Intel today launched the Core i9-14900KS Special Edition desktop processor, which forms the company's new flagship product in the desktop segment. The Core i9-14900KS is based on the same "Raptor Lake Refresh" silicon as the i9-14900K, and offers an 8P+16E core configuration. What's new is that Intel has increased clock speeds for both the P-cores and E-cores. The P-cores now boost up to 6.20 GHz, a 200 MHz increase over those of the i9-14900K; while the E-cores boost up to 4.50 GHz, a 100 MHz increase. But these tiny speed bumps aren't what make the i9-14900KS special. It's the 320 W Extreme Power Delivery Profile, something the regular i9-14900K lacks. On select Intel Z790 and Z690 motherboards with two 8-pin EPS power connectors, the processor is able to draw large amounts of power to hold onto its high boost frequencies. Intel also made the i9-14900KS from the highest bins of the "Raptor Lake Refresh" silicon.

The Core i9-14900KS comes with a 3.20 GHz base frequency for the P-cores. Each of the 8 "Raptor Cove" P-cores comes with 2 MB of dedicated L2 cache. The 16 "Gracemont" E-cores are arranged in four E-core clusters. Each cluster shares a 4 MB L2 cache among its four cores. The 8 P-cores and 4 E-core clusters share a 36 MB L3 cache. The processor comes with a base power value of 150 W—25 W higher than the 125 W of the i9-14900K. Its maximum turbo power is still 253 W, and is engaged on platforms capable of Intel Performance Power Delivery Profile. It's only with some of the more premium motherboards that the 320 W Extreme Power Delivery Profile is engaged. The Core i9-14900KS is a Special Edition SKU, meaning that it may not be available in all the markets where the i9-14900K sells. Intel is pricing this chip at $690, a $100 premium over the i9-14900K, though interestingly, $50 cheaper than what its predecessor, the i9-13900KS, launched at.

Be sure to check out the TechPowerUp Review of the Core i9-14900KS.The launch press-deck by Intel, along with its first-party performance claims, follows.

MeLE, designers of mini PCs, and industrial PCs, unveiled the Fanless Stick PC PCG02 Pro. Measuring 146 mm x 61 mm x 20 mm (LxWxH), this thing is about the size of two smartphones duct-taped together, but weighing as much as one (since there's no battery inside). Under the hood is an Intel N100 "Alder Lake-N" processor, which features one "Gracemont" E-core cluster for a 4-core/4-thread CPU, and an Intel Xe LP-based iGPU with 24 execution units. The N100 in the MeLE PCG02 Pro is wired to 8 GB or 16 GB of LPDDR4X-4266 memory. Storage is in the form of 128 GB or 256 GB of eMMC. The device comes with Windows 11 Home single language pre-installed.

The Fanless Stick PC PCG02 Pro comes with an impressive set of connectivity for its size. Networking options include Wi-Fi 5 ac + Bluetooth 5.1, and a 1 GbE wired Ethernet. Display outputs include two HDMI 2.1, which can power a pair of 4K Ultra HD displays at 60 Hz, each. USB connectivity includes two USB 3.2 Gen 1 (5 Gbps) type-A ports; and a USB 3.2 Gen 2 (10 Gbps) type-C. There is a second type-C port used for power input through any USB PD 3.0 source that can deliver 25 W. A power adapter is included. Depending on the memory- and storage size opted for, the Fanless Stick PC PCG02 Pro is priced between USD $270 to $290. It can power a range of applications between home-entertainment (it's completely fanless), through digital signage, or even most office applications.

Intel in its 2024 International CES presentation, unveiled its two new upcoming client microarchitectures, "Arrow Lake" and "Lunar Lake." Michelle Johnston Holthaus, EVP and GM of Intel's client computing group (CCG), in her keynote address, held up a next-generation Core Ultra "Lunar Lake" chip. This is the Lunar Lake-MX package, with MOP (memory on package). You have a Foveros base tile resembling "Meteor Lake," with on-package LPDDR5x memory stacks. With "Lunar Lake," Intel is reorganizing components across its various Foveros tiles—the Compute and Graphics tiles are combined into a single tile built on an Intel foundry node that's possibly the Intel 20A (we have no confirmation); and a smaller SoC tile that has all of the components of the current "Meteor Lake" SoC tile, and is possibly built on a TSMC node, such as N3.

"Lunar Lake" will pick up the mantle from "Meteor Lake" in the U-segment and H-segment (that's ultraportables, and thin-and-light), when it comes out later this year (we predict in the second half of 2024), with Core Ultra 2-series branding. Intel also referenced "Arrow Lake," which could finally bring light to the sluggish pace of development in its desktop segment. When it comes out later this year, "Arrow Lake" will debut Socket LGA1851, "Arrow Lake" will bring the AI Boost NPU to the desktop, along with Arc Xe-LPG integrated graphics. The biggest upgrade of course will be its new Compute tile, with its "Lion Cove" P-cores, and "Skymont" E-cores, that possibly offer a large IPC uplift over the current combination of "Raptor Cove" and "Gracemont" cores on the "Raptor Lake" silicon. It's also possible that Intel will try to bring "Meteor Lake" with its 6P+8E Compute tile, Xe-LPG iGPU, and NPU, to the LGA1851 socket, as part of some mid-range processor models. 2024 will see a Intel desktop processor based on a new architecture, which is the big takeaway here.

Intel's current generation mobile processor product stack is vast, to say the least. In Q4-2023, the company launched its Core Ultra "Meteor Lake" mobile processors spanning the U-segment (7 W to 28 W), and H-segment (35 W to 45 W). Today, the company capped the upper end of the stack with the 14th Gen Core HX-series mobile processors based on "Raptor Lake Refresh," which dial up core counts to 8P+16E. And now, the company is adding more choice to the U-segment with the Core Processor Series 1, based on a lower core-count variant of the "Raptor Lake Refresh" architecture.

The Core Processor Series 1 follows the same nomenclature as the Core Ultra, where the "Ultra" denotes the latest "Meteor Lake" architecture. Processor model numbering and case badges are similar between Core Processor Series 1 and Core Ultra, except the lack of the "Ultra" brand extension. These chips are built on the monolithic "Raptor Lake Refresh" dies on the Intel 7 foundry node, and lack innovations such as the Low-power Island cores, 3D Performance Hybrid architecture, the all important AI Boost and on-silicon NPU; as well as that 2x faster Arc Xe-LPG integrated graphics, but use existing combinations of "Raptor Cove" and "Gracemont" CPU cores, along with older Xe-LP graphics with up to 96 EU; and a mostly similar I/O.

Retailers in East Asia have reportedly chosen to go live with marketing imagery depicting Intel's initial batch of 14th Gen Core SKUs—this is roughly two weeks ahead of an official October 17 "Raptor Lake Refresh" product launch. A couple of "official" slides have been forwarded to VideoCardz—the information on display provides a quick overview of basic specifications for Core i9, Core i7 and Core i5 models. The 6 GHz clock and 24 core configuration is believed to be the upcoming flagship Core i9-14900K SKU. The Core i7-14700K would fit the bill of sporting a 5.6 GHz and 20-core setup—leaked marketing info confirms that this new model has been bolstered with an additional four Gracemont E-cores. The Core i5-14600K seems to align with the 5.3 GHz and 14-core spec. A premature publication of information is not all that surprising—a couple of marketing slides breaking embargo is relatively minor when compared to actual working hardware getting demoed during summertime. ITHome reported on Raptor Lake Refresh and Meteor Lake appearing on the showroom floor at July's Bilibili World.

A new entry has appeared on the Geekbench Browser database—very early this morning—apparently pointing to the lowest model sitting within the ranks of Intel's Alder Lake-N CPU series. According to leaked Geekbench 6.1 test results, as spotted by Benchleaks, the evaluated system was running on an "Intel N50" processor. This truly entry-point 6 W TDP SoC seems to feature two Gracemont efficiency cores with a 997 MHz base clock and a boost capability of 3.4 GHz. You are not getting any hyper-threading here. The integrated GPU is reported to sport 16 Execution Units—the more powerful N200, Core i3-N305 and Core i3-N300 Alder Lake-N siblings have double that count (32 EUs).

The benchmarked N50 system achieved overall scores of 1054 (single core) and 1388 (multi-threaded). Geekbench Browser states that the tested build had 8 GB of RAM (single channel), with the OS being Microsoft Windows 10 IoT Enterprise LTSC (64-bit) on a balanced power plan. VideoCardz was somewhat impressed with the plucky 6 W TDP chip offering comparable performance to decade old CPUs, albeit with a much lower power draw. The N50's single core result positions it closer to Intel fourth Gen Core Haswell mobile processors or first generation Ryzen CPUs, but the multi-core score is a bit of letdown—perhaps comparable to Intel Core 2 and AMD FX series models.

The upcoming Core i7-14700K "Raptor Lake Refresh" processor has a core configuration of 8P+12E. That's 8 "Raptor Cove" performance cores, and 12 "Gracemont" efficiency cores spread across 3 E-core clusters. Compared to the i7-13700K, which has been carved out of the "Raptor Lake-S" silicon by disabling 2 out of the 4 available E-core clusters and reducing the L3 cache size to 30 MB from the 36 MB present; the i7-14700K gets an additional E-core cluster, and increases the shared L3 cache size to 33 MB, besides dialing up the clock speeds on both the P-cores and E-cores in comparison to the i7-13700K.

The processor likely has a P-core base frequency of 3.70 GHz, with a 5.50 GHz P-core maximum boost. In comparison, the i7-13700K tops out at 5.40 GHz P-core boost. An alleged i7-14700K engineering sample in the wild has been put through Cinebench R23, where it scores 2192 points in the single-threaded test, and 36296 points in the multi-threaded test. The processor also scored 14988.5 points in the CPU-Z Bench multi-threaded test. Intel is expected to release its 14th Gen Core "Raptor Lake Refresh" desktop processors some time in October 2023.

A SemiAnalysis report sheds light on just how much smaller the "Zen 4c" CPU core is compared to the regular "Zen 4." AMD's upcoming high core-count enterprise processor for cloud data-center deployments, the EPYC "Bergamo," is based on the new "Zen 4c" microarchitecture. Although with the same ISA as "Zen 4," the "Zen 4c" is essentially a low-power, lite version of the core, with significantly higher performance/Watt. The core is physically smaller than a regular "Zen 4" core, which allows AMD to create CCDs (CPU core dies) with 16 cores, compared to the current "Zen 4" CCD with 8.

The 16-core "Zen 4c" CCD is built on the same 5 nm EUV foundry node as the 8-core "Zen 4" CCD, and internally features two CCX (CPU core complex), each with 8 "Zen 4c" cores. Each of the two CCX shares a 16 MB L3 cache among the cores. The SemiAnalysis report states that the dedicated L2 cache size of the "Zen 4c" core remains at 1 MB, just like that of the regular "Zen 4." Perhaps the biggest finding is their die-size estimation, which puts the 16-core "Zen 4c" CCD just 9.6% larger in die-area, than the 8-core "Zen 4" CCD. That's 72.7 mm² per CCD, compared to 66.3 mm² of the regular 8-core "Zen 4" CCD.

Giga Computing, a subsidiary of GIGABYTE and an industry leader in high-performance servers, workstations, and mini-PCs, today announced a newly designed ultra-compact, low power mini-PC for the GIGABYTE BRIX lineup - BNIP and BNi3 series, which adopt either the ultra-low power Intel Processor or Intel Core i3 N-Series Processors. With Intel's 7 nm manufacturing and E-core (Efficient Core) architecture, users can benefit not only in performance, but also in computing efficiency. Also, although the BRIX series is compact, these new models are even more so, and maintain respectable performance while fitting perfectly in any IoT employment, whether for office use, education use, home use, digital signage, medical care, or KIOSK.

The newly designed ultra-compact BRIX platform boasts a sleek and contemporary grid design on its exterior, creating a fresh and minimalist impression. The intricately patterned surface produces a pleasing interplay of light, shadow, reflection, and refraction from different angles, for an overall pleasant aesthetic. This design not only enhances the product's visual appeal, but also symbolizes the numerous possibilities of integrating and merging virtual and real elements within this compact mini-PC. And it only comes in a ultra small form factor that is about 1-inch tall and 0.5 L in size.

Intel today expanded its 13th Gen Core "Raptor Lake" client processor family into the commercial segment. Devices in this segment are bought in bulk by large businesses and enterprises; the processors powering them have certain in-built device security and remote management features that enforce company policy regardless of where the employees take the devices. Intel has been pioneering hardware-level remote-management and device security features for close to two decades now, with its vPro feature-set (originally launched as Centrino vPro in the 2000s. The new 13th Gen Core vPro lineup covers popular processor models across various device form-factor classes, spanning ultraportables (15 W or below), thin-and-light commercial notebooks (28 W to 35 W segment); mainstream commercial notebooks (45 W to 55 W); and commercial desktops (65 W).

The 13th Gen vPro series have a lot in common with the 12th Gen Core vPro series, At a hardware-level, these are processors based on the same "Raptor Lake" silicon as the mainstream-client 13th Gen Core processors, but with the enhanced vPro Management Engine that interacts with a compatible operating system to offer endpoint remote manageability for administrators. The desktop versions of these chips have identical SKU differentiation to 65 W 13th Gen Core processors already launched, but with either vPro or vPro Essentials feature-sets. Something similar applies to the 15 W U-segment, 28 W P-segment, and 35-45 W H-segment Core vPro processors. Along with these processors, Intel is debuting Q700-series chipsets specific to the form-factor. For U- and P-segment mobile processor SKUs, the core-logic is part of the processor package; while the H-segment and S-segment (desktop) processors come with discrete chipsets. With the 13th Gen, Intel is also standardizing a new WiFi 6E (Gig+) WLAN chipset with vPro features.

Earlier this week, we learned about the existence of the Intel Core i5-13490F, a China-exclusive processor SKU that's designed to strike "just the right" price-performance against AMD's Ryzen 5 7600 series, and possibly even the Ryzen 7 7700. It turns out that the i5-13490F isn't the only such SKU, there's also the Core i7-13790F. Positioned between the i7-13700 and i9-13900, the i7-13790F is a unique piece of silicon. It has the same 8P+8E core-configuration as the i7-13700 and i7-13700K, and even the same 5.20 GHz maximum P-core turbo frequency as the i7-13700, but comes with a little extra shared L3 cache of 33 MB.

Each of the 8 "Raptor Cove" P-cores on the i7-13790F has 2 MB of dedicated L2 cache, and each of the two available E-core clusters has 4 MB of L2 cache that's shared among the four "Gracemont" E-cores in the cluster. The L3 cache that's shared among the 8 P-cores and 2 E-core clusters, has been bumped up to 33 MB from 30 MB on the other 13th Gen Core i7 desktop SKUs. This is still short of the 36 MB physically present on the "Raptor Lake-S" silicon. There are a couple of gotchas, though. Firstly, this is still a "locked" (non-K) SKU having minimal overclocking capabilities, with a 65 W base power and possibly the same 219 W maximum turbo power as the i7-13700; and secondly, as an "F" SKU, it lacks integrated graphics. The i5-13490F and i7-13790F appear to be targeting the SI and retail channels, and come in an exclusive black paperboard box. The i7-13790F is priced at RMB ¥2,999 including taxes, or about $440.

Intel today debuted its 2023 N-series entry-level mobile processors targeting a range of low-cost notebook applications, such as educational notebooks for bulk purchase and distribution by public schools. These processors are built on the same Intel 7 (10 nm Enhanced SuperFin) node as the 13th Gen Core processors, but come with just "Gracemont" E-cores, and no P-cores.

The silicon physically features two "Gracemont" E-core clusters amounting to 8 E-cores, 6 MB of shared L3 cache, and an iGPU based on the Xe-LP graphics architecture, with 32 EUs (execution units). The silicon also features a GNI 3.0 for basic AI acceleration using a truncated version of the DLBoost instruction set, and an IPU (as in image processing unit), which can improve web-camera experience (on the fly background noise suppression). The iGPU also offers hardware-accelerated AV1 decoding. On the platform-side, the processor features a single-channel DDR5 memory interface that's backwards-compatible with DDR4, and also supports LPDDR5. Storage interfaces include eMMC, UFS 2.1, and NVMe SSD. Wireless networking options available with the platform include fast WiFi 6E and Bluetooth 5.2.

Intel today launched its 13th Gen Core "Raptor Lake" desktop processors, and companion 700-series motherboard chipset. These processors are built in the same LGA1700 package as the previous generation "Alder Lake," and are backwards-compatible with 600-series chipset motherboards through a BIOS update. Likewise, 700-series chipset motherboards support older "Alder Lake" processors. With the new 13th Gen Core, Intel is broadly promising an up to 15% uplift in single-threaded performance, which has a bigger bearing on gaming performance; and an up to 41% multi-threaded performance uplift; over the previous-generation, when comparing the top Core i9-13900K with its predecessor, the i9-12900K. Intel also claims to have outclassed the AMD Ryzen 9 5950X in multi-threaded performance, and the Ryzen 7 5800X3D in gaming performance.

Intel's performance claims are backed by some impressive hardware changes despite the company sticking with the same Intel 7 (10 nm Enhanced SuperFin) foundry node as "Alder Lake." To begin with, the single-thread performance uplift comes from the new "Raptor Cove" performance-core, which promises an IPC uplift over the previous-generation "Golden Cove," comes with more dedicated L2 cache of 2 MB per core (compared to 1.25 MB per core in the previous-generation); and significantly higher clock-speeds, going all the way up to 5.80 GHz. "Raptor Lake" has up to 8 P-cores, but the company has put in a lot of work in improving the contribution of E-cores to the processor's overall multi-threaded performance uplift. This is achieved by doubling the E-core count to 16. These are the same "Gracemont" E-cores as previous-generation, but Intel has doubled the L2 cache that's shared in a 4-core Gracemont cluster, from 2 MB per cluster to 4 MB. There are upgrades to even the hardware prefetchers of these cores.

OneRaichu, who has access to engineering samples of both the AMD "Raphael" Ryzen 7000-series, and Intel 13th Gen Core "Raptor Lake," performed IPC comparisons between the two, by disabling E-cores on the "Raptor Lake," fixing the clock speeds of both chips to 3.60 GHz, and testing them across a variety of DDR5 memory configurations. The IPC testing was done with SPEC, a mostly enterprise-relevant benchmark, but one that could prove useful in tracing where the moderately-clocked enterprise processors such as EPYC "Genoa" and Xeon Scalable "Sapphire Rapids" land in the performance charts. OneRaichu also threw in scores obtained from a 12th Gen Core "Alder Lake" processor for this reason, as its "Golden Cove" P-core powers "Sapphire Rapids" (albeit with more L2 cache).

With DDR5-4800 memory, and testing on SPECCPU2017 Rate 1, at 3.60 GHz, the AMD "Zen 4" core ends up with the highest scores in SPECint, topping even the "Raptor Cove" P-core. It scores 6.66, compared to 6.63 total of the "Raptor Cove," and 6.52 of the "Golden Cove." In the SPECfp tests, however, the "Zen 4" core falls beind "Raptor Cove." Here, scores a 9.99 total compared to 9.91 of the "Golden Cove," and 10.21 of the "Raptor Cove." Things get interesting at DDR5-6000, a frequency AMD considers its "sweetspot," The 13th Gen "Raptor Cove" P-core tops SPECint at 6.81, compared to 6.77 of the "Zen 4," and 6.71 of "Golden Cove." SPECfp sees the "Zen 4" fall behind even the "Golden Cove" at 10.04, compared to 10.20 of the "Golden Cove," and 10.46 of "Raptor Cove."

Remember how 12th Gen Core i5 non-K was vastly different in performance from the Core i5 K/KF on account of being 6P+0E processors in comparison to more L3 cache and a 6P+4E core-count of the i5-12600K/KF? Intel is doubling down on creating architectural confusion in the mid-range, according to a 3DCenter.org article citing a leaked slide from Intel's 13th Gen Core launch press-deck.

We had earlier thought that the 13th Gen non-K Core i5 will have a 6P+4E core-config, but still be based on "Raptor Lake" (i.e. "Raptor Cove" P-cores + "Gracemont" E-cores), in comparison to the i5-13600K/KF, which are confirmed "Raptor Lake" chips with 6P+8E configuration; but it turns out that Intel is basing the non-K 13th Gen Core i5 on the older "Alder Lake" microarchitecture. These chips will be 6P+4E (that's six "Golden Cove" P-cores + four "Gracemont" E-cores), which make them essentially identical to the i5-12600K, but without the unlocked multiplier, and a lower 65 W processor base power.

According to an investigative report by "Chips and Cheese," the larger L2 caches in Intel's 13th Gen Core "Raptor Lake-S" doesn't come with a proportionate increase in cache latency, and Intel seems to have contained the latency increase well. "Raptor Lake-S" significantly increases L2 cache sizes over the previous generation. Each of its 8 "Raptor Cove" P-cores has 2 MB of dedicated L2 cache, compared to the 1.25 MB with the "Golden Cove" P-cores powering the current-gen "Alder Lake-S," which amounts to a 60 percent increase in size. The "Gracemont" E-core clusters (group of four E-cores), sees a doubling in the size of the L2 cache that's shared among the four cores in the cluster, from 2 MB in "Alder Lake," to 4 MB. The last-level L3 cache shared among all P-cores and E-core clusters, sees a less remarkable increase in size, from 30 MB to 36 MB.

Larger caches have a direct impact on performance, as more data is available close to the CPU cores, sparing them a lengthy fetch/store operation to the main memory (RAM). However, making caches larger doesn't just cost die-area, transistor-count, and power/heat, but also latency, even though L2 cache is an order of magnitude faster than the L3 cache, which in turn is significantly faster than DRAM. Chips and Cheese tracked and tabulated the L2 cache latencies of past Intel client microarchitectures, and found a generational increase in latencies with increasing L2 cache sizes, leading up to "Alder Lake." This increase has somehow tapered with "Raptor Lake."

Intel's 13th Gen Core i5 "Raptor Lake" desktop processor lineup could see the top Core i5-13600K and i5-13600KF feature a 6P+8E core-configuration (that's six performance cores and eight efficiency cores). Each of the six P-cores has HyperThreading enabled, making this a 14-core/20-thread processor. Each of the six "Raptor Cove" P-cores has 2 MB of dedicated L2 cache. The eight "Gracemont" E-cores are spread across two E-core clusters with four cores, each. Each cluster shares 4 MB of L2 cache among the four E-cores (increased from 2 MB per cluster on "Alder Lake"). The P-cores and E-cores share 24 MB of L3 cache, increased from 20 MB on the i5-12600K.

A qualification sample (QS) of the Core i5-13600K made its way to social media, where it was put through a bunch of synthetic tests. In CPU-Z Bench, the i5-13600K QS scores 830 points in single-thread, compared to 648 points of the Ryzen 9 5950X "Zen 3," and trails it in the multi-threaded tests, with 10031.8 points, compared to 11906 points for the Ryzen. The QS comes with a Processor Base Power (PBP) value of 125 W, same as that of the i5-12600K. "Raptor Lake" is backwards compatible with Intel 600-series chipset motherboards, although it launches alongside the Intel 700-series chipset. It shares the LGA1700 socket with 12th Gen "Alder Lake," and is built on the same Intel 7 node (10 nm Enhanced SuperFin) as its predecessor.

Intel's 13th Gen Core "Raptor Lake" is shaping up to be another leadership desktop processor lineup, with an engineering sample clocking significant increases in gaming minimum framerates over the preceding 12th Gen Core i9-12900K "Alder Lake." Extreme Player, a tech-blogger on Chinese video streaming site Bilibili, posted a comprehensive gaming performance review of an i9-13900K engineering sample covering eight games across three resolutions, comparing it with a retail i9-12900KF. The games include CS:GO, Final Fantasy IX: Endwalker, PUBG, Forza Horizon 5, Far Cry 6, Red Dead Redemption 2, Horizon Zero Dawn, and the synthetic benchmark 3DMark. Both processors were tested with a GeForce RTX 3090 Ti graphics card, 32 GB of DDR5-6400 memory, and a 1.5 kW power supply.

The i9-13900K ES is shown posting performance leads ranging wildly between 1% to 2% in the graphics tests of 3DMark, but an incredible 36% to 38% gain in the CPU-intensive tests of the suite. This is explained not just by increased per-core performance of both the P-cores and E-cores, but also the addition of 8 more E-cores. Although the same "Gracemont" E-cores are used in "Raptor Lake," the L2 cache size per E-core cluster has been doubled in size. Horizon Zero Dawn sees -0.7% to 10.98% increase in frame rates. There are some anomalous 70% frame-rate increases in RDR2, discounting which, we still see a 2-9% increase. FC6 posts modest 2.4% increases. Forza Horizon 5, PUBG, Monster Hunter Rise, and FF IX, each report significant increases in minimum framerates, well above 20%.

Intel's next-generation Core "Meteor Lake" processors will debut the new Xe-LPG graphics architecture for its iGPU. A successor to the Xe-LP architecture powering iGPUs since 11th Gen Core "Tiger Lake," the Xe-LPG graphics architecture is tailored for small-scale GPU designs such as iGPUs. It sheds much of the bulk that the Xe-HPG has, which is optimized for discrete GPU designs. A leaked block diagram of "Meteor Lake" describes Xe-LPG as featuring a new "extended gaming mode," new Adaptix power sharing, which is probably a power-management optimization that prioritizes power share to the iGPU; and even more media encode acceleration capabilities.

The Core "Meteor Lake" compute tile will also feature the latest Gaussian Network Accelerator, GNA 3.5, which speeds up AI deep-learning neural net building and training. The chip features a purpose-build VPU (visual processing unit), similar to the ones in mobile SoCs, which improves the device's ability to recognize faces, or even augmented-reality applications. Lastly, with "Meteor Lake," Intel is debuting the new "Crestmont" E-core clusters that introduce an IPC improvement over the "Gracemont" E-cores powering "Alder Lake" and "Raptor Lake."

An engineering sample of a 13th Intel Core "Raptor Lake" Core i9 processor hit the web, courtesy of wxnod on Twitter, which confirms its 8P+16E core-configuration in a CPU-Z screenshot. Based on the same LGA1700 package as "Alder Lake," and backwards compatible with Intel 600-series chipset motherboards, besides new 700-series ones, "Raptor Lake" combines eight "Raptor Cove" performance cores (P-cores), with sixteen "Gracemont" efficiency cores (E-cores).

"Raptor Cove" features a generational IPC increase over the "Golden Cove" P-cores powering "Alder Lake," while the "Gracemont" E-cores, although identical to those on "Alder Lake," are expected to benefit from the doubling in L2 cache per cluster, from 2 MB to 4 MB. The ISA as detected by CPU-Z appears to be identical to that of "Alder Lake." The processor is a monolithic silicon chip built on the Intel 7 (10 nm Enhanced SuperFin) silicon fabrication process.

Back in January, we heard the first reports of Intel significantly increasing the on-die cache sizes on its 13th Gen Core "Raptor Lake-S" desktop processor, with the sum total of L2 and L3 caches on the silicon being 68 MB. A CPU-Z screenshot from the same source as the January story, confirmed the cache sizes. The "Raptor Lake-S" die in its full configuration features eight "Raptor Cove" performance cores (P-cores), and sixteen "Gracemont" efficiency cores (E-cores), making it a 24-core/32-thread chip.

Each "Raptor Cove" P-core features 2 MB of dedicated L2 cache even in its client variant, as previously reported, which is an increase from the 1.25 MB L2 cache of the "Golden Cove" P-cores on "Alder Lake-S." The sixteen "Gracemont" E-cores are spread across four E-core clusters, just like the eight E-cores of "Alder Lake-S" are spread across two such clusters. The four cores in each cluster share an L2 cache. Intel has doubled the size of this L2 cache from 2 MB on "Alder Lake" chips, up to 4 MB. The shared L3 cache on the silicon has increased in size to 36 MB. Eight P-cores with 2 MB each, and four E-core clusters with 4 MB, each, total 32 MB of L2 cache. Add this to 36 MB of L3 cache, and you get 68 MB of L2+L3 cache. Intel is expected to debut "Raptor Lake" in the second half of 2022 alongside the 700-series chipset, and backwards compatibility with 600-series chipset. It could go down as Intel's last client processor built on a monolithic silicon.

Le Comptoir du Hardware scored a die-shot of a 2P+8E core variant of the "Meteor Lake" compute tile, and Locuza annotated it. "Meteor Lake" will be Intel's first processor to implement the company's IDM 2.0 strategy to the fullest. The processor is a multi-chip module of various tiles (chiplets), each with a certain function, sitting on die made on a silicon fabrication node most suitable to that function. Under this strategy, for example, if Intel's chip-designers calculate that the iGPU will be the most power-hungry component on the processor, followed by the CPU cores, the graphics tile will be built on a more advanced process than the compute tile. Intel's "Meteor Lake" and "Arrow Lake" processors will implement chiplets built on the Intel 4, TSMC N3, and Intel 20A fabrication nodes, each with unique power and transistor-density characteristics. Learn more about the "Meteor Lake" MCM in our older article.

The 2P+8E (2 performance cores + 8 efficiency cores) compute tile is one among many variants of compute tiles Intel will develop for the various SKUs making up the next-generation Core mobile processor series. The die is annotated with the two large "Redwood Cove" P-cores and their cache slices taking up about 35% of the die area; and the two "Crestmount" E-core clusters (each with 4 E-cores), and their cache slices, taking up the rest. The two P-cores and two E-core clusters are interconnected by a Ring Bus, and share an L3 cache. The size of each L3 cache slice is either 2.5 MB or 3 MB. At 2.5 MB, the total L3 cache will be 10 MB, and at 3 MB, it will be 12 MB. As with all past generations, the L3 cache is fully accessible by all CPU cores in the compute tile.

Enthused with its IPC leadership, Intel is possibly planning a return to the high-end desktop (HEDT) market segment, with the "Alder Lake-X" line of processors, according to a Tom's Hardware report citing a curious-looking addition to an AIDA64 beta change-log. The exact nature of "Alder Lake-X" (ADL-X) still remains a mystery—one theory holds that ADL-X could be a consumer variant of the "Sapphire Rapids" microarchitecture, much like how the 10th Gen Core "Cascade Lake-X" was to "Cascade Lake," a server processor microarchitecture. Given that Intel is calling it "Alder Lake-X" and not "Sapphire Rapids-X," it could even be a whole new client-specific silicon. What's the difference between the two? It's all in the cores.

While both "Alder Lake" and "Sapphire Rapids" come with "Golden Cove" performance cores (P-cores), they use variants of it. Alder Lake has the client-specific variant with 1.25 MB L2 cache, a lighter client-relevant ISA, and other optimizations that enable it to run at higher clock speeds. Sapphire Rapids, on the other hand, will use a server-specific variant of "Golden Cove" that's optimized for the Mesh interconnect, has 2 MB of L2 cache, a server/HPC-relevant ISA, and a propensity to run at lower clock speeds, to support the silicon's overall TDP and high CPU core-count.

Intel is designing a "Halo" SKU of a future generation of mobile processors with a goal to match Apple's in-house silicon of the time. Slated for tape-out some time in 2023, with mass-production expected in 2024, the 15th Generation Core "Arrow Lake-P Halo" processor is being designed specifically to compete with Apple's "premium 14-inch laptop" (presumably the MacBook Pro) that the company could have around 2024, based on an in-house Apple silicon. This is to essentially tell its notebook partners that they will have an SoC capable of making their devices in the class truly competitive. Apple relies on a highly scaled out Arm-based SoC based on in-house IP blocks, with a software that's closely optimized for it. Intel's effort appears to chase down its performance and efficiency.

The Core "Arrow Lake" microarchitecture succeeds the 14th Gen "Meteor Lake." It is a multi-chip module (MCM) of three distinct dies built on different fabrication nodes, in line with the company's IDM 2.0 strategy. These nodes are Intel 4 (comparable to TSMC N7 or N6), Intel 20A (comparable to TSMC N5), and an "external" 3 nm-class node that's just the TSMC N3. The compute tile, or the die which houses the CPU cores, combines a hybrid CPU setup of 6 P-cores, and 8 E-cores. The performance cores are likely successors of the "Redwood Cove" P-cores powering the "Meteor Lake" compute tiles. Intel appears to be using one kind of E-cores across two generations (eg: Gracemont across Alder Lake and Raptor Lake). If this is any indication, Arrow Lake could continue to use "Crestmont" E-cores. Things get interesting with the Graphics tile.

Intel is allegedly advancing the launch of its 13th Gen Core "Raptor Lake-S" desktop processors to some time in Q3-2022, according to a report by Moore's Law is Dead. It was earlier believed to be a Q4 launch, much like "Alder Lake" was, in 2021. The report predicts the debut of "Raptor Lake" in the desktop segment in Q3-2022 (between July and September), with certain mobile SKUs expected toward the end of the year, in Q4. The Core "Raptor Lake-S" processor is built in the existing Socket LGA1700 package, and is being designed for compatibility with existing Intel 600-series chipset motherboards with a firmware update.

The "Raptor Lake-S" silicon is built on the existing Intel 7 (10 nm Enhanced SuperFin) node, and physically features eight "Raptor Cove" P-cores, along with sixteen "Gracemont" E-cores that are spread across four clusters. The chip has additional cache memory, too. Moore's Law is Dead predicts that the "Raptor Cove" P-core could introduce an IPC uplift in the region of 8 to 15 percent over the "Golden Cove" core, while the chip's overall multi-threaded performance could be anywhere between 30 to 40 percent over "Alder Lake-S," on account of not just increased IPC of the P-cores, but also eight additional E-cores.