This year, at the first digital CES of all time, Intel is announcing a series of new processors in several product families. There are new updates coming for the Pentium Silver and Celeron budget families, new 11th generation vPro and Evo vPro systems, future eight-core Tiger Lake CPUs, new next-generation Tiger Lake quad-core gaming and new data on future products like Rocket Lake. Here, we will focus mainly on the new CPU announcements.
Intel is placing a lot of emphasis on the Chromebook market, with the first 11th Generation CPUs being launched in that space at the end of this quarter. When Chromebooks first appeared, they were typically powered by ARM CPUs, but we’ve seen a number of x86 devices launch in recent years. Chromebooks were particularly popular in the COVID-19 era; PC market reports for the third quarter of 2019 showed a 1.9x growth in shipments in just one year. AMD has also been very interested in this market lately, so we’ll see how things evolve this year. We can see a low-cost grudge match between the two companies in this space if the Chromebook market remains high.
Tiger Lake gains high power
As far as Tiger Lake is concerned, Intel unveiled a new Tiger Lake (H35) gaming platform and tried out the next eight Lake Tiger eight-core CPUs. There are three new quad-core TGL chips – the Core i7-11375H “Special Edition”, the Core i7-11370H and the Core i5-11300H. All of these are 35 W chips:
Note: Intel has stopped providing basic frequencies for its processors on the 15W TDPs it had previously reported. Instead, it now reports the minimum CPU frequency if run in cTDP Down (28W) and cTDP Up (35W). This is a hostile change to the consumer that overshadows basic information about the minimum operating frequencies that customers should expect. Without knowing whether a laptop is designed to operate in one TDP range or another, the consumer has no way of comparing the expected performance. The end user deserves to be aware of the minimum CPU clock expected in all cases.
One of the problems with cell phones is the degree to which they overshadow information about their own real clock speeds and often market chips based on wristwatches only. This is not a trend that the PC industry should copy.
With that said, these CPUs offer the advantages you would expect from a higher TDP – higher boost clocks and probably more sustained turbos, delivering higher overall frequencies. They won’t make a big difference, but they will probably provide better experiences under load.
In the meantime, Intel is not yet ready to talk about the name of its eight-core Tiger Lake CPU, but is willing to confirm the chip’s existence. This CPU will offer up to eight Willow Cove performance cores, with full support for PCIe 4.0 and 20 support tracks. Since GPUs do not currently benefit from an x16 PCIe 4.0 connection – PCIe 3.0 can power any modern card – it will be interesting to see if any companies are using some of these avenues to enable multiple x4 connections for M.2 storage arrays on mobile devices , instead of using them only for GPUs. Unfortunately, we will probably see a much more focused use of GPU than M.2, but we can wait.
Presumably, the eight-core TGL CPU will fall in a TDP range between 35W – 65W. Intel’s 10th generation Core i9-10980HK has a base frequency of 2.43 GHz, but a TDP range of 45W – 65W, so the next TGL CPU can come anywhere in this space. With peak frequencies up to 5 GHz, we think that this is not a low power chip.
This eight-core CPU will represent the most advanced and most powerful variant of an Intel architecture that you can buy in 2021, although it may not be the fastest in absolute terms. Rocket Lake, which we will discuss below, is based on Cypress Cove, also known as Sunny Cove, also known as the CPU within Ice Lake – not Tiger Lake.
Rocket Lake arrives
Finally, we have the imminent arrival of Rocket Lake, which Intel also confirmed as arriving in the first quarter of 2021. The Core i9-11900K is capable of a 5.3 GHz turbo and a 4.8 GHz all-core turbo, and will be provided with support for DDR4 -3200, above DDR4-2933. Normally, it is possible to operate Intel CPUs at higher RAM frequencies than that, but Intel chips do not benefit from higher RAM clocks to the same degree as AMD chips.
Rocket Lake will be backward compatible with Intel 400, and Comet Lake-S CPUs will be compatible with 500 series motherboards, except Celeron CPUs with only 2 MB of cache. These specific 10th generation CPUs will not be supported. This should not affect many people, as anyone with a state-of-the-art Celeron on a 10th generation motherboard must have enough upgrade space within the Comet Lake family to make a new chip worthwhile without having to change the motherboard platforms.
Intel is forecasting a 1.19x improvement in the IPC for Rocket Lake, which would be in line with our own expectations and indicates the full benefit of Sunny Cove from Ice Lake successfully backported to the Rocket Lake 14 nm process. There will be some new features to improve the chipset’s bandwidth, add AV1 decoding and better integrated graphics performance compared to previous generation CPUs. Intel also believes it can tie or outperform AMD gaming with this next chip:
The 5900X must be a fair comparison to the 11900K in games – the 5000 series does not show the same gap in gaming performance of the high end CPU as the 3000 series shows. Intel image, all recommended benchmarks with salt.
It is possible that the company can actually do this. AMD’s Zen 3 architecture is faster than Intel’s current 14nm CPUs across the board, with the possible exception of some well-tuned AVX-512 workloads, but games are probably the place where the leadership position from AMD is weaker. Intel may not be able to work with 12-16 core CPUs at absolute performance, but strong single-threaded performance may allow the company to recover games. We will see in a few months how accurate the above benchmarks, provided by Intel, are compared to the verified results independently.
Now read: