What’s the best CPU for your next laptop? That was an easy answer just a few years ago, when Intel’s product line was far and away the strongest. But with multiple generations of AMD’s game-changing Ryzen chips finally giving Intel some real competition, you have more to think about.
We’re here to help you navigate this wider landscape, but without thousands of words and stacks of charts. We’ll start with a quick primer on the strengths and weaknesses of each chip, then we’ll discuss how to pick the right one for you. To keep this from getting too overwhelming, we’ll stick only to the mainstream CPUs that typically go into three-pound, thin-and-light laptops, rather than get into the high-performance chips that go into thicker and heavier gaming laptops.
Does it even matter which CPU you have?
One of the guidelines we’ll emphasize is to avoid paying for performance you don’t need. For most people who browse the web, check email, watch YouTube, or run Microsoft Office, just about any modern Core or Ryzen CPU with an SSD and enough RAM (at least 8GB) offers more than enough performance.
Rather than worrying about how the CPU was manufactured, or about some special feature it has, you may get more satisfaction by prioritizing the screen, keyboard, design, and price. This is especially true for 13-inch thin-and-light laptops.
There are still real upsides to a faster CPU. Zoom videoconferencing, for example, will not let you turn on its nifty virtual background feature without a newer quad-core CPU. Gaming is faster on newer chips, multi-tasking is better, and heavy content editing will, indeed, run faster on a new CPU.
One other new development is security features. Microsoft has said that Windows 11 will support only the latest CPU generations for security reasons. Even Microsoft’s older Surface devices don’t make the cut. Windows 10 will be supported until 2025, but if you want maximum future-proofing you should stick with newer chips.
One last warning: This guide is based on the performance of the Ryzen 7 or Core i7 versions. Lower-end models typically give you fewer cores, and slower graphics.
Intel’s 11th-gen Tiger Lake processors feature Iris Xe graphics, improved AI performance, and the ability to hit very high clock speeds with its four cores.
Intel’s 11th-gen Tiger Lake processors feature Iris Xe graphics, improved AI performance, and the ability to hit very high clock speeds with its four cores.
Intel’s 11th-gen Tiger Lake processors feature Iris Xe graphics, improved AI performance, and the ability to hit very high clock speeds with its four cores.
Intel 11th-gen ‘Tiger Lake’
Intel’s 11th-gen Tiger Lake CPU family is the company’s latest chip, built on a 10nm process with new SuperFIN transistor technology. Besides giving you more megahertz, it also features an all-new graphics core called Iris Xe.
Example model: Core i7-1185G7. You know this is an 11th-gen CPU, by the ‘11’ in the four digits after the Core designation. Larger numbers means more performance on the CPU side. For graphics, the G and a larger number means more performance. So far, there are only G4 and G7.
Strengths: The main strength of 11th-gen chips that most will feel is in pure megahertz. The CPU can reach 4.8GHz on very easy tasks, and 4.3GHz on more CPU-intensive tasks. This typically gives 11th-gen laptops a very “snappy” feel. The HP Elite Dragonfly Max with Core i7-1185G7, for instance, blew past similar competition in mainstream applications (though its thin-and-light chassis constrained it in other tests).
Intel’s 11th-gen series have other strengths. They are the only CPUs that support PCIe Gen 4, which means when paired with a Gen 4 SSD, you’ll see storage performance unmatched by AMD or by older Intel generations. AI performance has also improved, letting the chips excel over older Intel CPUs for doing such things as sorting your photos or upscaling an image. The 11th-gen CPUs with G7-level graphics enjoy the fastest integrated graphics today, and can give GeForce MX GPUs a run for the money.
Microsoft and hardware vendors continue to add security features directly into the CPUs. If you want the most secure CPU with the best performance, you want the newest chip, which gives the 11th-gen a big leg up over 10th-gen and older chips.
Weaknesses: The main weakness with 11th-gen is that it has “only” four cores. Because of other improvements it easily matches or sprints away from its 10th-gen predecessors—even those with six cores. AMD’s Ryzen 4000 and 5000 series, however, feature four to eight cores. While Ryzen 4000 runs at a slight deficit to Intel’s 11th-gen chips, the Ryzen 5000 largely pulls up dead-even or slightly faster on lighter tasks—and achieves better performance on multi-core tasks.
Recommended use case: Intel’s 11th-gen Tiger Lake supersedes all previous Intel chips for basically all needs. Chips with ‘G7’ in the name have the fastest version of Intel’s new Iris Xe graphics, meaning they can run recent AAA games with few compromises–and they surpass what’s available on AMD’s CPUs. The CPU should also be the snappiest in Office, web browsing, and even many photo chores. Although still rare, advanced AI-accelerated applications also favor Tiger Lake. It’s arguably the best overall CPU for what most people do on small laptops.
Intel
Intel’s 10th-gen Ice Lake chip was its first widely-used 10nm chip. Its inability to hit higher clock speeds gave AMD’s Ryzen 4000 an edge.
Intel 10th-gen ‘Ice Lake’
Ice Lake was Intel’s first 10nm CPU. It brought redesigned cores and was the first laptop chip with support for AI. It featured an updated graphics engine, too. Although it’s being phased, out, we still see 10th-gen Ice Lake laptops for sale.
Example model: Core i7-1065G7. You can tell this is a 10th-gen by the four digits starting with ‘10’ after the Core designation. Intel changed up its naming scheme with the new generation, so we show it below to help you decipher the rest. For our purposes, note that the fourth digit denotes power level, so a ‘5’ in that spot means the chip uses more power than a sibling chip with a ‘0’ in that spot. With graphics, there are three levels from G1 to G7, with bigger numbers again meaning better performance.
Rob Schultz / IDG
Intel’s use of 10nm 10th gen and 14nm 10th gen made for a confusing mess for consumers.
Strengths: The 10th-gen’s main strength is its efficiency improvement. Despite not hitting the high clock speeds or megahertz of Intel’s other 10th-gen chips, performance didn’t suffer as you might expect. The chips support DL Boost, which helped with AI image processing. New graphics cores improved noticeably over the company’s very tired HD and UHD graphics.
Weaknesses: In addition to its lower clock speeds, Intel’s 10th-gen Ice Lake series’ main weakness is the lack of CPU cores, topping out at four. AMD’s Ryzen 4000 increased the number of CPU cores a consumer could expect in a thin-and-light laptop.
Recommended use case: Intel’s 10th-gen is a fine CPU for general use, lighter-duty gaming, and advanced AI workloads. The 11th-gen Tiger Lake is preferred, but for someone looking for a premium laptop for Office, browsing and photo editing, it’s still a very good CPU.
Intel 10th-gen ‘Comet Lake U’
Intel’s 10th-gen family gets a bad rap because of the confusion Intel created by releasing an advanced 10nm 10th-gen chip along with a 10th-gen chip using older 14nm technology. It got so confusing, we even wrote a story on how to pick between 10th-gen Ice Lake and 10th-gen Comet Lake. Although largely being phased out now (and even discontinued) we still see a few older stock laptops using 10th-gen chips.
Example Model: Core i7-10710U. The 5 digits in the Comet Lake U family tell you that it’s built on older 14nm technology. Graphics on the chips were all universally mediocre. As is with other CPUs, the larger number denotes more performance.
Strengths: The 10th-gen Comet Lake U is, for the most part, a direct descendant of the 8th gen “Whiskey Lake U,” and most of the CPUs offer virtually the same performance. The one standout was the Core i7-10710U, which had six cores and Hyper-Threading. It also had a decently high Turbo Boost of 4.7GHz. It’s hard to recommend Comet Lake U over an 11th-gen chip, but the reason to consider it would mostly be cost. We also would recommend it over 7th-gen and 8th-gen CPUs in the used laptop market, to ensure you get the latest security updates in the cores.
Weaknesses: The 10th-gen Comet Lake U gave you essentially the same graphics core as the previous five generations of CPUs. On CPU performance, Comet Lake U’s older 14nm technology also meant it could not run at very high clock speeds for very long. Even worse, using all six cores simultaneously drove the chip to the point of exhaustion.
Recommended use case: If all you want is good-enough performance and don’t care about gaming, multi-core applications, or AI, Comet Lake U is an adequate candidate. If that doesn’t sound like a ringing endorsement, you’re right—it isn’t. This chip has not aged well, and 11th-gen Core and Ryzen run circles around it. The honest truth, however, is that for most people on a budget—this is still more than enough performance.
Intel
Intel’s 8th-gen series bolted on more cores compared to its 7th-gen predecessors, but kept the same anemic graphics.
Intel 8th-gen ‘Whiskey Lake’
Intel’s 8th-generation CPU sounds old, but it’s still seen on occasion in laptops, especially those with commercial CPUs aimed at corporate customers. And obviously, if you’re buying a used laptop, you still see them. It’s built on an older 14nm technology, it doesn’t have the cores of Ryzen, the clock speeds of Tiger Lake, or the gaming chops of either. And yes, Intel actually had two different 8th-gen chips, the more current “Whiskey Lake” and the older “Kaby Lake-R” under the same umbrella.
Example Model: Core i7-8665U. Like all Intel CPUs, the larger the number, the higher the speed. The four-digit number that starts with ‘8’ tells you that it’s an 8th-gen chip. Graphics are essentially the same as in the previous 7th gen—nothing special. Strengths: The main strength of the 8th-gen CPU are its corporate security and management features in many of the models still being sold. If, for example, you needed 10,000 laptops that can be centrally managed and secured, an 8th-gen corporate CPU is the way to go. Laptops using 8th-gen chips (with Whiskey Lake being better) are preferred to older 7th-gen chips, as the latest security features are supported and Windows 11 (today anyway) does not support 7th-gen CPUs. Weaknesses: From a performance standpoint, there’s simply nothing special about this four-core chip anymore, beyond the fact that Intel turns on the corporate features on many of them. Graphics are mediocre, and it can’t hold the high clock speeds of the newest CPUs, nor match a Ryzen 5000 in anything performance-related.
Recommended use case: If you really don’t need a top-shelf CPU for most of what people do on a small laptop, this chip is still very solid. Yes, a new Ryzen or 11th-gen Core is much faster, but you’re unlikely to feel it all that much if all you do is run Outlook, Word, and a browser, and attend videoconferences. If you also need fleet management, the 8th-gen with vPro is still the go-to chip.
For a consumer, the main appeal of an 8th-gen Whiskey Lake laptop is that it’s still more than enough performance for what most consumers do on a laptop, such as browse and use Microsoft Office. It’s also a far, far better option than a lower-end Pentium N or Celeron N CPU, which we’d say is almost less than adequate for general use in Windows.
AMD
AMD’s 7nm Ryzen 5000 offers blazing multi-core performance with equally blazing single-core performance.
Ryzen 5000 ‘Cezanne’ (and ‘Renoir’)
Watching AMD’s CPU launches has been like watching a chessmaster’s strategic moves for the last few years. Ryzen 5000 was the latest stunning move by the company. Although not quite the crushing blow AMD pulled off on the Zen 3 in desktops, it’s still arguably the best CPU for many performance-minded shoppers.
Example Model: Ryzen 7 5800U. Like the previous 4000-series, the 5000-series adopts a 3/5/7 schema. Not all of the models feature the newest Zen 3 cores, though. You can tell the difference based on the model number. Even-numbered Ryzen 5000 U-class chips are Zen 3, while odd-numbered Ryzen 5000 are Zen 2-based (the same cores in Ryzen 4000). For example, the Ryzen 7 5800U and Ryzen 5 5600U and Ryzen 5 5400U are Zen 3-based, while the Ryzen 7 5700U, Ryzen 5 5500U, and Ryzen 3 5300U use Zen 2 cores.
AMD argues this is fair because it increased the clock speeds on the Zen 2 cores, which justifies the “5000” model number. For the purposes of this article, we speak of the Zen 3-based chips.
Strengths: The strengh of the Zen 3-based Ryzen 5000 chips is clear: multi-core performance that sets a new bar of what to expect from a thin-and-light laptop, and single-core performance on a par with, or better than, Intel’s 11th-gen chips. Graphics performance is also very impressive—only Intel’s best Tiger Lake chips with G7 graphics outpace it (and even that’s open to debate.) The newest Zen 3 version also supports Microsoft’s latest security designs, giving them a key advantage against older Zen chips.
Weaknesses: Obviously one weakness the CPU has against Intel is AI performance. Intel’s AI performance push felt fruitless at first, but we’ve seen enough software that uses it to believe it’s finally paying off. Ryzen 5000 is still not available in enough laptops yet. The ones it’s in are pretty spectacular and premium designs. But on premium laptops, Intel still far outnumbers AMD.
For those into the advanced features of Thunderbolt 3 or Thunderbolt 4, you simply can’t find it yet in Ryzen 5000. There is some concern over Ryzen 5000 performance on battery, but we recommend you read the test showdown linked earlier in this sentence to learn the nuances.
Recommended use case: You really can’t lose with AMD’s Ryzen 5000 U-series. You get awesome multi-core performance, awesome single-core performance, and great graphics performance too (for a thin-and-light laptop CPU). It also has the latest security features for Windows and even comes in many premium laptops. Frankly, if you can get one in a laptop that you like, don’t think twice about it.
AMD
AMD’s 7nm Ryzen 4000 is a game-changing CPU.
Ryzen 4000 ‘Renoir’
AMD’s Ryzen 4000 was a game-changing CPU that made the company’s chips the superior choice at its debut. It has really strong graphics capabilities with its Radeon graphics cores. Intel’s 11th-gen Tiger Lake and AMD’s newer Ryzen 5000 have taken some shine off the once-impressive Ryzen 4000, though.
Example Model: Ryzen 7 4800U. AMD adopts a similar model number schema as Intel’s, with the 7, 5, 3 family indicating best, better, good. The first digit in the model indicates the generation, which is 4th, with the following three numbers indicating better performance the larger the number gets. The ‘U’ at the end indicates the power class, which is 15 watts here. A gaming laptop would actually use the same chip, but tuned for more heat and given an ‘H’ designation.
Strengths: AMD’s Ryzen 4000 is built using an advanced 7nm technology, which lets AMD put up to 8 cores in each CPU along with its Radeon graphics cores. The CPU is shockingly fast in multi-core performance and can even make much larger laptops using Intel 9th-gen and 10th-gen CPUs sweat.
Weaknesses: Ryzen 4000 at launch was a true ground-breaking CPU in multi-core performance in a thin-and-light laptop. Intel’s 11th-gen Tiger Lake didn’t change that, but because most people do light-duty tasks in ultra light laptops, it put a dent in Ryzen 4000. We’d say Ryzen 5000 set it back further by being good at light tasks and heavy all-core tasks. It also has a weakness in AI, and the lack of Thunderbolt 3 and 4 stings a little too.
Recommended use case: Ryzen 4000 today has a good place as a step or two down from the top-shelf but still pretty awesome. In multi-core performance, it still shines, and its light-duty performance is fine—it just doesn’t sing like its Ryzen 5000 sibling or Intel’s newest chips. For someone on more of a budget, it’s still a great chip and choice.
Reddit
A picture of the Microsoft Surface Edition Ryzen 7 3780U surfaced soon after the new laptop was released. It’ll feature a quad-core chip built on a 12nm process and up to 11 graphics compute units.
AMD Ryzen 3000 ‘Picasso’
AMD’s Ryzen 5000 gets all the love and for budget buyers, and Ryzen 4000 ain’t bad either. We mostly include the Ryzen 3000 chip as there might be one or two laptops still using it. Of course, used laptops still feature it too.
Example model: Ryzen 7 3780U. AMD uses the familiar Intel-like 3/5/7 to denote good, better, best. The first digit in the model tells you it’s 3rd-generation chip, while larger numbers indicate more performance. The ‘U’ tells us it’s a lower-power CPU.
Strengths: Ryzen 7’s strength was mostly in graphics, where it held a lead over most CPUs with Intel HD graphics. It was also mostly competitive with Intel’s 10th-gen chips in compute tasks.
Weaknesses: The Ryzen 3000 is showing its age these days. We’d recommend a newer-generation Ryzen, or Intel’s 11th-gen or even the older 10th-gen over the Ryzen 3000 today. Far worse, though, is the battery life, which is unimpressive for Ryzen 3000.
Recommended use case: Like 8th-gen Whiskey Lake CPUs, Ryzen 3000 is best left to people that just don’t care all that much about performance. If a half-second or slightly less responsiveness doesn’t matter to you, then like most users, you won’t be able to tell if it’s a Ryzen 3000 or Ryzen 4000 inside. And yes, battery life matters, but if that only counts when moving from the kitchen to the den, then it probably doesn’t matter all that much.