In a recent interview, Intel CEO Pat Gelsinger claimed that Intel had returned for performance leadership and that the leadership of its current rival AMD was “ending”. This statement was enough to highlight the company’s emphasis on the 12th generation Alder Lake platform.
AMD’s latest Ryzen 5000 “Zen 3” processors have managed to outperform Intel’s in both single and multi-threaded performance. The red team has continued to put pressure on Intel since the birth of the Ryzen name in 2017. Intel was comfortable selling 4-core and 8-thread processors in the mainstream segment, not spending much time on innovations because it had no geared competitor. While the company’s transition from 14nm to 10nm has slowed, Intel has focused itself on IPC gains since 2016 and has struggled to improve since the 8th generation.
The blue team has reached mass production of 10nm by 2019, but market demand has skyrocketed, causing a shortage of chips. Intel has probably decided to dedicate all of its foundry capacity to produce mobile chips that can take advantage of 10nm technology, and so far has been pushing the desktop with 14nm in the classroom. New 12th Gen Core Alder Lake processors, technically 10nm Enhanced SuperFin ” Intel 7” process. Intel’s hand is now more comfortable as this process offers transistor density and power characteristics comparable to TSMC or Samsung’s 7nm class DUV technologies.
The move to Intel 7 has allowed the company to increase transistor counts and develop the latest high-performance “Golden Cove” cores. As you know, while the number of cores increased to 10 in the 10th generation processors, this number decreased to 8 in the 11th generation. AMD, on the other hand, can reach much higher core counts. In response, Intel engineers came up with an innovative solution: adding two quad-core “Gracemont” low-power core clusters instead of the 9th and 10th CPU cores. With Alder Lake, it is now possible to offer processors up to 16 cores with 8 “performance cores” and 8 “efficiency cores”.
Intel found that Golden Cove (P-core) cores provide a 28% IPC gain over the “Skylake” core (about 15-20% over AMD’s Zen 3). The reasonably fast Gracemont (E-core) cores would allow the flagship processor to compete with a 16-core Ryzen. Games don’t need 16 cores and benefit from the full weight of the performance cores’ IPC gain, while multi-threaded productivity tasks benefit from the performance of both core types.
While gaming or in different scenarios, putting the productivity cores into sleep mode by running them at low power and letting the performance cores take care of their workloads provides consumers with significant benefits in terms of high productivity. Now let’s get into the details of the inner workings of the 12th generation Alder Lake.
12. Generation Processors
Intel is releasing its 12th Gen Core desktop processors in two phases. The new announcement only features the top-of-the-line Z690 motherboard chipset along with unlocked “K” and “KF” chips. Therefore, gamers and computer enthusiasts can switch to the next generation early. The company is saving the other half of this series and a large number of non-K processors for the first quarter of 2022 at a more affordable price.
The Core i9-12900K and i9-12900KF processors are the leaders of the family, followed by the Core i7-12700K and i7-12700KF, and the Core i5-12600K and i5-12600KF are in the middle tier. The introduction of e-core Gracemont cores gives Intel a new dimension in partitioning, specifically between Core i9 and Core i7 SKUs. Core numbers are represented as P+E (performance + efficiency cores). Performance cores support HyperThreag technology, while productivity cores lack this technology.
Core i9-12900K and i9-12900KF are two processors with a total of 16 cores, with an 8+8 structure that uses all of the CPU cores physically available on silicon. The Core i7-12700K and i7-12700KF come in 8+4 core configurations with one of the two E-core clusters disabled. In addition to cores, Intel has also made changes to the cache architecture.
Core i9 processors can use the full 30 MB L3 cache physically available. Core i7 series has 25 MB and Core i5 models have 20 MB L3 cache. In addition, Core i9 and Core i7 series chips support Turbo Boost Max 3.0, while Core i5 SKUs do not use this technology.
Mavililer introduced the Thermal Velocity Boost (TVB) feature a few generations ago. This feature is not used on any Alder Lake model and this has been verified by Intel. Additionally, AVX512 command support has been removed.
The price of the new CPUs seems close to the 11th Gen “Rocket Lake” family. The i9-12900K was priced at $589 and the i9-12900KF at $564. The Core i7-12700K has a price tag of $ 409, while the i7-12700KF has a price tag of $ 384. Moving on to the lower segment, the Core i5-12600K was introduced with a list price of $289 and the Core i5-12600KF $264.
You may notice that the KF series is priced at $25 less. This represents Intel’s value to the UHD 770 integrated graphics (iGPU), which is not available on the KF models. These processors can appeal to gamers who do not need an iGPU and use external graphics cards. Since none of the six processors released today have a stock cooler, this unpaid money can be used to purchase an aftermarket cooling solution.
This especially upsets gamers a bit because these processors are completely different from the old LGA1200 or LGA115x socketollar series with a new LGA1700 socket, different mounting design and IHS structure. Therefore, it is necessary to purchase a compatible new cooler, or if the manufacturer of your current cooler has offered free upgrades to the kidollar series, you can use them. Some motherboard vendors have decided to add mounting holes for both existing LGA115x/LGA1200 coolers and new LGA1700, so it’s easier to use older coolers.
|Price||Core/Imprint||P-Core Frequency||E-Core Frequency||TDP / PBP / MTP||Memory||L3 Cache|
|Ryzen 9 5950X||$799||16P | 32||3. 4 / 4.9 GHz||–||105W||DDR4-3200||64MB (2×32)|
|Core i9-12900K / KF||$589 (K) – $564 (KF)||8P + 8E | 16 / 24||3. 2 / 5. 2GHz||2. 4 / 3.9 GHz||125W / 241W||DDR4-3200 / DDR5-4800||30MB|
|Ryzen 9 5900X||$549||12P | 24||3. 7 / 4.8 GHz||–||105W||DDR4-3200||32MB (1×32)|
|Core i9-11900K||$549||8P | 16||3. 5 / 5.3 GHz||–||125W||DDR4-3200||16MB|
|Core i7-12700K / KF||$409 (K) – $384 (KF)||8P + 4E | 12 / 20||3. 6 / 4.9 GHz||2. 7 / 3.8 GHz||125W / 190W||DDR4-3200 / DDR5-4800||25MB|
|Core i7-11700K||$409||8P | 16||3. 6 / 5.0 GHz||–||125W||DDR4-3200||16MB|
|Ryzen 7 5800X||$449||8P | 16||3. 8 / 4.7 GHz||–||105W||DDR4-3200||32MB|
|Core i5-12600K / KF||$289 (K) – $264 (KF)||6P + 4E | 20 / 16||3. 7 / 4.9 GHz||2. 8 / 3.6 GHz||125W / 150W||DDR4-3200 / DDR5-4800||16MB|
|Core i5-11600K||$272||6P | 12||3. 9 / 4.9 GHz||–||95W||DDR4-3200||12MB|
|Ryzen 5 5600X||$299||6P | 12||3. 7 / 4.6 GHz||–||65W||DDR4-3200|
Alder Lake-S Silicone
Intel does not find the modular CPU core chip approach sensible to build the high core processors used by AMD. The hybrid architecture approach adopted in this context is also an important innovation in the x86 market. Alder Lake-S silicon is a monolithic die fabricated on the Intel 7 process formerly known as 10nm Enhanced SuperFin. Silicon uses the same layout as the Intel client processor from the past, with a central last-level cache (LLC or L3 cache) and a bidirectional ringbus connecting various blocks (parts of the processor).
The most important blocks are CPU cores, and as we said, up to 16 can be used in Alder Lake-S. The much larger Golden Cove performance cores make up half of that, while the other half are clustered with Gracemont cores. Each of the eight performance cores has its own special ring-stop feature. Groups of four efficiency cores (E-cores) form a single core cluster. There are two such E-core bands in Alder Lake-S, each with a ring stopper. In other words, it is possible to disable all P-cores in the processor individually, but E-cores can only be disabled on a cluster basis. Also, at least one performance core must be active on the CPU; therefore it is not possible to run the CPU with efficiency cores only. On the contrary, it is possible to disable all efficiency cores for a processor consisting of performance cores.
The Alder Lake-S silicon has 30MB of shared L3 cache accessible by all elements in the silicon, including P core, E-core clusters, iGPU (integrated GPU), memory controllers, PCIe root complex, and other components. Each of the eight P cores has a dedicated 1.25MB L2 cache. Each E-core cluster benefits from 2MB of L2 cache shared between the four cores.
Intel UHD 770 Graphics (Xe LP)
The Alder Lake-S die houses an iGPU based on the same Xe LP graphics architecture as the previous generation “Rocket Lake-S”. The Intel UHD 770 iGPU is used in six processors currently released. In detail, it includes all 32 Xe threads available within the GPU.
Extensive testing was done on the Xe LP based iGPU on the previous generation Core i9-11900K. While this GPU is sufficient for modern PCs for tasks such as high resolution displays and high frame rate video playback, it is not enough for games, even at the lowest possible settings.
You can use hardware accelerated decoding of most modern HDR standards, as well as AV1, HEVC and other modern video formats. The iGPU’s display engine looks different from the previous generation, with support for HDMI 2.1 and DisplayPort 1.4.
Memory and PCIe
The 12th-gen Core Alder Lake architecture heralds the biggest platform I/O (I/O) update in several generations, putting Intel ahead of AMD in this area. In addition, all the developments are helping with the spread of new RAM and PCIe standards.
The most important feature of Alder Lake is that it provides DDR5 memory support without losing the current generation DDR4 support. The processor can support up to 128GB of DDR5-4800 memory with a total of four 40-bit wide channels (two per DIMM). Unlike DDR4, which uses a single 64-bit wide channel per DIMM, DDR5 includes two 40-bit channels. In this sense, the DDR5 interface in Alder Lake can be called “quad-channel”, although each of these channels has a lower bus width. The processors retain support for dual-channel DDR4-3200 (64 bits per DIMM), as we mentioned.
Wanting to facilitate the transition between DDR4 and DDR5, especially in the current environment where hardware prices are high, Intel decided to enable motherboard partners to launch motherboard series with DDR4 memory slots in addition to DDR5. Motherboards with both DDR4 and DDR5 slots are not yet available. However, the end result is that systems use only one type of memory and are unnecessary for almost all users.
Another major I/O update is on the PCI Express side. Alder Lake-S is the first platform to offer PCI-Express Gen 5 technology to consumers. The processor comes with 16 PCI-Express Gen 5 lanes, typically dedicated to the motherboard’s PCI-Express Graphics (PEG) slot.
Some motherboards use up to two x16 slots (x8/x8 when both are used), while others carry a single x16 slot. Speaking of which, besides the 16 Gen 5 lines, the processor also offers four Gen 4 lines towards a CPU-attached M.
The chipset bus is also getting a big upgrade, with Intel introducing the new Direct Media Interface (DMI) Gen 4. The Alder Lake-S platform effectively doubles the bandwidth between processor and chipset, supporting up to eight DMI Gen 4 lines. The Z690 chipset, which has been released, uses all eight DMI lines. We don’t know if other future chipsets will have narrower interfaces, but Intel will likely use the DMI bus width more sparingly in the lower-end chipset.
For example, the B560 only supports four-lane DMI Gen 3, even on eight-lane Rocket Lake-S processors. DMI Gen 4 x8 provides a significant chipset bandwidth increase with 128 Gbps per direction. 10th generation Core processors offered speeds of 32 Gbps per direction. Such an increase in chipset bus bandwidth was necessary due to the increasing popularity of modern downlink standards such as Thunderbolt 4, USB 3.2×2, 10GbE Ethernet, Wi-Fi 6E, or NVMe RAID among video editing professionals.
Performance and Efficiency Cores
Alder Lake is not Intel’s first processor to feature Hybrid. We first saw the hybrid name in the Lakefield series, a pioneering mobile processor that includes a Sunny Cove P core and four Tremont E cores. Now with Alder Lake, the blue team is taking this concept not one but two steps further. The processors feature four Golden Cove P cores, two generations ahead of Sunny Cove, and Gracemont E cores, one generation ahead of Tremont.
Intel claims that Golden Cove provides a massive 19% IPC increase over Cypress Cove cores running 11th Gen Rocket Lake and a massive 28% IPC boost over 10th Gen “Comet Lake” (Skylake cores). However, Intel engineers seem to have done a good job on the Gracemont side. Not designed for high power and occupying roughly a quarter of Golden Cove’s mold area, Gracemont manages to keep a close eye on Skylake cores in terms of IPC.
The Gracemont kernel has received major upgrades to all three core components (front-end, execution, and loadstore) to bridge the ISA gap between itself and the P-core. The cores share up to 4MB of L2 cache with four other cores communicating with 30MB of L3 cache in Alder Lake.
Much of Intel’s efforts to improve the performance of its efficiency cores seem to be related to ISA consistency between the two core types. The Gracemont kernel supports the AVX2 and AVX-VNNI (256-bit) instruction sedollar series, something that shouldn’t be on Intel’s “small” kernels. The net result of the blue team’s efforts is that Gracemont achieves 40% more performance at ISO power or 40% less power at ISO performance than a Skylake core. Intel has struck a balance between the two, using the right power and clock speeds to achieve some performance parity between the E-core and Skylake cores used in Alder Lake.
All eight Golden Cove performance cores have received significant enhancements to Alder Lake-S chips. The tech giant claims these have 28% higher IPC than the Skylake core and an impressive 19% IPC gain over the 12th Gen Cypress Cove core.
Intel, which made improvements in its processors starting from the cache to all the fine details, surprisingly removed the AVX-512. This is probably because there isn’t a lot of demand for 512-bit AVX right now, and Intel wants some ISA consistency between P-core and E-core.
Hybrid Architecture and Intel Thread Director
Intel Thread Director is a highly specialized middleware that interfaces with the operating system and software on one side and two sets of CPU cores on the other. Its job is to analyze a workload and help the OS scheduler distribute it at a granular level (both at the process level and at the thread level) among P-core or E-core clusters. It also requires the Windows 11 operating system for this process, but Windows 10 should still work fine with chips because Alder Lake also includes support for “preferred cores”.
Windows 11 also introduces the concept of quality of service (QoS) for software. This basically allows applications to communicate with the operating system about workloads and gives Windows Scheduler a clue as to whether P cores deserve their resources or whether they can be relegated or even restricted to E cores.
To achieve this, Thread Director can monitor the operation of the processor with nanosecond precision. There is mostly ISA consistency between the two types of cores, but some processes may need features available only on P cores, such as AMX or DLBoost. Thread Director, on the other hand, will ensure that such operations are allocated to performance cores only. Dialogue between Thread Director and OS scheduler also ensures that processes that are in the background or idle are routed to the productivity cores.
Intel is only releasing the Z690 chipset along with the 12th generation K and KF processors. Mid and entry-level chip sedollars will join the range in 2022. Most motherboard manufacturers have two different classes of motherboards based on the Z690. One of them uses newer DDR5 memory and the other uses current generation DDR4 memory. Alder Lake chips offer support for both standards.
The Z690 motherboards are generally of interest to hardware enthusiasts, as they provide all the overclocking features found in unlocked processors. The Z690 chipset communicates with the processor via a DMI Gen 4 x8 chipset bus that we detailed earlier. With a bandwidth of 128 Gbps per direction, this bus is superior to the standard ones offered today.
Just as the Alder Lake processor has a partial Gen 5 and a partial Gen 4 PCIe interface, the Z690 chipset has a Gen 4 and Gen 3 general purpose PCIe interface. The chipset uses 12 PCI-Express Gen 4 stream lanes and 16 PCI-Express Gen 3 stream lanes. The 4th generation stripedollarser will come in handy for PCIe 4.0 NVMe SSDs or NVMe RAID setups that can take advantage of the bandwidth, as well as next-generation Intel’s Thunderbolt controllers or third-party USB4 controllers. The 16 Gen 3 lanes provide network interfaces and additional USB3 controllers, etc. It will cover almost everything.
The Z690 PCH features eight SATA 6 Gbps ports and an integrated SATA RAID controller with full Optane Memory support. As for USB interfaces, up to four 20 Gbps USB 3.2 x 2 ports, up to ten 10 Gbps USB 3.2 x1 ports, up to ten 5 Gbps USB 3.1 1 x 1 ports and up to fourteen USB 2.0 ports available. The chipset also integrates a two-wire Ethernet MAC supporting the latest generation 2.5GbE and 1GbE controllers. Kardollarsar with WLAN will support the latest Wi-Fi 6E denedollar receivers.
Overclocking also gets a little more complicated with the new Hybrid core architecture. Features offered include the ability to overclock E cores separately from P cores, internal and external base frequency options, XMP 3.0 SPD enhancements, and the ability to instantly overclock both DDR5 and DDR4 memory types while the system is running, which is not yet available on the AMD platform.
Efficiency cores have a base clock multiplier called xE that works separately from the core rate of P cores called xP. The xG multiplier determines the iGPU frequency and the frequency at which the xR Ringbus interconnect and L3 cache operate.
12th generation processors have a built-in clock-generator, which can be sufficient for all areas and reduces motherboard cost. However, some premium motherboards have a separate, external clock generator with a wider or more detailed frequency range. This external BCLK can be used to tune the frequency of certain areas or all. The added complexity of the SoC brings with it additional voltage fields and offsets.
On the other hand, there are 20 new tuning buttons for overclockers to fine-tune their CPUs. Intel has announced the new Extreme Tuning Utility version 7.6, which gives you access to all these controls and simplified overclocking with the new Speed Optimizer feature. Speed Optimizer will add a fixed 100 MHz overclock option to your CPU.
Along with the new DDR5 standard, it introduces Extreme Memory Profile (XMP) 3.0, a feature that makes it easy to implement frequency and latency on an unlocked platform. XMP 3.0 features longer profile names of up to 16 characters, three of which are manufacturer-supplied and two of which are rewritable, and up to five stored per DIMM, and support for DDR5 PMIC (on-module power management).
Intel is constantly researching how to best transfer heat from the die to your heatsink, and with the 12th generation it has made another change. Much like Comet Lake, the new Alder Lake silicon uses a very thin silicon substrate. But the solder thermal interface material (STIM) layer is now much thinner. Besides, the copper integrated heat spreader is much thicker and can absorb more heat. As you know LGA1700 socket is longer than LGA1200 and therefore has more surface area for heat transfer.
The new Dynamic Memory Boost feature takes advantage of the instant memory overclocking feature unique to Intel memory controllers. This feature will allow you to switch between the JEDEC default and a variety of XMP 3.0 profiles on your memory modules depending on system workloads to conserve power.
With the 12th generation, the definition of “PL” power levels also changes completely. PL1 is now an abstract concept and no longer exactly matches the advertised TDP for a particular processor. As you know all unlocked K/KF models have long been rated 125W “processor base power”.
The company will no longer use the term “TDP” in its documentation or marketing process. The PL2 value we know will now be referred to as “maximum turbo power”. The default value of these two definitions for all chips available today is 241W, so we can say PL1=PL2=241W.
Moreover, Intel has deprecated the Thermal Velocity Boost feature, which comes into play on well-cooled chips. The Core i7 and Core i9 SKUs carry the Turbo Boost Max 3.0 and Turbo Boost 2.0 algorithms, while the use of Turbo Boost Max 3.0 is based on preferred cores (managed by the Thread Director). The Core i5 series is only supported with Turbo Boost 2.0.
Intel claims its new 12th generation processors are the best in the world for gaming. It is emphasized that the Core i9-12900K is on average 13% faster than the 11900K in the 31 games tested. This comparison was performed at 1080p with High settings using the tesdollarseries GeForce RTX 3090.
Blues also state that the 12900K is about 30% faster than the Ryzen 9 5950X. However, Intel has Before performance patch for AMD CPUs Acknowledges that it was made on the Windows 11 operating system. The manufacturer claims that the 12900K will still be the “world’s fastest gaming CPU” when using DDR4 memory, but no data has been provided to support this.
On the other hand, it is worth noting that tesdollarserin 11 tested it with virtualization-based security feature enabled. In a recent test with Windows 11 and Windows 10, it was revealed that the VBS feature negatively affects the Core i9-11900K performance in games. However, the level of performance impact depends on the game.
Intel is aware that it will have many difficulties on the software side with its hybrid CPU architecture and is working on this issue. We mentioned that Alder Lake processors may bring incompatibility that closely concerns players, and that old games may not work on processors. Denuvo DRM is not initially compatible with Alder Lake, but Intel is working with Denuvo to fix this and this will need to be applied as a patch to affected games. Intel says 91 games are affected and 32 are yet to be fixed, but 16 of them will be fixed before Alder Lake is released.
In terms of efficiency, it is noted that Adobe applications have gained over 30% when compared to 12900K and 11900K, and similar gains are achieved in applications such as Autodesk Revit. The blue team expects gains of at least 15% in the single-threaded productivity app, as with UL Procyon.
No part of the presentations compared 12th generation CPUs to AMD competitors in terms of productivity. They also don’t claim to offer the world’s best chips for productivity as with games. This suggests that Intel is unlikely to surpass AMD in terms of productivity, and the 10900K turned out to be faster than the 11900K at times.
Performance cores are 28% faster than the 10th gen and 14% faster than the 11th gen, while the efficiency cores are almost on par with the 10th gen in terms of IPC, although they are significantly more efficient. In fact, Intel believes the 12900K is 50% faster than the 11900K at the same peak power level and 30% faster when capped at 125W. in other words, it is noted that this new architecture is significantly more efficient, especially at low power levels.
- Gaming performance: Core i9-12900K, up to 25% more FPS in Troy: A Total War Saga, up to 28% more FPS in Hitman 3 and up to 23% more in Far Cry It provides serious performance increases from generation to generation, including FPS.
- Performance leap on the content development side: Intel says it has made improvements to multi-threaded performance specifically, and shared the following values:
-36% faster photo editing performance
-32% faster video editing performance
– Up to 37% faster 3D modeling performance
-100% faster multi-frame rendering
- Best overclocking performance: Offers overclocking tools for new processors, efficient cores, and ultimate performance customization, including DDR5 memory overclocking. Enthusiasts and gamers can try these new platform overclocking features as part of the latest Intel Extreme Tuning Utility (XTU) 7.5. Starting with the Core i9-12900K, XTU will also offer one-click overclocking support with Intel Speed Optimizer for unlocked 12th generation processors. In addition, Intel is introducing the latest Intel Extreme Memory Profile (XMP) 3.0 with DDR5 support, which offers additional profiles, including custom rewritable profiles and flexible tuning for memory overclocking.
New Architecture, Platform and Summary
Intel seems pretty confident that it has a product that can reclaim its performance and efficiency leadership in the market. The key to all of this is the massive 28% IPC gain that Golden Cove (P-core) cores provide compared to Skylake cores. In addition, small kernels (Gracemont) are also very successful and the name should not be deceived. Because efficiency cores allow Intel to catch up with its rival by comprehensively increasing multi-threaded performance.
Behind this work, the production process also plays an important role. Intel sonunda dökümhane kaynaklarını masaüstü segment için yeterli 10nm üretimi için ayırmayı başardı. 14nm++’dan 10nm Enhanced SuperFin’e (“Intel 7”) geçiş, Intel’in temel tasarımındaki birçok önemli değişikliği gerçekleştirmesine ve IPC hedeflerine ulaşmasına olanak tanıdı. Ayrıca bu çekirdekleri bol miktarda orta (L2) ve son seviye (L3) önbellekle desteklemek mümkün hale geldi.
241W’lık maksimum turbo güç değerleri ve 125W’lık temel güç değeriyle Intel, AMD’yi yakalamak için tüm termal boşluğu kullanmış gibi görünüyor. Şimdiye kadar Intel’in yayınladığı güç rakamları görmedik, ancak yeni işlemcilerin güç tüketiminin yüksek olması bekleniyor. Mavililer eğer 16 çekirdekli AMD Ryzen işlemcinin önüne geçersek, şüphesiz övgülerin büyük kısmı Gracemont’u tasarlayan ekibe gitmeli.
I/O tarafına gelirsek, Intel PCIe Gen 5 ve DDR5 dahil olmak üzere yeni nesil bağlantıları masaüstü pazarına getiren ilk isim oldu. Şirket geçen yıla kadar hala PCIe Gen 3 + DDR4 kombinasyonunu kullanıyordu. Sunulan tüm bant genişliğinden tam anlamıyla yararlanan PCIe 5. 0 ekran kardollarsarını görmemiz biraz zaman alacak. Ancak PCIe 5. 0 hadollarsarından bazılarını NVMe yuvalarına yönlendirebilen bazı anakart üreticileri, kullanıcılara önemli depolama avantajları sunabilir. Buna ek olarak SSD kontrolcüleri üreten şirkedollarser, 2022 yılının ikinci yarısında ilk PCIe 5. 0 SSD’lerin piyasaya çıkacağını tahmin ediyor. NVMe RAID kurulumları, Thunderbolt 4, USB 3. 2×2, 10 GbE LAN gibi birçok ek özelliği destekleyen, hem işlemciden hem de yonga setinden gelen bol miktarda bol miktarda PCIe Gen 4 bağlantısı görmek de bir diğer önemli artı.
DDR5 standardı bellek alt sistemine muazzam değişiklikler getirecek ve veri hızlarını nesilden nesle ikiye kadollarsayacak. DDR5, modül üzerinde güç yönetimi özellikleri, modül başına bağımsız bellek kanalları, dahili ECC, daha düşük modül voltajları ve nesil DRAM yoğunluğunda iki katına kadar artış sunuyor.
16 GB bellekler artık 8 GB’ın yerini almaya başlıyor ve çift kanallı kullanımla birlikte sistemlerde 32 GB bellekleri daha fazla göreceğiz. Müşteriler DDR4 yuvalarına sahip anakardollarsarı seçip eski belleklerini kullanmaya devam edebilir veya ucuz DDR4 RAM’lerden herhangi birini seçebilecek. Yeni teknolojilere geçmeyi seven ve yeterli bütçeye sahip kullanıcılar ise bellek üreticilerinin sunduğu sayısız DDR5 modellerinden birini tercih edebilecek.