Intel Core Ultra 9 285K: #1 in multi-threaded tasks even without HT

A new CPU has taken the throne of absolute performance – the Intel Core Ultra 9 285K. Yes, it’s a tight race, it’s with higher power consumption and lower efficiency, but that’s a necessary tax to pay for that “absolute peak”. Sure, someone will gladly sacrifice that leadership in favor of better operating characteristics, but either way, the 24-core CPU is the one that sets the pace in compute tasks on a mainstream platform. In this case, it’s Intel LGA 1851.

Intel Core Ultra 9 285K in detail

Currently, the most powerful processor for the new Intel LGA 1851 platform is the 24-core Core Ultra 9 285K model. The number of cores hasn’t changed between generations (compared to the Core i9-14900K), not even the ratio of performance/P (8) to efficient/E (16) cores, but what has changed is the number of threads. Those are now “only” 24 (instead of 32 when Intel Core i9 processors used HT on E cores) due to the removal of the iconic Hyper Threading technology.

However, the efficient (E) cores of Arrow Lake, or rather Core Ultra 200S processors, are significantly more powerful, so multi-threaded performance shouldn’t suffer from this move, and as we already know from our tests, it hasn’t. The E cores of the new Core Ultra 9 processors (as well as this entire generation of Intel CPUs) are significantly more powerful than they used to be with the Ci9-14900K (and Ci9-13900K or Ci9-12900K). You can read more about their Skymont architecture in this detailed analysis. About Lion Cove, the architecture of the P cores, then here (link to article).

The new thing about the Arrow Lake generation is also in the chiplet structure of the processor itself. Previously, Intel’s processor design was monolithic, including Raptor Lake (Refresh) processors. The results of the Core Ultra 7 265K are already a testament to the success of this revamp by Intel. The higher-end Core Ultra 9 285K model has more active E cores (16 instead of 12) and utilizes the full capabilities of a 3-nanometer chip with an area of approximately 117 mm².

Also higher (than with the CU7 265K) are the maximum core clock speeds, of both the performance and the efficient cores. Under such circumstances, without power limits, the power consumption of the CU7 285K is naturally higher (than that of the CU7 265K). In case you limit the processor to the same power consumption, for example to the TDP level (125 W), the core clock speeds of the Core Ultra 7 285K will already end up at lower values. That is, in terms of all-core boost. In a single-threaded load, the processor is far from the TDP/PL1 power limits even with an SC boost at 5.7 GHz. The CPU core clock speeds are more conservative than those of the Core i9-14900K, which can be attributed to the new TSMC N3B manufacturing process node.

The Ultra 9 285K’s natural rival is the Ryzen 9 9950X, to which we’ll also be comparing the Intel processor.

Please note: The article continues in the following chapters.

Wordpress Table Plugin

* We do not have certainty on this parameter. AMD does not specify the maximum resolution and maximum refresh rate in publicly available materials. However, it is possible that it will be the same as for Ryzen 7000s, i.e. 3840 × 2160 px (60 Hz).