Category: Article

The Lenovo ThinkPad L14 is successor to last year’s 14-inch ThinkPad L490. During last year’s review, we noticed that neither the display nor the CPU performed particularly well despite what turned out to be an overall decent package. This year, for the first time in the history of Lenovo’s L-series, you can choose between an Intel and an AMD processor. Lenovo’s L-series sits right in-between the higher-quality T-series and the more affordable entry-level E-series and is aimed primarily at business users. Given that the L14 shares many of its exterior attributes with its predecessor we are going to focus mainly on the new AMD CPU in this review.

Case – Typical ThinkPad

As expected of Lenovo overall build quality is solid. The ThinkPad L14 neither creaks nor flexes considerably, and the entire top of the 14-inch case is made of roughened plastic despite of which it manages to attract fingerprints. The thin display bezels are slightly rubberized and sit flush with the palm rests with the display lid closed. The two hinges are firm and offer the perfect amount of resistance required to ensure that one-handed opening remains possible. The maximum opening angle is 180 degrees.

Maintenance is simple and straightforward on this model as well given that the bottom panel can be easily removed by undoing a few screws. This will give you access to the most important components, such as for example the RAM which can be upgraded up to 64 GB. The battery, unfortunately, is not user replaceable.

Connectivity – AMD brings HDMI 2.0

Connectivity remained unchanged over last year’s L490. The L-series continues to lack a Thunderbolt port, which is only available on the T-series. Port layout has also remained unchanged over the ThinkPad L490. The main difference is the new HDMI port, which now supports 4K resolutions at 60 instead of 30 Hz.

Keep in mind that the HDMI port differs between the Intel and AMD models. While AMD SKUs support HDMI 2.0 (4k at 60 Hz) Intel SKUs are still limited to HDMI 1.4 (4K at 24 Hz).

Display – Only Marginally Better

Last year’s L490 featured a different panel than its predecessor, and it was barely any brighter than the L480’s display that was significantly too dim. This year, we see yet another panel in the ThinkPad L14, the Innolux N140HCA-EAC – basically a cheaper model of the ThinkPad T14s’s Innolux N140HCG-GQ2 panel.

The differences between the L and the T-series are very obvious in regard to maximum brightness (244 vs. 371 nits) as well as color space coverage, and the L14 is far behind its more expensive business sibling. When compared with the L490, brightness remained practically identical with some improvements in contrast ratio and black level. That said our review unit still managed to outperform the HP and Dell.

Outdoor usability was decent even in bright environments, but you should definitely avoid direct light sources. In order to get comfortable, you need an overcast day or at least full shade.

The panel itself is a relatively inexpensive IPS panel that continued to suffer from major color deviations even post calibration. The L14 is thus unsuitable for serious photo editing but should work perfectly fine in typical office and business environments.

Performance – No Throttling with AMD

Processor

AMD’s Ryzen 5 4500U does not support Hyperthreading on its six Zen 2 cores and thus features exactly as many threads as it has cores to offer. It does, however, run at a slightly higher base clock speed of 2.3 GHz with a turbo boost of up to 4 GHz. Additional details can be found on our dedicated page.

In multi-core benchmarks, the AMD Ryzen 5-powered ThinkPad L14 performed similarly to the recently reviewed ThinkPad E14 and outperformed its identically equipped HP ProBook 445 and Dell Latitude 14 competitors. We also found no evidence of throttling as performance remained stable and consistent even after 20 consecutive runs into our CineBench loop, much unlike Intel CPUs whose performance tends to drop significantly after the first run. Long story short: Intel CPUs seem to be more susceptible to performance throttling than AMD CPUs.

All Ryzen 5 notebooks performed almost identically in our single-core benchmark and managed to outperform Intel’s Core i5 by a small margin.

System Performance

Compared to its competitors, the ThinkPad L14 performed average in PCMark 10. Nevertheless, it performed above average for an office and business notebook, which became noticeable not only in this particular synthetic benchmark but also during everyday use. 

Storage Devices

Our review unit featured a 512 GB Wester Digital SSD (WDC PC SN730 SDBQNTY-512GB) whose very important (from the operating system’s point of view at least) CrystalDiskMark 5.2 read/write 4K performance was lower than on all of its competitors. In AS SSD, its 4K read performance was lower than expected as well. We also found a significant performance difference between running on battery and running on mains. Without an active power supply, the read seq numbers dropped by almost 40 %.

GPU Performance

The Ryzen 5 features an AMD Radeon RX Vega 6 GPU with 8 CUs running at 1.2 GHz. This new GPU is a giant leap compared to the L490’s ancient Intel UHD Graphics 620, which is best exemplified by the fact that scores in 3DMark 11 doubled. Thus, the AMD-powered L-series is capable of taking on its Intel-equipped T-series that continues to include Intel’s UHD Graphics 620, and thus only performed half as well in our benchmarks. That said our review unit could have achieved even better scores with dual-channel RAM instead of its default single-channel configuration. Take for example the ThinkPad E14 with the very same iGPU as the L14 but dual-channel instead of single-channel mode RAM: it ran up to 29 % faster.

The L14 passed our stress test with flying colors. You will find more details in the emissions section under temperature below. Performance dropped slightly on battery. In 3DMark 11, we noticed a 5 % decrease in CPU (6,664 vs. 7,061 in Physics) and 9 % decrease in GPU performance (3,092 vs. 3,401 in Graphics).Gaming performance was right where we would have expected it to be. Overall, the L14 performed slightly below average for an AMD Radeon RX Vega 6. AMD’s Radeon RX Vega 7, such as we find for example in Lenovo’s T14s, was significantly faster, and outperformed our L14 by 10 FPS on average in FHD, a very common gaming display resolution.

Gaming Performance

Gaming performance was right where we would have expected it to be. Overall, the L14 performed slightly below average for an AMD Radeon RX Vega 6. AMD’s Radeon RX Vega 7, such as we find for example in Lenovo’s T14s, was significantly faster, and outperformed our L14 by 10 FPS on average in FHD, a very common gaming display resolution.

Emissions – Quiet Fan, Quiet Speaker

Noise Emissions

Just like its predecessor and the ThinkPad T14s our review unit remained pleasantly quiet overall. It peaked at just 32.8 dB(A) during our stress test, which was a pleasant surprise. Fan noise frequency was also very pleasant overall and not at all annoying even after long periods of time. We found no evidence of coil whine on our review unit.

Temperature

Overall, the ThinkPad L14’s temperatures around the keyboard area were better than its L490 predecessor’s with Whiskey Lake that tended to reach up to 46 °C in the top right corner. In comparison, our AMD-powered review unit peaked at 43 °C. Unfortunately, temperatures around the palm rests are around 7 °C higher than on the L490. The L14’s hot spot is located at the bottom rear where it peaked at around 50 °C – too hot for lap use.

Power consumption very quickly settled at 25 W during our stress test resulting in around 2.8 GHz and around 890 MHz for CPU and GPU, respectively.

Energy Management – AMD Is Power Hungry

Power Consumption

Compared to its predecessor, power consumption increased by around 13 %. At maximum load it drew 50.6 W through its 65 W power supply. Our comparison table also reveals that Lenovo’s AMD-powered notebooks on average tend to have a higher power consumption than their Intel-equipped counterparts.

Battery Life

Compared to its Whiskey Lake-equipped predecessor the AMD Ryzen 5-powered L14 ran about an hour longer under full load but shorter during video playback in return. In this particular test, the L14 landed in last place of its comparison group.

Verdict – A Breath of Fresh Air thanks to AMD

Lenovo did an overall very good job with the ThinkPad L14, which continues to emphasize Lenovo’s traditional strengths such as its sturdy high-quality case, its pleasant-to-use keyboard, straightforward maintenance, and an interesting collection of ports and connectivity including a smart card reader as well as a docking port for business users.

On the other hand, Lenovo missed its opportunity to improve upon the L490’s weaknesses, such as for example the dim display or its overall poor speakers that lack both volume and bass. Lenovo’s choice of display panel is a sore point in particular, and we would have wished for a wiser choice.

The most important achievement for the L-series is the introduction of AMD’s Ryzen 5 4500U. In our reviews of the ThinkPad E14 and T14 the AMD SKUs proved to perform much better than the Intel SKUs, and the same holds true with the L14 as well. AMD is simply better.

There was no noticeable throttling and thus no drop in performance during our CineBench loop. In contrast, Intel CPUs tend to lose a lot of performance after the first run due to performance throttling. The price you pay for this increase in performance is a slightly higher power consumption as well as higher surface temperatures at the bottom under maximum load. The GPU’s comparatively poor performance can be easily remedied by adding a second RAM module.

AMD brings a breath of fresh air to Lenovo’s ThinkPad series, and changes the traditional rules. Intel’s dominance is about to falter as AMD not only offers lower prices but also a significant performance boost. Thus, we have no choice but to once again highly recommend the AMD SKU over its Intel counterpart.

Microsoft has added a new built-in command line tool called DiskUsage to Insider builds of Windows 10 that reports how much disk space folders use.

As first reported by BleepingComputer, the news outlet discovered the new utility when testing Windows Insider builds 20277 and 21277. DiskUsage is capable of scanning entire drives or specified folders and reporting how much drive space each folder is using.

The utility is located in System 32 and is still in the early stages of development. Currently some features are now working as expected and there are even some spelling errors in the the DiskUsage’s instructions.

Still though, admins may find it useful to analyze disk space from terminal windows though most Windows users will likely continue using popular tools such as WizTree or TreeSize.

DiskUsage

As DiskUsage is still in development by Microsoft, the tool currently reports file and folder sizes in bytes as opposed to doing so in megabytes or gigabytes. However, the software giant has included a /h argument to display file and folder sizes in a more conventional way.

It’s also worth noting that DiskUsage requires administrative privileges so Windows Insiders that want to test it out for themselves will need to open an elevated command prompt in Windows 10 before running the tool or else it will display an error.

The tool can be used for general disk space scans but users can also configure it to say, display all of the folders on a drive that are larger that 1GB in size. DiskUsage also includes other features such as creating configuration files with the options you want to use automatically, output customization and the ability to skip over certain folders depending on their type.

We’ll likely hear more about DiskUsage from Microsoft itself once the company is ready to ship the tool in an upcoming Windows 10 release. For now though, Windows Insiders running builds 20277 and 21277 can begin experimenting with it.

Since the debut of ASRock’s Taichi brand, we’ve been fans of its quality blend of price, performance, and features. It’s also one of few sub brands that doesn’t go all in on a ‘gaming’ theme, which is something many people appreciate. I know I do. With the Razer influence, however, that’s changed somewhat here. The traditional yin/yang and cog machinery aesthetic has been left off with the release of the B550 Taichi Razer Edition. 

This reinvented ASRock Taichi board features a minimalist, yet unmistakably premium appearance that integrates Razer Chroma RGB support. It sure is a lovely looking board, and if you’re invested into the Razer ecosystem, it’s a board you’ll definitely want to consider for a Ryzen 5000 series build.

Depending on which country you live in, at $299 USD the B550 Taichi Razer Edition may be the most expensive B550 board of all, it’s certainly one of the best B550 motherboards. This isn’t some bargain basement cheapie that belongs in the budget bin. It’s a fully featured premium tier board with almost all the bells and whistles you’d expect from an enthusiast motherboard. Do you want a to run an overclocked 5950X with 64GB of RAM? Pair it with an Nvidia RTX 3090 or AMD 6900XT? Go right ahead.

As the name suggests, the biggest marketing talking point of the board is its support for the Razer Chroma RGB ecosystem. On the surface it’s just another flavour of RGB, but Razer does go beyond a typical implementation to include a greater range of effects while offering extensive customization with the Chroma Studio app. There’s also game integration, support for many external apps and Amazon Alexa integration. So, if you want to RGB your RGB, the chances are Chroma can do it. 

It really does look lovely, though as is often the case, a graphics card will cover a big chunk of the PCH heatsink. But, with all the other RGB devices you’re likely to add to such a system, that’s not really a problem.

The other notable differentiator is the inclusion of Killer E3100 2.5G and AX1650 WiFi with DoubleShot support. The gaming community tends to have a bit of a love/hate relationship with Killer. A lot of it goes back to the clunky Manager app. The days of LAN gaming are behind us, but that would have been a great environment to test. Throw in all the leeching and you’d really stress your networking! The network outside of your PC and home is the true bottleneck these days. In the course of our testing the Killer networking operated exactly as it should. 

The board is dominated by the subtle large black heatsinks featuring that stylish RGB lighting. RGB and ARGB fan headers can be seen just above the RAM slots, which themselves support up to 128GB at DDR4-4733. While that speed sounds good, actually you’re better off at 1:1 with the Infinity Fabric up to around DDR4-4000.

The primary M.2 slot is covered by its own heatsink. Thankfully this means you’re not required to remove the entire heatsink to connect a drive. First world reviewer problem perhaps? You also get eight SATA ports, something that’s becoming rarer these days even on high end boards as the switch the NVMe and M.2 accelerates. File hoarders take note. 

Also noteworthy is the inclusion of a USB Type-C case header. At the bottom of the board you’ll find another set or RGB headers and power and reset buttons. This is becoming more and more common to many cases.

The VRM used on the ASRock Taichi Razer Edition is among the best you’ll find on a B550 board. A 16-phase setup with 50a mosfets will easily power an overclocked 5950X. Twin 8pin EPS connectors provide the juice.

The I/O is fairly typical of a premium board. We like that there are eight USB ports on the back panel, though curiously they aren’t color coded. There are two labelled USB 3.2 Gen 2 ports, four Gen 1 ports and a pair of 2.0 ports. (Side note: We really hope the USB naming mess is fixed with USB 4. It really is BS.) We’re not displeased to see HDMI 2.1 and DP 1.4a ports. If you’re a gamer with a discrete card it won’t matter, but a Zen 3 based APU with a monolithic die might surprise with its performance, so it’s better to have them than not. 

As we often say when testing motherboards, the performance between them is usually within a margin of error or a couple of percent, and that’s the case again here. The B550 Taichi Razer Edition proved to be quite strong under most of the multi-threaded tests with our AMD Ryzen 9 5950X, often leading the other X570 and B550 boards we’ve tested to date. The board was more mid pack in the gaming tests, however.

Now that we’re familiar with the characteristics of our 5950X sample, it’s a lot easier to know what to expect from a board. We were again able to hit 4.5GHz with a 1.26v load but its gets very hot running 32 threads at 100%! Well into the 80’s in fact, so we call it quits there. The board has more headroom but the CPU just gets too hot even with a 360mm AIO. The VRM didn’t break a sweat with a 55°C maximum temperature recorded. That’s a long way from any danger or concern.

As far as memory overclocking goes, DDR4-4000 or 2,000 MHz with a 1:1 Infinity Fabric clock was easy. We have achieved higher on the Asus Crosshair VIII Dark Hero but that’s a top tier board. A good set of DDR4-3600 with your choice of capacity at a 1:1 Infinity Fabric ratio will sit right in the sweet spot.

The ASRock B550 Taichi Razer Edition is arguably right at the top of the B550 tree. Whether a B550 board costing as much as the Taichi Razer Edition is right for you may come down to how important RGB is to you and how many M.2 drives you plan to run. Many X570 boards are available at this price with three M.2 slots and feature more PCIe 4.0 general purpose lanes. There is an ASRock X570 Taichi Razer Edition, and that is some $100 more. Is it worth the extra? We’d say probably not unless your needs are specific. 

The ASRock B550 Taichi Razer Edition can overclock as far as your cooling will allow, it can run fast memory, and it’s got Killer 2.5G LAN and WiFi 6 along with decent audio. It really does look great and will strongly appeal to RGB lovers. Ask yourself if you need extra M.2 slots or things like 10G LAN. If not, then this is a board that’s definitely worth considering. Even if for the aesthetic alone…

A benchmark of AMD’s Ryzen 7 5800H has allegedly leaked on Geekbench, with some positive results. The benchmark was posted to Twitter by well-known leaker ASPIK (@TUM_APISAK) and can be compared against its predecessor, the Ryzen 7 4800H, and also the existing Comet Lake-H CPUs from Intel.

As ever, treat any leak with caution, but if this benchmark score is a genuine reflection of performance then the AMD Ryzen 7 5800H would be giving some serious competition to Intel’s Tiger Lake-H family of high-end processors that will be launching in early 2021.

Trouble for Tiger Lake?

AMD’s Ryzen 7 5800H CPU was featured in the upcoming Acer Nitro AN515-45 notebook which comes configured with 16GB of DDR4 memory. In regards to performance, the 5800H features eight cores and 16 threads and was running at a base clock of 3.20 GHz and a boost speed of 4.45 GHz, with the average across all cores being a stable 4.40 GHz.

The 5800H hit a score of 1,475 in Geekbench 5 single-core, and managed 7,630 for multi-core. That’s a staggering 35% faster than the Ryzen 7 4800H in single-core, something that should greatly concern Intel with the impending release of its Tiger Lake CPUs. Not only that, but the ability to boost at 4.45 GHz would mean the Ryzen 7 5800H could defeat the Intel Core i9-10900K by 5% in single-core performance.

On paper, the difference in performance seems insignificant, but it’s worth bearing in mind that the Ryzen 7 5800H is a mobile chip iwithin a notebook, whilst the Intel Core i9-10900K is a full desktop CPU with a much higher power limit (35W vs 250W).

Things have certainly been heating up in the competitive world of CPUs this year, and with the Ryzen 5000 Mobility CPUs expected to be announced at CES 2021, competition is only going to get fiercer. Whether your preference lays with AMD or Intel, there are exciting things to come in the next few months.

Intel is taking Optane to the next level with its SSD P5800X, which it bills as the fastest SSD in the world. The next-gen Optane SSD leverages the PCIe 4.0 interface, second-gen Optane media, and a new SSD controller to deliver truly astounding performance and endurance specifications.

Compared to the first-gen Optane DC P4800X, Intel says the 5800X offers an impressive 3X performance improvement in random read/write workloads, peaking at 1.5 million 4K random IOPS, and 3X more sequential performance at a peak of 7.2 GB/s. 

The drive also delivers up to 4.6 million IOPS in random 512B workloads (useful for certain types of caching workloads) and up to 1.8 million IOPS in mixed workloads. 

Additionally, the drive offers up to 67% more write endurance than the first-gen P4800X, and now peaks at a staggering 100 drive writes per day (DWPD). That also makes it the most endurant SSD on the market. For perspective, there are very few ‘regular’ NAND-based SSDs that offer even 10DWPD of endurance, meaning the P5800X is truly in a class of its own. 

The drive leverages Intel’s second-gen Optane media, which comes with two decks/layers instead of one, and a new PCIe 4.0 x4 SSD controller that allows the company to finally break the shackles of the PCIe 3.0 interface. 

Intel isn’t sharing the finer-grained technical specifications behind the P5800X’s new PCIe 4.0 SSD controller. Still, we know that the drives will come in the tried-and-true U.2 form factor in capacities of 400GB, 800GB, 1.6TB, and 3.2TB, marking a healthy capacity increase that is more than twice the storage of the previous-gen DC P4800X. 

Just like any other Optane-powered drive, the exotic underlying 3D XPoint technology provides far more speed and endurance than any NAND-based SSD, delivering a level of performance that lands between flash SSDs and system memory. The drives are also designed to offer a superior quality of service (QoS) over competing NAND-based SSDs, meaning performance consistency is a key focus area. Intel’s first-gen P4800X had already established itself as the most consistent SSD on the market, but Intel says the P5800X extends that with up to a 40% better QoS in 4K random read workloads. 

Perhaps one of the most valuable aspects of an Optane SSD is its ability to deliver tremendously consistent performance even while it is under heavy load. As outlined in the slide above, the P5800X reaches new heights with more than 8 GB/s of throughput while under heavy load, easily outclassing all NAND-based SSDs on this logarithmic scale and also notching a substantial lead over its previous-gen counterpart. Intel claims the drive can deliver less than 66us of 99.999th percentile latency, which is a massive step forward over the first-gen DC P4800X, not to mention standard SSDs. 

The Optane Memory P5800X can even outstrip what we typically classify as the throughput limit of the PCIe 4.0 bus. That comes as a byproduct of the drive’s ability to leverage PCIe’s bi-directional interface. By saturating both sides of the interface at full line-rate read and write performance, the drive delivers even higher performance during mixed workloads. We can see that ability in the 70/30 read/write workload outlined above where the SSD peaked at 2 million IOPS. This allows the drive to deliver 3.7x more performance than the previous-gen model in mixed random workloads. 

Intel hasn’t released pricing information yet, but the company says the drives are already shipping to OEM companies. Naturally, as a standards-based NVMe block-addressable device, these SSDs will also be compatible with AMD’s EPYC processors. 

The Surface Pro 8 has already leaked plenty of times, but now Roland Quandt has revealed the memory configurations of Microsoft’s upcoming 2-in-1. So far, we know that the design of the Surface Pro 8 will remain unchanged from its predecessor, which featured an outdated design itself. The Redmond-based company will switch the Surface Pro to Intel’s Tiger Lake-U platform though, which should bring a healthy graphics boost for the Surface Pro 8 compared to the Surface Pro 7.

According to Quandt, LTE connectivity will return as of the Surface Pro 8, which is unsurprising. We suspect that Microsoft will restrict LTE connectivity to the Business version of the Surface Pro 8 though, as it does with many of its Surface products. Hence, there will be no 5G connectivity for the Surface Pro in 2021. The Core i5-1135G7 and Core i7-1165G7 should be available for business and regular customers, though.

Additionally, Quandt claims that the Surface Pro 8 will be configurable with up to 32 GB of RAM and a 1 TB SSD. We imagine that the Surface Pro 8 will be just as improbable to upgrade as the Surface Pro 7 though, if it follows the same design. It would be good if the Surface Pro 8 features a replaceable SSD like the Surface Laptop 3 and Surface Pro X do, but we doubt that will be the case.

The Surface Pro 8 should debut in early 2021 alongside the Surface Laptop 4. Microsoft will offer the Surface Pro 8 in the following configurations, according to Quandt:

Surface Pro 8 LTE i5 8 GB/128 GB

Surface Pro 8 LTE i5 8 GB/256 GB

Surface Pro 8 LTE i5 16 GB/256 GB

Surface Pro 8 i5 8 GB/128 GB

Surface Pro 8 i5 8 GB/256 GB

Surface Pro 8 i5 16 GB/256 GB

Surface Pro 8 i7 16 GB/256 GB

Surface Pro 8 i7 16 GB/512 GB

Surface Pro 8 i7 16 GB/1 TB

Surface Pro 8 i7 32 GB/1 TB

While we all wish we could afford an Nvidia RTX 3090 or AMD RX 6900 XT, us rank and file PC gamers are more interested in more affordable and worldly options such as the RTX 3060 Ti. We’ve already checked out the Nvidia RTX 3060 Ti Founders Edition card here at PCG and were impressed by its generational performance uplift and general characteristics. 

There’s always room for improvement though, and we’ve come to expect that from the premium tier partner cards such as in the MSI RTX 3060 Ti Gaming X Trio card we have here. It’s an absolute monster of a card. Is it worth the premium over entry level cards?

The RTX 3060 Ti shares a lot in common with the Nvidia RTX 3070. It uses a cut down version of the GA104 GPU with 4,864 activated cores versus the 5,888 of the RTX 3070. They both share 8GB of 14Gbps memory and of course you get all the Ampere features including improved second generation ray tracing support, DLSS and more. Check out our launch article for a closer look at the features and tech of the RTX 30-series’ architecture.

Moving on to the MSI GeForce RTX 3060 Ti Gaming X Trio itself—say that 10 times fast—the first thing which becomes apparent is its sheer size. It would make a pretty good murder weapon (that’s where your mind goes first?! – Ed.). 

The Tri Frozr cooler is the same that’s used to cool the RTX 3080 Gaming X version with its much higher TDP, so we know it’s more than capable of cooling this less demanding GA104 GPU. It’s a triple slot card that’s 323mm x 140mm x 56mm so you’ll need to make sure it will fit in your case. You get a neat RGB light bar along the edge of the card, along with a lit MSI logo and another strip running across the fan shroud. It’s a good-looking card. At $489 it’s also quite a step up over the cheaper Founders Edition and reference cards, but it’s not out of line with other top tier cards like the Strix or Aorus from Asus and Gigabyte respectively.

The MSI RTX 3060 Ti, however, features a very large and surprisingly sparse PCB with an 8-phase VRM. There’s really no need for a Lightning-esque VRM with this fourth-tier Ampere GPU. It features a boost clock of 1,830MHz but the official clocks are all but irrelevant unless you’re mining with it or have a case with poor airflow. The TDP of the card has been increased to a rather lofty 240W compared to the 200W of an RTX 3060 Ti Founders Edition. The proprietary 12-pin power connector that you’ll find on the Nvidia-built cards is omitted in favour of a pair of standard 8-pin PCIe connectors. 

This might be a good thing for the obsessive cable managers out there who dislike the central placement of the 12 pin connector. The card features the now typical set of display outputs including three DisplayPort 1.4a and a single HDMI 2.1 port. Type-C USB has been omitted from the Ampere generation cos who needs VirtualLink really.

At the end of the day, however, it’s the performance that matters most and the RTX 3060 Ti makes a massive jump over not only the RTX 2060, but it beats out the RTX 2070 and RTX 2080, and the Super models at that. In fact only the RTX 2080 Ti is faster out of all the Turing generation GeForce cards. The MSI Gaming X makes a mockery of the advertised 1,830MHz boost clock. After running for over 10 minutes at full load the card settled into a consistent boost clock of between 1,930 to 1,970MHz, albeit on an open test bench.

MSI RTX 3060 Ti Gaming X Trio benchmarks

MSI’s Gaming X Trio options are often the quietest models of graphics cards you can buy. The RTX 3060 Ti flavour continues that trend. It simply cannot be heard over the sounds of the other fans and AIO pump in our test system and it manages these low noise levels without compromising on performance one jot. We know this cooler can easily tame a much hotter RTX 3080 GPU so when it’s tasked with cooling a GA104 GPU, which demands 100W less power, well, it’s a walk in the park really. If silence is high on your list of priorities, it will be hard to do better.

A quick overclocking test resulted in a consistent boost clock of between 2,020 and 2,050MHz. That’s not bad, but given the way modern GPUs boost so effectively within their power budgets, a performance gain of around 5% doesn’t really seem worth it. We were also able to hit 2,000MHz on the memory with ease (or 16Gbps up from the default 14Gbps). More is likely possible if you put the time into it, though how relevant that will really be in terms of extra performance.

If you game at 1440p and you’re looking at a card to play at high settings then the RTX 3060 Ti is a great option. Something like Cyberpunk 2077 at maximum settings and 4K resolution will be beyond it, but then that’s beyond pretty much every card on the market. But at least you can play with ray tracing and DLSS enabled. These are things that are currently unavailable to AMD users. 

When you look at a stunning game like Cyberpunk 2077, you’re left thinking it’s the ‘killer app’ that pushes ray tracing over a threshold to where it will really become a core PC graphics technology. Ray tracing isn’t just a cool bonus tech anymore. Adoption will only increase from here. Though, let’s see how AMD and the new consoles look with it once its available.

Now that that piece of commentary is out of the way, let’s get back to the MSI Gaming X Trio. Its much bigger cooler means that thermals and noise levels are hugely improved over the entry level RTX 3060 Ti cards, but then it costs an extra $70 or more too. There’s also the 40W increase in power consumption to consider. Whether the extra money is worth it is up to you. At this position in the market this kind of price premium isn’t as drastic compared to hundreds of dollars extra for premium RTX 3080’s but if you’re on a tight budget, it can make a big difference.

There’s also the thorny question of availability. Nvidia’s RTX 30-series, and to an even greater extent AMD’s RX 6000 series cards, are suffering from poor or non existent stock levels. Waiting months for a GPU to come into stock is unacceptable. So, which card you buy may come down to what’s in stock at the split second you happen to be on the page. 

It won’t last forever though, and the MSI RTX 3060 Ti Gaming X is a great, chunky card and one that’s aimed at users who value quiet running and premium mid-range performance. If you want a nice splash of RGB on top of a card that leans towards low noise ahead of a few degrees of temperature then the MSI Gaming X Trio is one you’ll want to consider. Y’know, when you have the luxury of options…

The first consumer DDR5 memory kits will arrive in 2021, presumably just in time for Intel Alder Lake and AMD’s Zen 4.

This year will come to a close without any consumer platforms supporting DDR5 memory modules, but next year will likely see at least two of them—Intel’s Alder Lake and AMD’s Zen 4 architectures. In preparation, Team Group says it has developed consumer-grade DDR5 memory, and claims it is the first to enter the validation phase with key motherboard partners.

They include all the big names, those being ASRock, Asus, Gigabyte, and MSI. Team Group says it has been working the research and development divisions of each company to validate its DDR5 memory to ensure it works as it should, and tweak the parameters as necessary.

“The successful completion of the validation phase will confirm that the frequencies of standard DDR5 products surpass those of overclocked DDR4 products, and will represent another big step forward in the evolution of computer memory,” Team Group says.

Team Group is referring to factory overclocked modules, because if it took into consideration extreme overclocking results obtained with liquid nitrogen cooling, it would have to surpass 7GHz, as an overclocker recently obtained with a kit of Crucial Ballistix DDR4 RAM. That’s not in the cards.

Extreme overclocks aside, DDR5 RAM will bring faster memory speeds in general to PCs, picking up where higher end DDR4 memory kits leave off. More specifically, Team Group is getting the DDR5 party started with a single 16GB DDR5 memory module that runs at 4,800MHz at 1.1V. In a previous release, Team Group suggest it could go as high as 5,200MHz.

“It is expected that consumers will not need to enter BIOS to enable the overclocking function as the DDR5 generation begins. After installing DDR5 memory, consumers can directly boot up their computers and experience the high performance without the overclocking step of DDR4,” Team Group says.

Earlier this month, Team Group said its first consumer DDR5 memory product is expected to be available in the third quarter of next year. The company also noted it was coordinating the launch with Intel and AMD, which suggests that both will have DDR5 platforms ready around the same time.

Coming in at the middle of the Zephyrus gaming laptop line-up, the Asus ROG Zephyrus M15 GU502 packs an Intel Core i7 10750H and an Nvidia GTX 1660 Ti into its svelte form. This nifty combo affords some steady gaming performance, especially when you consider the relatively reasonable price point of $1,299 (£1,000). But for this price, I’d still expect a little more from its other features than what I actually got.

I will admit I genuinely gasped when taking the M15 from its packaging. As the box opens, the laptop pushes out to meet you like some fancy, overpriced pop-up book. It doesn’t have any bearing on performance, but it’s a nice bit of flair on Asus’ part.

The chassis itself is gorgeously understated. It’s slimmer and lighter than a lot of gaming laptops around. The half brushed aluminium, half ‘prismatic dot design’ finish even gives the M15 a classy look and feel, and means it’s super grease-mark resistant. Being milled from solid aluminium makes for a super robust laptop—though the screen does still feel a tiny bit wobbly. 

The major drawback to a metal chassis is that, while it may be sturdy, it sure does conduct heat. So, when you accidentally cover the exhausts for more than a minute, it’s very difficult to get it back down to an acceptable temperature, until you exit the game and leave it to rest for a bit. Considering the CPU temps tend to reach an elevated 96-100°C (~205°F), with the GPU not far behind at 83°C (181°F), that translates to some pretty unwieldy external temps when the whole thing’s encased in metal.

In the machine’s vain attempts to cool down, it does get a little screechy as well. At a maximum of around 37 decibels, it probably doesn’t pass our Jacob’s ‘train carriage test,’ i.e. you might feel embarrassed to use it on a train’s quiet coach. However, if you just want to drown out the noise of the cooling system, the downward facing speakers are pretty decent. Though I’ve heard louder, there’s no real issues with sound distortion at high volume, and as long as you have the machine on a hard surface, the resonation of sound through your table does help to enrich the experience.

There are also settings to nullify the noise of the rather expressive cooling system within Asus’ included ‘Armory Crate’ software, as well as settings for tweaking performance. The software includes an array of pre-set visual profiles, including an ‘eyecare’ setting that limits the blue-light beaming into your eyeholes, as well as configurable profiles that allow you to save personal system and app settings for different situations.

There’s some nice customisation here, and connectivity isn’t bad at all. All the ports and connections are placed on the side for easy access, and to avoid blocking the screen as it opens. That satisfies my preference, though I’m aware it’s subjective—it might get in the way for some users. N.B. Although you can easily remove the power cable, I suggest you don’t take it out during gameplay. Not only are there some serious performance throttling issues when it runs off the battery, it’ll also only last about an hour and ten minutes without a power lead. Not ideal.

System performance

The lack of a webcam is sure to be a downside for some, but at least it helps keep the price down. Also contributing to the affordability factor is the mediocre keyboard. It’s a little stiff on the keypress, and the backlights are not per-key programmable, so the single-zone lighting effects are very limited. Same goes for other models with the ‘prism’ design—perhaps another attempt to shy away from the edgy gamer aesthetic and add more corporate uniformity to the design. 

Considering it is still a gaming laptop, one of my biggest issues with this machine is the paucity of storage capacity. With only a single 512GB SSD at my disposal I had to install then uninstall each individual game for benchmarking, which was rather tedious. So don’t expect to have your whole gaming library at the ready.

Drawbacks considered, what matters most is that the M15 can bash out some solid gaming performance for the price, but don’t expect to be turning all the settings up to full whack.

That said, the GU502L manages to tie with, and in some cases supersede, the similarly specced (yet more expensive) XPG Xenia 15. It charts a smooth 78 FPS average on Far Cry New Dawn at the screen’s native 1080p on ultra settings, and runs Shadow of the Tomb Raider at a very consistent 70 FPS, with settings at highest, and shadows on ultra.

Gaming performance

With the most graphically intensive games, such as Metro Exodus, the Asus M15 only managed to push 43 FPS average—that’s with super special hair and PhysX switched off. And though that’s not the worst score we’ve seen, it might need a tune down to get the most out of it.

When it comes to more CPU heavy games like Total War: Warhammer 2, it only manages to push a 50 FPS average, which isn’t too impressive in the grand scheme of things. Still, it backs up those frames with an above average 144Hz refresh rate display, which gives you a better chance of your reflexes actually meaning something, as long as the frame rates actually allow it.

Sadly, it’s easy to see where corners have been cut with this machine, meaning there are more than a few trade-offs if you want portability, steady FPS, and a speedy display for that price. Overall, the M15 seems best for those who only play a select few games at any one time, and are looking to ditch all the fancy greebles that some of the more expensive gaming laptops have to offer. 

It’s totally possible to get some good old fashioned 1080p gaming done at mid-high settings, and it’ll even do so in impeccable, corporate style—so you can take it to work without looking too conspicuous. But don’t expect much more from it performance wise, bar some nice productivity scores.

Still, it is a sturdy, slim, stylish machine, with punchy speakers enough horsepower to get the more speed-oriented gamers off the ground. And it proves that it’s possible to design a well performing gaming laptop, without it looking like it was birthed by a transformer. Though, some more exhaust ports wouldn’t go amiss.

Windows 10 has witnessed a taskbar change which, if it stays in place – at the moment it’s just in testing – could have an impact on those who haven’t paid for their copy of the operating system, and haven’t yet activated it.

The plan is to take some of the options currently available when you right-click the taskbar, and move them into the taskbar category under the Settings app (in personalization).

As Windows Latest reports, in fresh preview builds of Windows 10, Microsoft is experimenting with moving some options to the Settings app, including the ability to show or hide various buttons (such as Task View, or Windows Ink Workspace), plus options for the Search box in the taskbar (including the ability to hide that).

This would further centralize the various taskbar options under Settings, and perhaps hints at a future for Windows 10 where some of these options don’t appear when right-clicking the taskbar – although that’s far from certain, and it wouldn’t exactly be convenient either.

Activation stations

Another point to bear in mind is that it’ll be a bit more of a blow to the convenience of users who haven’t activated Windows 10, too, seeing as if the OS isn’t activated, then it’s not possible to access personalization options. That already forbids doing things like automatically hiding the taskbar, but it could potentially see the removal of more abilities for these users, like the aforementioned search box and button options.

Of course, it’s unlikely that you haven’t activated Windows 10, but there are some users out there who keep using Microsoft’s operating system in this manner.

In fact, it’s perfectly possible to keep on trucking with Windows 10 without needing to activate it, although you do forfeit any ability to personalize the desktop as mentioned, and there are other restrictions including a watermark being present at all times. Otherwise, the core functions of the OS run fine in this manner – although there’s no guarantee that this will remain the case, and it could change as soon as tomorrow (or more abilities could be stripped away).

In other words, inactivated users are still (and always will be) running at their own risk, although perhaps Microsoft is happy to have the added Windows 10 user numbers – at least for the time being – and maybe these folks are still spending elsewhere in the Microsoft ecosystem via Windows 10.