NVIDIA's Breakthrough in Computational Lithography to Accelerate Next-Gen Chip Design For ASML, TSMC
NVIDIA has announced a major development in computational lithography that will revolutionize the design and manufacture of next-generation chips. The company's new software library, cuLitho, will enable semiconductor leaders like TSMC, Synopsys, and ASML to speed up their manufacturing processes using NVIDIA Hopper architecture GPUs.
With the limits of physics rapidly approaching, this breakthrough technology is poised to change the game in the field of computational lithography. TSMC, the world's leading foundry, and electronic design automation giant, Synopsys, are already integrating the new cuLitho software library into their systems and manufacturing processes. ASML, a major equipment maker, is also working closely with NVIDIA on this project and plans to integrate GPU support into all of its computational lithography software products.
This innovation promises to significantly accelerate the design and production of next-generation chips, positioning NVIDIA at the forefront of the industry's technological advancements. The integration of this software library into existing manufacturing processes will be a game-changer, giving manufacturers a competitive edge in the semiconductor market.
The advance will enable chips with tinier transistors and wires than is now achievable, while accelerating time to market and boosting energy efficiency of the massive data centers that run 24/7 to drive the manufacturing process.
"The chip industry is the foundation of nearly every other industry in the world," said Jensen Huang, founder and CEO of NVIDIA. "With lithography at the limits of physics, NVIDIA's introduction of cuLitho and collaboration with our partners TSMC, ASML and Synopsys allows fabs to increase throughput, reduce their carbon footprint and set the foundation for 2 nm and beyond."
Running on GPUs, cuLitho delivers a performance leap of up to 40x beyond current lithography—the process of creating patterns on a silicon wafer—accelerating the massive computational workloads that currently consume tens of billions of CPU hours every year.
It enables 500 NVIDIA DGX H100 systems to achieve the work of 40,000 CPU systems, running all parts of the computational lithography process in parallel, helping reduce power needs and potential environmental impact.
In the near term, fabs using cuLitho could help produce each day 3-5x more photomasks—the templates for a chip's design—using 9x less power than current configurations. A photomask that required two weeks can now be processed overnight.
Longer term, cuLitho will enable better design rules, higher density, higher yields and AI-powered lithography.
Support From Industry Leaders
NVIDIA is working with key partners to smooth the speedy adoption of these new technologies.
"The cuLitho team has made admirable progress on speeding up computational lithography by moving expensive operations to GPU," said Dr. C.C. Wei, CEO of TSMC. "This development opens up new possibilities for TSMC to deploy lithography solutions like inverse lithography technology and deep learning more broadly in chip manufacturing, making important contributions to the continuation of semiconductor scaling."
"We are planning to integrate support for GPUs into all of our computational lithography software products," said Peter Wennink, CEO of ASML. "Our collaboration with NVIDIA on GPUs and cuLitho should result in tremendous benefit to computational lithography, and therefore to semiconductor scaling. This will be especially true in the era of high NA extreme ultraviolet lithography."
"Computational lithography, specifically optical proximity correction, or OPC, is pushing the boundaries of compute workloads for the most advanced chips," said Aart de Geus, chair and CEO of Synopsys. "By collaborating with our partner NVIDIA to run Synopsys OPC software on the cuLitho platform, we massively accelerated the performance from weeks to days! The team-up of our two leading companies continues to force amazing advances in the industry."
Enabling Semiconductor Scaling
The cost of the computational time needed for the largest workloads in semi manufacturing has in recent years been outpacing Moore's law, due to both the larger number of transistors in newer nodes and more stringent accuracy requirements. Future nodes require more detailed calculations, not all of which can feasibly fit into the available computational bandwidth provided by the current platforms, slowing the pace of innovation in semiconductors.
A fab process change often requires an OPC revision, creating bottlenecks. cuLitho helps remove these bottlenecks, and it makes possible novel solutions and innovative techniques like curvilinear masks, high NA EUV lithography, and sub-atomic photoresist modeling needed for new technology nodes.
European prices for NVIDIA's GeForce RTX 40 series dropped by 5%. - 12/05/2022 07:15 PM
The NVIDIA GeForce RTX 4090 is now only €1,859 (before it was €1,949), while the GeForce RTX 4080 has decreased to €1,399 (previously it was €1,469). According to the official NVIDIA store, th...
Marvel Spider-Man Remastered bundled with NVIDIA's RTX 3090 and RTX 3080 series - 09/07/2022 06:57 PM
NVIDIA revealed the Be Greater game bundle, in which the company is giving away Marvel's Spider-Man Remastered (2022) with new retail purchases of GeForce RTX 3090 Ti, RTX 3090, RTX 3080 Ti, and RTX ...
AMD, Intel, and Nvidia's CPU and GPU prices could rise another 20% in the next months. - 01/19/2022 10:47 AM
It is predicted that the costs of AMD, Intel, Nvidia, and other processors will rise by between 10 and 20 percent in the next months as the result of TSMC boosting chip prices....
Nvidia's Ada Lovelace GPUs are manufactured on TSMC's 5nm process rather than Samsung's. - 08/30/2021 03:29 PM
Kopite7Kimi, a well-known Nvidia GPU enthusiast an, has shared information on the future Ada Lovelace GPUs. The entirely new architecture would be based on TSMC's 5nm fabrication technology....
Battlefield 2042 to be Bundled with NVIDIA's RTX 30-series Desktops and Laptops - 08/25/2021 09:08 AM
NVIDIA announced a bundle that includes the highly anticipated online multiplayer shooter "Battlefield 2042" as well as a selection of pre-built desktops and laptops that are equipped with...
Senior Member
Posts: 695
Joined: 2017-02-04
What amazes me is that we landed on the moon, and came back, with the processing power of a calculator from the 60's. Now computing is taken for granted and trusted to fly your airliner, drive your car and let you waste your life away playing video games. Ain't science great?
Senior Member
Posts: 2484
Joined: 2005-05-04
Thank you science

17132
Senior Member
Posts: 14592
Joined: 2014-07-21
You can say what you want, but Nvidia is here to stay.
Senior Member
Posts: 8131
Joined: 2020-08-03
Doesn't it seem to you that whatever effort amd is making to advance mcm's involvement in the process of making gpus, nvidia is doing twice as much to still squeeze everything out of single die design ? Using their own R&D resources to make their contractor's make their products even better is not something I saw coming. Though I don't think there is a mention of amd gpus not being able to take advantage of this, so ultimately I think it'll benefit everyone in the business.
Senior Member
Posts: 665
Joined: 2017-08-16
So in a extremely short and surely wrong summary, ASML made nvidia which helps ASML make better nvidia. (and amd)