Intel lifts the lid on 13th Gen Core: Official specs reveal $589 for 13900K champ


Intel is moving forward with the release of 13th Gen Core processors for the mainstream desktop processor market. Officially announced today at an Innovation event in San Jose and available in stores October 20, let’s take a look at the massive firepower Intel is bringing in response to the AMD 7000 series processors released yesterday.

Intel is in no doubt about the performance reach of the 13th Gen Core, codenamed Raptor Lake. Although rumors are on the rise, this is the first time the Core team has leaked any real specs. Key legends are faster P cores, double the number of E cores for each family, and larger caches.

Slotting into the same LGA1700 socket as the incumbent 12th Gen Core Alder Lake CPUs released late last year, the 13th Gen Core is a refinement rather than a basic design. This is completely normal and expected for this next iteration.

Intel claims that adding both speed and cores increases Core i9-13900K performance in single-threaded applications by up to 15% over an already fast Core i9-12900K, while multi-threaded gains measure a very respectable 41%.

What’s new

This purported gain is achieved without switching from the enhanced 10nm “Intel 7” SuperFin process used on Alder Lake. An impressive feat considering that more cores and cache are driven at higher frequencies.

Extrapolating from the slide above, Intel is adding more electronic cores for the Core i5, Core i7, and Core i9 families. That means there’s more processing power across the leaked stack. On top of that, the hybrid memory controller is now specified for native operation at DDR5-5600 speeds, up from 4800 MT/s on 12th Gen Core.

Intel is also announcing 700 series chipsets to fully support 13th generation Core processors. The Flagship Z790, however, doesn’t differ much from today’s Z690, which will also support Raptor Lake processors via a BIOS update.

Evolving over time, PCIe 3.0 chipset lanes are reduced from 16 to 8, while 4.0 lanes are inversely reduced from 12 to 20, primarily to provide high-speed expandability for PCH-suspended devices. There’s also another 2×2 20 Gbps USB 3.2 port, but that’s your lot.

Fuel performance with more cache

Intel comments that it was able to boost Raptor Lake frequencies using an improved version of the third-generation Intel 7 process. Like the latest Ryzens, this results in a reasonable shift in the voltage/frequency curve.

The 13th-gen core can run at the same speed as the 12th-gen core with 50mV less juice – leading to the appetizing idea of ​​heaps of performance at reduced TDPs – or offer an extra 200MHz at the same voltage. Boosting speed with an extended TDP range is how the 13th Gen Core i9 scales up to 600MHz higher.

We know there are more electronic cores this generation and Intel is sticking to the proven Gracemont microarchitecture – the smaller cores are identical on Alder and Raptor Lake. That said, performance should be better as Intel doubles the L2 cache from 2MB per quad-core E-core cluster to 4MB. A small but important detail.

P cores also benefit from more L2 cache. You may recall that 12th Gen offered 1.25MB per core – 13th Gen increases that to 2MB. Simply put, an Alder Lake Gen 8P+8E (16 cores, 24 threads) chip carries a total of 14 MB L2 (8 × 1.25 + 2 × 2) while Raptor Lake increases to 24 MB (8 × 2 + 2 × 4). More cache never hurts if latency is kept at roughly the same levels, and Intel is also adding to L3 capacity for every family. Do you have the impression that Intel is racing without even thinking about blowing up transistor budgets?

Raptor Cove works differently than Golden Cove

A new addition is what Intel calls a significantly optimized prefetch algorithm. Speaking to Adi Yoaz, Chief Architect of P-Cores, this is actually a feature that’s also built into Alder Lake. A firmware-upgradable power management microcontroller is integrated into Alder Lake’s Golden Cove P-cores, which operates per-core and offers precise adjustments.

For the P Raptor Cove cores, Intel extends the scope of this microcontroller to adjust other aspects of the core in addition to telemetry monitoring and frequency adjustment. Yoaz told Club386 that Intel can control the behavior of the prefetch unit and whether L2 and L3 caches are included or not by introducing new algorithms through firmware updates. Knowing that the microcontroller affects parameters every 200 microseconds, many on-the-fly optimizations take place on Raptor Cove. Plus, constant machine learning updates algorithms so application performance improves over time.

In the future, thanks to this microcontroller, “Intel will have more and more algorithms that will adapt and modify the machine to the real application running”, continues Yoaz.

Since this is a firmware-led upgrade and the microcontroller technology already exists in Alder Lake, an interesting question is whether it can be backported to 12th Gen Core for improved performance. Intel chose not to answer this pressing question during my conversation with Adi during a technical tour in Israel a few weeks ago. I think Raptor Cove’s improved prefetch and cache algorithms are primarily marketing decisions, to help differentiate it from the direct predecessor which effectively uses the same core technology. Widely applied in the server industry, Intel is no stranger to extended performance through firmware updates.

Models and comparisons

The notion of more, more, more translates to this slice of six processors. Notice how the Core i5 jumps to 44MB of combined L2 and L3 cache, the Core i7 to 54MB, and the Core i9 to 68MB. That’s quite a bit more than the 12th Gen Core, as we’ll explain in the table below.

The devil is in the details as Intel has not announced any non-K(F) chips. Rumors abound that CPUs under the Core i5-13600K(F) will use Alder Lake technology rather than Raptor Lake, with core and thread counts and frequency massaged to fit the planned Core i5 and Core i3 segments. . Time will tell us.

CPU Cores Son L2 L3 turbo Basic PDT Max TDP MSRP
Core i9-13900K 24 (8P+16E) 32 32 MB 36 MB 5.8GHz 125W 253W $589
Core i9-12900K 16 (8P+8E) 24 14 MB 30 MB 5.2GHz 125W 241W $589
Core i7-13700K 16 (8P+8E) 24 24 MB 30 MB 5.4GHz 125W 253W $409
Core i7-12700K 12 (8P+4E) 24 12 MB 25 MB 4.9GHz 125W 190W $409
Core i5-13600K 14 (6P+8E) 20 20 MB 24 MB 5.1GHz 125W 181W $319
Core i5-12600K 10 (6P+4E) 16 9.5 MB 20 MB 4.9GHz 125W 150W $289

Here are some of the best chips from the 12th and 13th generation families. Core i9 adds another eight E-cores and, as previously explained, significantly increases the cache size. Not only that, but the frequency also sees a healthy increase, so it’s inevitable that the Core i9-13900K will be faster than the Core i9-12900K in all significant scenarios. For higher clocks, Intel increases the maximum power to 253W.

The 13th Gen Core i7 looks a lot like the 12th Gen Core i9. It will likely end up being faster thanks to better caching and higher clocks, while the new Core i5 looks set to match today’s Core i7. Indeed, Intel is therefore mounting a class of processors for this new generation.

Price hike rumors are unfounded. High-end Core i9 and Core i7 chips come in at the same price of $589 and $409 as their immediate predecessors and it’s only Core i5 that sees a 10% increase: Intel offers more performance for around the same price . It’s reasonable to expect the premium 12th-gen chips to get a price drop on October 20, to keep them attractive for those who can’t justify the 13th-gen cash.

It’s clear that Intel has prioritized total performance over power and, arguably, profit. It can’t be cheap to increase the number of threads, caches and frequency, on the same manufacturing node, while keeping the prices essentially the same as the last generation, which, let’s not forget, isn’t bad at first.

Likely performance

A key question is where do these new Core processors fit into the broader mainstream PC landscape? Intel was unable to answer this question because, understandably, it only had access to rival AMD Ryzen 5000-series processors at the time of testing. Maybe Club386 can lend a hand.

Citing multi-core gains of between 16-69% for the Core i9-13900K over the Ryzen 9 5950X, we can extrapolate that the new big boss should compare to the same ballpark as the Ryzen 9 7950X. Let the high-stakes battle begin!

Choosing the games that run best on the Core architecture, Intel’s slides reveal performance that it defines as “leadership”. Generally above the cache-rich AMD Ryzen 7 5800X3D on these results, the proof will be once we have the Core i9-13900K in our labs.

Let’s leave with one last slide that Intel is rather fond of, detailing how a Core i9-13900K is able to deliver the same performance as a 241W Core i9-12900K at just 65W. Aren’t you amused?


Comments are closed.