The competition between AMD and Intel continues to simmer, giving us great processors to test and use in more powerful systems. AMD's Ryzen chips are built on a more efficient platform, allowing the company to get more performance without using too much power. Intel is struggling with this in its manufacturing facilities, but we're confident about the upcoming launch of the 15th Gen Arrow Lake processors.
That's right, all eyes are on Intel's upcoming 15th Gen Arrow Lake launch as this could potentially bring AMD back and maintain some healthy competition. There's still a lot we don't know about these chips, but here's everything we know so far for those who are curious and can't wait to see what Intel has in store for us.
Intel will use TSMC's foundry … again
Should still bring noticeable improvements
Intel had originally planned to launch Arrow Lake processors using its 20A process, but has since confirmed that the chips will be manufactured on TSMC's platforms. Regardless of who makes the processors, these new chips should have significant advantages over Intel's 14th-gen SKUs. 14th-gen chips are manufactured on a 7nm process, putting them behind AMD and TSMC. The move to switch Arrow Lake to a similar TSMC process is another chip in the American foundry's armoury. The reason for this delay is that the smaller the process, the more transistors can be packed onto the silicon and the better the performance or efficiency.
This would be a significant upgrade for Intel, which has lagged behind the competition due to its internal issues. With TSMC, the company can still bring solid processors to market, but keeping everything in-house would have been a masterstroke. There have been rumors and speculation online about how much 20A would impact Intel's 15th generation processors. A post on AnandTech's forum suggested that we would have seen a roughly 15% increase in multi-core performance. With the move to TSMC, we still expect significant improvements over the current generation of processors.
Raptor Lake Refresh wasn't brilliant, but it gave Intel some time to focus on getting everything ready to make Arrow Lake a reality. 18A is also just around the corner.
Details about the Intel LGA1851 socket “Arrow Lake”
For next-generation Arrow Lake-S CPUs and beyond
Details about the next-gen LGA1851 socket have been known for a while now, thanks to 2D and 3D renderings created by Igor's Lab. Word has it that Intel's next-gen LGA1851 socket was originally intended for Meteor Lake-S desktop processors, but now it's making its way to 800-series chipset motherboards for Arrow Lake-S desktop processors, since Meteor Lake is, well, a mobile-only processor generation.
The overall dimensions of the LGA1851 socket remain largely the same as the LGA1700 socket, with only a slight difference in the Z-height. This distance, in case you were wondering, is the difference between the top of the motherboard and the top of the CPU IHS. This could prompt CPU cooler manufacturers to produce the necessary hardware. However, this is not a major problem, as Intel may have already sent the necessary documents to manufacturers detailing the changes we will see in Arrow Lake-S desktop CPUs and the LGA1851 socket platform.
The LGA1851 socket is expected to bring a whole host of changes, the first of which is the ability to connect a dedicated PCIe 5.0 x4 interface directly to the CPU. This is in addition to the x16 lanes leading to the GPU, meaning there's native support for PCIe 5.0 SSDs, which can reach insanely high read and write speeds.
AI inside
There is that buzzword again
Artificial intelligence (AI) continues to be the hot topic of the day and we expect Intel to capitalize on the demand. Arrow Lake processors could use the Intel Ce-LPG Plusm GPU architecture, which will benefit from Extended Matrix Extensions (XMX), an AI acceleration engine. This could allow the new Intel chips to speed up data processing and increase graphics and video performance. Intel markets Arrow Lake as the “world’s first gaming CPUs with [an] AI accelerator.”
With the advent of the AI PC, Intel is positioning itself with accelerators integrated into the CPU package, giving it a strong position against AMD.
Arrow Lake could launch in late 2024
Without Hyper-Threading and DDR4 support
Intel announced a launch of Arrow Lake in 2024 or 2025, but we firmly believe the company will launch the new processor family later this year.
What we will see in terms of cores and threads is pure speculation. But if the leaks and rumors are to be believed, it looks like Intel Arrow Lake-S desktop processors will have up to 8 P-cores and 16 E-cores for a total of 24 cores and 32 threads. That, in case you were wondering, is similar to the existing 14th and 13th generation flagship CPUs on the market. The chips are also said to have up to 125W PL1 TDP and a maximum base clock of 3.5GHz.
The leaks also suggest no Hyper-Threading with mentions of 8 IA cores and 8 threads. Some other sources have published similar information suggesting that the Arrow Lake chipset will not have Hyper-Threading support. However, the Arrow Lake chips are said to have an NPU that should offer similar performance to the Meteor Lake NPU. The Arrow Lake chips are also said to drop support for DDR4 memory, so you might have to shell out money for some DDR5 memory kits.
Related
Best DDR5 RAM
If you are looking for the best DDR5 RAM modules on the market, we have something for you.
Wait for the start
As with AMD processors and Nvidia GPUs, I would recommend waiting until we know more before launch and performance figures are confirmed. Modern processors (and I'm talking about AMD Ryzen 6000 and Intel Core 12000) are fantastic chips with plenty of power to handle more demanding games and software. There's no need to upgrade to the latest and greatest version to enjoy a smooth and stable experience. The same goes for Arrow Lake, so we'll have to see how good these chips are.