When Chinese chip maker Rockchip launched its RK3288 processor earlier this year there was some confusion about whether it was based on ARM Cortex-A12 or Cortex-A17 designs.

It turns out it doesn’t matter: ARM says that while the two types of chips were originally designed to offer different performance levels, but due to demand for better performance and efficiency, ARM Cortex-A12 and Cortex-A17 are effectively identical.


In fact, ARM doesn’t even plan to use the Cortex-A12 name anymore and will refer to all of the designs in this category as Cortex-A17.

Cortex-A17 chips offer 60 percent more single-threader performance than a Cortex-A9 chip at the same clock speed, while also offering better efficiency.

These are still mid-range chips that aren’t in the same class as the Cortex-A57 designs based on ARMv8 architecture. But if you’ve been spending the last year wondering whether Rockchip’s latest processor was a Cortex-A12 or Cortex-A17 chip now you can move on to other pursuits.

via MiniMachines

Support Liliputing

Liliputing's primary sources of revenue are advertising and affiliate links (if you click the "Shop" button at the top of the page and buy something on Amazon, for example, we'll get a small commission).

But there are several ways you can support the site directly even if you're using an ad blocker* and hate online shopping.

Contribute to our Patreon campaign


Contribute via PayPal

* If you are using an ad blocker like uBlock Origin and seeing a pop-up message at the bottom of the screen, we have a guide that may help you disable it.

Subscribe to Liliputing via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 9,544 other subscribers

4 replies on “ARM Cortex-A12 and Cortex-A17 chips are identical”

  1. While the Cortex-A12 and Cortex-A17 are approximately equivalent, I still do not believe the Cortex-A12 as implemented in the RK3288 as as fast as the real Cortex-A17 running at the same clock speed. ARM specifically states that the latest revision of Cortex-A12 performs at the level of the Cortex-A17, and the same most likely is true for older revision of Cortex-A12 used in the RK3288 (that is, of course, until Rockchip updates their Cortex-A12 core to a recent version).

    Based on my analysis of the Geekbench results database (https://browser.primatelabs.com/), it looks to me like the real Cortex-A17 in new chips such as MediaTek’s MT6595 is about 13% faster clock-for-clock than the Cortex-A12 inside the RK3288, which would also put RK3288’s Cortex-A12 somewhat below typical Cortex-A15 performance (reference: https://mobilesemi.blogspot.com/2014/11/analysis-of-tablet-processors-by-chip.html).

    Rockchip is one of the few known customers (virtually the only one) of Cortex-A12 through its foundry Global Foundries (which made Cortex-A12 available early) and it would have been convenient for both ARM and Rockchip to settle this issue since Rockchip had already been advertising the RK3288 as having Cortex-A17 for some time.

    Overall, it does not matter very much because both the Cortex-A12 inside the RK3288 produces performance in the same ballpark as Cortex-A15 and Cortex-A17, which is a lot faster than previous cores such as Cortex-A9 or power-efficient cores such as Cortex-A7. The CPU performance is plenty for a mobile device. More critical would be power efficiency and other factors (such as the memory interface and memory bandwidth which seem to be a bit low for the RK3288 despite the use of a dual-channel (64-bit total) external memory interface, which can explain the performance drop-off from 720p to 1080p in games).

Comments are closed.