CPUsNews

AMD Zen 3+ (Warhol) May Get an IPC Increase of Up to 10%; Zen 4 (Raphael) to be Up to 50% Faster than Zen 2

AMD is expected to launch its next processor lineup at the end of this year in the form of Warhol, under the Ryzen 6000 or 5000 XT series. Although we’re expected to see the same Zen 3 core architecture, the chips are supposedly going to be fabbed on an enhanced version of the 7nm node, codenamed N6 (6nm). Technically, if the core is the same, the IPC should stay more or less the same as well, with some improvements in the boosting behavior due to the adoption of a more mature process.

However, as per details shared by RGT, the Zen 3+ based Warhol CPUs will feature a double-digit IPC gain. In most applications, the processors will reportedly show an IPC uplift of 4-7%, with certain cases seeing a gain of as much as 11-13%. Frankly speaking, I’m a bit skeptical about this as we’ve heard nothing new about the Zen 3+ design, meaning that it’s likely the same as Zen 3. Furthermore, the 6nm node that is supposed to be utilized for the Warhol lineup is basically going to be the same 7nm process with some improved libraries and possibly EUV lithography.

As such, while the boost clocks should be a bit higher than Vermeer, the IPC should remain more or less unchanged. Furthermore, it’s unclear whether AMD will name the new lineup as Ryzen 6000 or 5000 XT. Moving on, there are the Ryzen 7000 chips based on the 5nm EUV process and the Zen 4 core. These CPUs are likely going to feature another 15-20% IPC gain over Zen 3, bringing up the total figure close to 40% when compared to Zen 2.

While Zen 3+ (Warhol) will use the same AM4 socket, Zen 4 (Raphael) will be based on the newer AM5 socket. This means that the new memory (DDR5) and socket improvements will also factor into the IPC in most cases, and we’re likely going to see the largest generational gain of any Zen-based core.

Source

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.
Back to top button