CPUsNews

Intel Rocket Lake-S Based i9 Fails to Beat the Ryzen 9 5900X in ST or MT Performance

Intel’s upcoming 11th Gen Rocket Lake-S processors will feature the Sunny Cove core architecture which the company is calling Cypress core for some reason. It will be the first architectural upgrade since 2016’s Skylake: That’s nearly 5 years since a new core was introduced in the desktop space. AMD launched Zen, Zen 2, and Zen 3 in the same interval while Intel kept adding pluses to its 14nm node. Regardless, Rocket Lake’s single-threaded performance is looking pretty solid.

As per CapFrameX, the RKL-S Core i9 just manages to edge past AMD’s Ryzen 9 5950X. However, this is when the chip is overclocked by 200MHz to 5.5GHz. The stock boost clock is 5.3GHz for single-threaded workloads and 4.8GHz for multi-threaded workloads such as content creation.

At the stock settings, the ES1 (Engineering sample v1) scores 616 points in the CB R20 ST benchmark and 5,740 in the multi-threaded benchmark. In comparison, the Comet Lake-S flagship, the Core i9-10900K nets 6,399, while the 5900X achieves 636 points. This isn’t surprising as Rocket Lake-S is going to top out at 8 cores while the former features 10 cores for the i9 part. The 12-core Ryzen 9 5900X yields 7,100 points in the multi-core test which is around 25% faster than the RKL-S flagship.

Comparing the ST scores of the 10th Gen Core i9 chips, we can see an IPC uplift of Rocket Lake is rather significant. The 10900K at 5.1GHz scores 525 in the ST test while the 11900K gets 616 points. That means an IPC gain of around 15-17%. That’s about the same as Zen 3. Intel will either level with AMD in gaming workloads or beat it by a small margin. Rocket Lake-S is slated to launch in Q1 2021. It’ll be the company’s last 14nm part, after which it’s planning to shift to a hybrid-core architecture fabbed on the 10nm SuperFin node.

Areej

Computer Engineering dropout (3 years), writer, journalist, and amateur poet. I started my first technology blog, Techquila while in college to address my hardware passion. Although largely successful, it was a classic example of too many people trying out multiple different things but getting nothing done. Left in late 2019 and been working on Hardware Times ever since.

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button