5)","offset":0.33},{"transform":"scale(0.95)","offset":0.66},{"transform":"scale(1)","offset":1}],"delay":0,"duration":1450,"easing":"ease-in-out","fill":"both","iterations":1}] [{"selector":"#anim-565aa352-7349-438c-b4a7-55bf149e1c40 [data-leaf-element=\"true\"]","keyframes":{"transform":["translate3d(5.796185826645434%, 0, 0)","translate3d(0%, 0, 0)"]},"delay":0,"duration":2000,"easing":"cubic-bezier(.3,0,.55,1)","fill":"both"}]

The Omniverse-focused, L40-powered second-generation OVX system is paired with the workstation-focused RTX 6000, L40, and L40.

[{"selector":"#anim-fceb0541-870b-4d54-b7ec-3f5954c98b4f","keyframes":[{"transform":"scale(1)","offset":0},{"transform":"scale(1.5)","offset":0.33},{"transform":"scale(0.95)","offset":0.66},{"transform":"scale(1)","offset":1}],"delay":0,"duration":1450,"easing":"ease-in-out","fill":"both","iterations":1}] [{"selector":"#anim-a76b3376-72c1-4f42-bc8e-fb49972d1c64 [data-leaf-element=\"true\"]","keyframes":{"transform":["translate(20.760504069175152%, 0%) scale(1.5)","translate(0%, 0%) scale(1)"]},"delay":0,"duration":2000,"easing":"cubic-bezier(.3,0,.55,1)","fill":"forwards"}]

Finally, Nvidia announced the launch of Omniverse Cloud, new software and infrastructure-as-a-service.

[{"selector":"#anim-64a31c6f-c4a6-436c-b367-6767a794f368","keyframes":[{"transform":"scale(1)","offset":0},{"transform":"scale(1.5)","offset":0.33},{"transform":"scale(0.95)","offset":0.66},{"transform":"scale(1)","offset":1}],"delay":0,"duration":1450,"easing":"ease-in-out","fill":"both","iterations":1}] [{"selector":"#anim-a23ac5bc-6c43-4dda-887a-ac1c215143ec [data-leaf-element=\"true\"]","keyframes":{"transform":["translate3d(-62.56318898136048%, 0, 0)","translate3d(0%, 0, 0)"]},"delay":0,"duration":2000,"easing":"cubic-bezier(.3,0,.55,1)","fill":"both"}]

Ada Lovelace is not part of Nvidia’s Hopper GPU architecture

[{"selector":"#anim-9cb28caa-6dfb-4eeb-85df-b0a9f5326c53","keyframes":[{"transform":"scale(1)","offset":0},{"transform":"scale(1.5)","offset":0.33},{"transform":"scale(0.95)","offset":0.66},{"transform":"scale(1)","offset":1}],"delay":0,"duration":1450,"easing":"ease-in-out","fill":"both","iterations":1}] [{"selector":"#anim-210b7270-a33a-4410-a440-4e2f3a5df0dc [data-leaf-element=\"true\"]","keyframes":{"transform":["translate3d(5.796185826645434%, 0, 0)","translate3d(0%, 0, 0)"]},"delay":0,"duration":2000,"easing":"cubic-bezier(.3,0,.55,1)","fill":"both"}]

Ada Lovelace is as good for graphics workloads as Hopper is for AI or HPC workloads.

[{"selector":"#anim-f2a9fefe-97be-4200-8d75-77c48542cba6","keyframes":[{"transform":"scale(1)","offset":0},{"transform":"scale(1.5)","offset":0.33},{"transform":"scale(0.95)","offset":0.66},{"transform":"scale(1)","offset":1}],"delay":0,"duration":1450,"easing":"ease-in-out","fill":"both","iterations":1}] [{"selector":"#anim-be3fef1a-fc42-440a-be80-18fdc67080a9 [data-leaf-element=\"true\"]","keyframes":{"transform":["translate(20.760504069175152%, 0%) scale(1.5)","translate(0%, 0%) scale(1)"]},"delay":0,"duration":2000,"easing":"cubic-bezier(.3,0,.55,1)","fill":"forwards"}]

The GPU is also the best graphics.

[{"selector":"#anim-d6b5dc19-0fa0-4873-96b1-15295e4f2208","keyframes":[{"transform":"scale(1)","offset":0},{"transform":"scale(1.5)","offset":0.33},{"transform":"scale(0.95)","offset":0.66},{"transform":"scale(1)","offset":1}],"delay":0,"duration":1450,"easing":"ease-in-out","fill":"both","iterations":1}] [{"selector":"#anim-584cab5e-81b5-42d9-85d4-d8e4c390e1f8 [data-leaf-element=\"true\"]","keyframes":{"transform":["translate3d(5.796185826645434%, 0, 0) translate(25%, 0%) scale(1.5)","translate3d(0%, 0, 0) translate(0%, 0%) scale(1)"]},"delay":0,"duration":2000,"easing":"cubic-bezier(.14,.34,.47,.9)","fill":"forwards"}]

Ada Lovelace's new GPU is equipped with 76.3 billion transistors and 18,176 next-gen CUDA Cores, 568 fourth-gen Tensor Cores, and 48GB GDDR6 memory.

[{"selector":"#anim-9cb01116-d76d-4f65-80f2-80db3461d800","keyframes":[{"transform":"scale(1)","offset":0},{"transform":"scale(1.5)","offset":0.33},{"transform":"scale(0.95)","offset":0.66},{"transform":"scale(1)","offset":1}],"delay":0,"duration":1450,"easing":"ease-in-out","fill":"both","iterations":1}] [{"selector":"#anim-ec4f8244-d422-4f3c-bf08-c206376c3fb3 [data-leaf-element=\"true\"]","keyframes":{"transform":["translate3d(-62.56318898136048%, 0, 0)","translate3d(0%, 0, 0)"]},"delay":0,"duration":2000,"easing":"cubic-bezier(.3,0,.55,1)","fill":"both"}]