CPU, APU & Chipsets News - Page 4
Intel has just confirmed that its next-next-gen 14th Gen Core "Meteor Lake" compute tile has taped in, meaning that the design is finalized and can be completed to tape out the whole chip.
Intel confirmed the news with Executive Vice President & GM of Intel's Client Computing Group, Gregory M Bryant, tweeting out that it was a "Great way to start the week! We are taping in our 7nm Meteor Lake compute tile right now. A well-deserved celebration by the team on this milestone".
The new Meteor Lake architecture should be powering Intel's next-next-gen 14th Gen Core CPUs which are expected in 2023, they will be following up the next-gen 12th Gen "Alder Lake" CPUs coming in 2022 with DDR5 and PCIe 5.0 support and the 13th Gen Core "Raptor Lake" CPUs after that.
AMD has been ahead of the curve for a while now, with its chiplet-based Zen CPU architecture making waves through the industry... but the industry moves so fast it's always time for change. Well, that change comes in the form of "X3D" packaging, and it's coming to codename Milan-X.
Patrick Schur tweeted out today that AMD is "working on a new CPU" that is codenamed Milan-X that will use stacked dies. ExecutiveFix backed up the leak, saying that codename Milan-X will use X3D stacking technology and will feature the Genesis-IO die, which is the codename of the current-gen Zen 3-based EPYC architecture.
This won't be inside of your new Ryzen CPU, but rather it will be what AMD will use to fight its competitors in the data center market. AMD recently teased a simplified diagram during its Financial Analyst Day that packed four compute tiles in a 2 x 2 pattern, with stacked memory surrounding the chip. This hybrid approach is called X3D, as it combines Hybrid 2.5 and 3D stacking technologies.
AMD's next-gen Zen 4 "Raphael" LGA1718 package has been teased, this time through a new mockup that will eventually be what we'll see as the Ryzen 7000 series CPUs.
The new Raphael CPUs will reportedly have a flagship 170W TDP model, but for the most part will be 120W TDP designs. AMD will have DDR5 support with its new AM5-based motherboards, meanwhile, Intel will support DDR5 on its higher-end boards, but will offer DDR4 support on other Intel 600-series chipsets.
Zen 4 will have 28 PCIe 4.0 lanes on offer, an upgrade over the 24 PCIe 4.0 lanes on the Zen 3 series. Intel will have next-gen PCIe 5.0 connectivity on its upcoming 12th Gen Core "Alder Lake-S" processors, which will be directly battling against AMD's new Zen 4 "Raphael" CPUs.
We last heard about AMD's next-gen Zen 4-powered EPYC processors a few months ago now back in February 2021, and now there's a new (but old) roadmap teasing Zen 4 embedded products.
The roadmap is sold old here that the Zen 3-based EPYC 7003 series processors are a "concept" yet they have already been released, so the roadmap is a few years old. On this roadmap, we can see the Zen 1-based EPYC 7001 series CPUs with their planned up to 32 cores and 64 threads meanwhile the Zen 2-based EPYC 7003 series had up to 64 cores and 128 threads.
But AMD wasn't finished there, as you can see the Zen 4-based EPYC 7004 series will have over 64 cores, with previous rumors pegging it to offer a monstrous 96 cores and 192 threads. You will also notice the new Zen 4-based EPYC 7004 series has a TDP of between 120W and 230W+ or more, while the other EPYC chips tapped out at 225W.
It wasn't too long ago that the AMD 4700S PC Desktop Kit was spotted online, and now it's out in the wild in some markets at least -- and someone has taken some pictures of the motherboard
The pictures were posted on Twitter by "DISCLOSUZEN" but the CPU cooler wasn't removed, so we don't have a juicy shot of the chip itself. They are the first pictures of the AMD 4700S PC Desktop Kit, with the AMD 4700S processor based on the Zen 2 architecture and packing 8 cores and 16 threads at up to 4GHz.
AMD also has 16GB of GDDR6 memory on its 4700S processor, which should be on the back of the motherboard and under what looks to be a custom heat sink. There's a PCIe 3.0 x16 slot for a discrete GPU, which makes sense given that AMD has most likely disabled the integrated GPU, or it defective and thus, disabled anyway.
Intel has just launched its new 11th Gen Core "Tiger Lake-H" processors with the shining star of the show being the flagship Core i9-11980HK processor, which promises ultra-performance inside of the next generation wave of laptops in 2021.
The new Intel Core i9-11980HK offers up 8 cores and 16 threads on the new Willow Cove architecture, on Intel's in-house 10nm SuperFin technology. We have a fully unlocked processor with Intel Extreme Tuning Utility (XTU) for "enthusiast overclocking" and Intel Speed Optimizer for easy overclocking.
There's 20 PCIe 4.0 lanes on the Core i9-11980HK as well as 24 PCIe 3.0 lanes, more than enough to cover the fastest discrete GPU that can go alongside the new CPU -- being NVIDIA's GeForce RTX 3080 GPU -- as well as a couple of ultra-fast PCIe 4.0 SSDs.
Intel has finally rolled out its new Tiger Lake-H processors in both 6-core and 8-core CPU form and all on the new 10nm node. The new flagship Core i9-11980HK processor is an 8-core, 16-thread chip at up to 5.0GHz and is the first overclockable 10nm CPU from Intel.
Intel is debuting its new Willow Cove core architecture inside of the new 11th Gen Core-H series CPUs and on its 10nm SuperFin technology. The company is touting a 19% performance improvement over the previous generation, with Intel using a 45W power envelope on the new 11th Gen CPUs -- except for the flagship Core i9-11980HK processor which has a TDP between 45-65W. The rest of the stack are 35-45W offerings with between 6 and 8 cores.
You've got Thunderbolt 4 support, up to DDR4-3200 memory support, and Intel Optane Memory H20 offering second-gen Optane + NAND hybrid SSD for what Intel says will deliver "amazing responsiveness in real-world client usages".
Samsung is hard at work on its next-gen Exynos SoC which will be competing against Qualcomm's next-gen SoC, and Apple's current-gen M1 chip and next-gen M2 chip.
But what will be inside of Samsung's new SoC is going to be interesting: it's aimed at notebooks and will pack an integrated custom GPU from AMD which I'm sure is based on the RDNA 2 architecture. The news is coming from The Korean Economic Daily which said an industry source told them: "The new Exynos will offer improved functions, including extraordinary computing power and battery efficiency, by utilizing a 5-nanometer processing technology. It's good for both laptops and smartphones".
We should expect Samsung to unveil a new family of Exynos SoC-powered laptops and smartphones using its new chip, which will be made on the fresh 5nm node. We should see the Exynos SoC scale up from smartphones with lower power and TDP through to laptops with varying power and TDPs.
Intel's largest launch in a very long time will be later this year with its 12th Gen Core "Alder Lake-S" CPU family, which will reportedly be launching in November 2021 according to the latest rumors.
The introduction of Alder Lake-S will usher in Intel as the first with PCIe 5.0 and DDR5 memory support, after AMD held PCIe 4.0 over Intel's head for a couple of years there, Intel will be the first with both PCIe 5.0 and DDR5 support. PCIe 5.0 will be on all 600-series motherboards, but we will see some boards launching with DDR5 or DDR4 support -- some of that will be up to motherboard makers.
Wccftech says that Intel has told its partners that it is "planning to launch Alder Lake S in November of 2021. As with all timelines, this can move backward or forwards depending on a few variables but for now this has been communicated to partners and is mostly fixed".
We are hearing some new rumbles on AMD's next-gen Ryzen 6000 series APUs codenamed Rembrandt, with leaker ExecutibleFix pumping out some new details.
In the tweets we're told that AMD's new Rembrandt APUs will be using the new Zen 3+ architecture, RDNA 2 GPU cores and be on TSMC's new 6nm node. We heard all of this nearly a year ago now with previous rumors: Zen 3+ architecture + RDNA 2 architecture + TSMC 6nm node.
But the new rumors are teasing that the Rembrandt APU packs 12 Compute Units of RDNA 2 GPU cores, which will be interesting to see how gaming performance will end up when the new APUs launch in 2022. The current-gen Cezanne APUs feature 512 stream processors based on the previous-gen Vega GPU architecture, while the new Rembrandt APU cores will pack 768 stream processors -- a 50% increase.