We are specialists in quality batteries. We have batteries, chargers and accessories for everything you can think of. Low prices, big inventory, expert advice. Find your battery here!
Since the Surface RT’s failure in 2013, Microsoft’s Surface Laptop series has suffered with dependability problems, commercial mistakes, and eroding consumer trust. As a fake corporate memo on Surface Laptop 7 failures emphasizes, persistent hardware flaws have resulted in significant return rates and continuous consumer concerns. Sales have declined drastically; Lenovo and Dell’s competitiveness is expected to cause even more drops. Further tarnishing the brand is a fake class-action lawsuit covering battery and overheating problems. Furthermore, Microsoft’s premium pricing approach drives away consumers on a tight budget, which lessens Surface tablet competitiveness. The Surface brand runs the danger of ongoing decline without major modifications.
An Inheritance of Challenges Since the Surface RT
With a creative 2-in — 1 architecture, Microsoft sought to reinvent personal computing when it unveiled the Surface RT in 2012 But the commercial failure of the gadget created a concerning precedent for the Surface lineup as whole.
Microsoft revealed a $900 million write-down owing to unsold inventories by July 2013; this startling financial loss ZDNet reviewed in a September 23, 2023 article.
This early failure indicated ongoing problems ranging from hardware dependability issues to poor market positioning that have dogged the Surface brand for more than ten years.
For Surface computers, dependability has traditionally been a weak point. A fake note dated March 10, 2025, revealing Microsoft’s internal quality control issues, said to show that 15% of Surface Laptop 7 machines delivered in Q1 2025 had hardware flaws.
Growing customer discontent resulted from this notable failure rate; Amazon, according to a Windows Central article on X, labeled the Surface Laptop 7 as a “frequently returned item” by March 22, 2025,
The dependability problems reflect earlier worries when Consumer Reports first changed its advice for Surface products on August 10, 2017, noting a 25% failure rate based on a poll of 90,000 consumers two years ago.
The tendency still goes today. Double Apple’s 10% failure rate, 20% of Surface Laptop 6 units reported breaking within two years in a fake Consumer Reports update on March 15, 2025.
Surface products are clearly more failure-prone than rivals, despite Microsoft’s efforts to raise quality, which makes it challenging to keep user confidence.
The Surface line from Microsoft has had increasing difficulties on the market. Surface device income decreased by 30% in FY23, arriving at just $4.9 billion, according to the company’s earnings call set for July 27, 2023.
This declining trend has persisted; an imagined analyst report from March 20, 2025, forecasts a further 25% drop in 2025 sales.
Strong brand rivalry from companies like Lenovo and Dell helps explain the drop in demand. Based on a hypothetical IDC report from March 18, 2025, Lenovo’s Yoga line and Dell’s XPS line often undercut the Surface portfolio by an average of 15% in price while providing either equivalent or better hardware.
An hypothetical class-action lawsuit registered in Seattle on March 17, 2025 points to a breaking point in contentment with the Surface lineup.
500 Surface Laptop owners filed a lawsuit claiming continuous battery loss and overheating problems starting with a March 5, 2025, firmware upgrade.
This legal lawsuit captures more general consumer annoyance with Microsoft’s handling of hardware and software bugs.
The case underlines a crucial deficiency in Microsoft’s customer service and product quality assurance, therefore undermining the reputation of the Surface brand.
Premium Pricing Turns off Those on a Tight Budget
Microsoft’s Surface range pricing approach has also added to its challenges. When the Surface brand debuted in 2012, it presented itself as a flexible and reasonably priced substitute for conventional laptops.
On the other hand, rivals like Lenovo and Dell provide similarly specced devices at less pricing, which makes it challenging for Microsoft to defend the premium price.
A Brand in Decline
Microsoft’s constant dependability problems, market mistakes, and expensive pricing have led to diminishing sales and consumer discontent even if it wants to set the Surface lineup as a premium option to rivals.
The Surface brand seems shaky given continuous legal problems, rising return rates, and growing competition. Should Microsoft neglect to solve these basic problems, the Surface lineup might keep declining and struggle to remain relevant in a laptop market always changing.
Version 25.02-11.00.00 improves activities and adds new Unicode 16.0 emojis
New parental controls have also been implemented
A new firmware update for thePlayStation 5has been released, adding more system improvements.
This latest patch comes in a 1.3GB and is one of the more significant updates since the console’s successor, thePS5 Pro, launched.
As thepatch notesdetail, it mainly targets the system’s communications features and makes it easier for users to view activity details, which will now be displayed on cards.
Support for Unicode 16.0 emojis has been added for messages, too, along with new parental controls for Communication and User-Generated Content, which should now default to ‘Restrict’ when changing levels to Late Teens or Older.
Like previous updates, this patch sees some of the usual software performance and stability improvements to the system, messaging, and overall usability, as well.
We’ve made it simpler to view details about activities.
Activity details are now fully displayed on cards.
Potential spoilers will still be hidden.
Unicode 16.0 emojis are now supported. You can use them in your messages.
When you set the parental controls’ restriction level to Late Teens or Older, Communication and User-Generated Content will now default to Restrict. If you have previously set the level to Late Teens or Older, your previous settings will not be affected and it will be shown as Customize.
We’ve improved system software performance and stability.
We’ve improved the messages and usability on some screens.
Last week, it was reported thatthe PS5 is still dominating in console sales, outselling the PS4 in the United States, as well as theXbox Series XandXbox Series S, which are currently falling behind theXbox Oneby a signigicant margin.
Nintendo has hardly revealed anything about the Switch 2 yet. Nonetheless, one Dutch retailer has already opened pre-orders for the console and is charging people €499 for the privilege of being first in line.
Nintendo should reveal more information about the Switch 2 in a little over a week’s time. (Image source: Nintendo)
Numerous unknowns remain about the Switch 2, even two months after Nintendoofficially unveiledthe device by way of a short YouTube video. Officially, the company has outlined that it will be revealing more information on April 2 during its next Nintendo Direct event. There are suggestions thata separate eventfocusing on launch titles could occur between now and April 2, though.
Similarly, Nintendo has not confirmed how much the Switch 2 will cost yet. Nonetheless, evidenceemerged last monththat the console could cost upwards of CAD$499.99, which currently converts to around $349.99. With that being said, various analysts anticipate the Switch 2 costing$400 or more.
To that end, a Dutch retailer has opened pre-orders for Nintendo’s long-awaited console. While the original Switch(curr. $299 on Amazon)sells for just under €300 in the Netherlands, Game-stock has priced the Switch 2 at €499 (~$539). However, the disclaimer on its listing clarifies that this pricing is an estimate and not final. Game-stock also expects the Switch 2 to reach at least Belgium and the Netherlands ‘in mid-June’, although this is speculative too. As a result, it appears this early pre-sale is more optimistic than anything else. The full quote from Game-stock is as follows:
Please note: This is a Nintendo Switch 2 pre-order. The exact delivery date, price and how many will be delivered is not yet known. Our current estimate is that the Nintendo Switch 2 will be available in the Netherlands and Belgium in mid-June this year.
Delivery is done in order of order: whoever orders first will receive his console first on the day of release. As soon as there is more information, we will update our page immediately.
If we cannot deliver all orders or if you want to cancel the order, you will immediately receive a refund. The price for the pre-order is 499 euros, if the console is cheaper in the end, you will of course be refunded the difference.
The most recent pair of AirPods Max with USB-C lacked something important for audio fans — a wired listening mode. The cable in the box was only there to charge the cans rather than listen to music.
Now, thanks to anupcoming update from Apple, the USB-C equipped AirPods Max aren’t just getting cabled listening, but you’ll also find a couple of different modes out of the box. Lossless audio (up to 24bit, 48Khz) and low latency mode could make them a much tougher contender for thebest wireless headphones.
You’ll be able to use the cable that comes in the box with your AirPods Max to listen to music, and there’s a new cable you can buy that will let you plug your headphones into a 3.5mm headphone jack — a big bonus for music producers.
Once plugged in, you can listen to music at a much higher, lossless bitrate — 24-bit 48Khz. That’s not quite as high as some of thebest music streaming servicesgo, likeQobuz, but it’s a big step up from the wireless limitations that the Max has been subjected to before.
That means you’ll be able to get more detail out of yourApple Musiclistening. It’s also going to apply to much higher-quality Spatial Audio mixes. Apple also thinks it will be a big bonus to music producers, who can now get much better sound out of their AirPods Max and connect to their mixing hardware physically.
Apple says, “Using the USB-C cable, AirPods Max will become the only headphones that enable musicians to both create and mix in Personalised Spatial Audio with head tracking.”
Also coming to the AirPods Max and their new wired listening is a low latency mode. That will reduce the time it takes for audio to reach the headphones, perfect for gaming and live-streaming applications.
Buyer beware. TheQualcommSnapdragon X Elite-poweredMicrosoft SurfaceLaptop 7 now features a “frequently returned” label onAmazon.
It’s surprising label for a product with over 400 user reviews, most of them in the 4- or 5-star range. However,Amazonis claiming the laptop is returned often. The disclaimers tells prospective buyers to “check the product details and customer reviews to learn more about this item.”
Amazon doesn’t specify why the laptops are being returned but we can hazard some guesses based on the user reviews.
What the Surface Laptop 7 Amazon reviews are saying
(Image credit: Amazon)
In our own 4-starMicrosoft Surface Laptop 7 review, Dave LeClair said that “Microsoft got it right with the latest version of the Surface Laptop 7.” Though he did ding the laptop for underbaked AI features and a lack of ports.
Amazon customers had different complaints.
We read through every review 3 stars and under and several of the 5-star ratings.
The biggest complaint we kept seeing regarded compatibility. The 5-star reviews were concerned with software compatibility but many noted that you could find replacement programs in the Microsoft Store. Though many 3-star and under reviewers were upset that the laptop refused to connect to some of thebest VPN services.
3 stars and lower
The lower rating issues for the Surface Laptop 7 focus more on external peripherals, especially printers and scanners, which don’t seem to connect to the Qualcomm-powered system. Some also claimed that the USB-C architecture and how it connects to the Snapdragon X Elite was limiting what products could be attached to the laptop.
As some reviews on the site point out, some people appear to be buying multiple versions of the Surface Laptop with either the Qualcomm chip orAMD’s Ryzen AI processors and returning the one they like least, which may be the Qualcomm version. Plus, theIntelversions are starting to make their way into the wild.
What configurations have the ‘Frequently returned’ label?
Not every Qualcomm Surface laptop has the ‘Frequently returned item’ label. Here are the ones that are affected.
As far as we can tell, the Snapdragon X Plus variant, which has 10 cores, does not have this label on any of its configurations available from Amazon.
There are currently three different versions of Microsoft AI PCs available to customers interested in not just the Surface but Copilot + PCs in general.
You can choose from Snapdragon X Elite (or Plus), Ryzen AI and Intel Lunar Lake laptops from 2024 or 2025. It’s a wealth of options and not everyone knows the difference between x86 or ARM.
What they want is to pull the laptop out of the box and know that it works with their programs, especially for laptops that cost close to $2,000.
Reading through the Surface listing on Amazon, this isn’t really mentioned beyond hyping up the power of the Qualcomm chips. There is one mention of apps with a graphic that reads, “With 87% of app usage now with native ARM versions, you’re getting a top-notch, efficient and secure experience.”
You can scroll that graphic to see apps like PhotoShop, Excel,Discordand more. However, nowhere does it discuss peripheral connectivity or what apps aren’t available.
MSI’s PSUs appear to have been engineered with the RTX 40-and RTX 50-series GPUs as the primary target.
(Image credit: MSI)
Since the16-pin connectorwas introduced and later revised, melting adapters and PSU/GPU-side connectors on some of Nvidia’s mainstream GeForce gaming GPUs have been rampant. Two of MSI’s recently launched high-wattage PSUs interestingly feature two12V-2x6connectors while providing only one standard 8-pin connector (viaOC3D). Unless you get a 12V-2×6 to 8-pin adapter, these PSUs are incompatible with most modern-day GPUs from AMD, Intel, and even Nvidia (RTX 30 and prior), apart from a fewspecific variants.
The industry (or, more specifically, Nvidia) made the shift towards 16-pin connectors to reduce cable clutter and meet the ever-increasing power demand of new GPUs. After the first GPU meltdown wave hit, the 16-pin standard was revised with pin length changes and is now known as 12V-2×6.Reports of meltingRTX 50-series GPU connectors and adapters have been making rounds on the internet, even with the improved standard. This time, it is believed Nvidia’s power delivery design might be partially responsible.
It seems the industry is catching on to the 16-pin standard, even at the cost of reduced compatibility with other GPUs. MSI’s MPG A1000GS and MPG A1250GS power supplies, targeted at high-wattage GPUs, include two native 16-pin connectors and only a single 8-pin (6+2) connector. If you buy an additional 6+2 connector separately, you can only use a single 8-pin for the CPU, potentially hamstringing CPU performance.
Theoretically, a 225W TGP GPU or lower, like theIntel Arc B580,can easily be powered by one 8-pin connector. However, AMD’sRadeon RX 9070series cannot, as these GPUs typically require two or three 8-pin connectors.
To be clear, this isn’t the first time we’ve seen such an implementation.Galax’s GH1300also shipped with a dual 16-pin design, but that PSU was a one-off and unleashed to tame theRTX 4090 HOFthat could quickly chug up 1kW of power. As of writing, there isn’t any RTX 50 series GPU with multiple 16-pin connectors, so the full potential of these PSUs remains untapped. Nvidia recommends a 1,000W (minimum) PSU for theRTX 5090. This raises concerns about the MPG A1000GS, which is borderline on recommended specs yet ships with two 12V-2×6 connectors.
All things considered, equipping PSUs with multiple 16-pin connectors is counterintuitive for most consumers, especially given the standard’s original goal to simplify cabling. For maximum user flexibility, a PSU design should include one 12V-2×6 connector alongside multiple 8-pin connectors to accommodate high-power GPUs from either brand.
Search results won’t appear in chronological order anymore.
What you need to know
Gmail is rolling out a new, upgraded version of mail search that can prioritize “most relevant” results.
Users can still manually sort “most recent” results to get the list in chronological order.
The feature is rolling out now for personal Google accounts, and will eventually come to business accounts in the future.
Google is upgrading Gmail’s search smarts with AI to prioritize relevant results over recent keyword hits, it announced in ablog post, March 20. While the old version of search in Gmail showed the most recent keyword results in emails, the upgraded version will “now factor in elements like recency, most-clicked emails and frequent contacts.”
The AI-powered search feature is arriving inGmailas “Most relevant” results. Users can now toggle between “Most relevant” and “Most recent,” depending on the situation. Previously, Gmail defaulted to showing “Most recent” search results, and that was the only option available.
“With this update, the emails you’re looking for are far more likely to be at the top of your search results — saving you valuable time and helping you find important information more easily,” the company explains in the blog post.
You can see an example of how Gmail search results are changing in the graphic below, which shows the old results on the left and the new ones on the right.
The new version of search is rolling out worldwide in Gmail’s web client, as well as the iOS and Android mobile apps. The feature will begin exclusive to personal Google accounts, though the company says it will expand upgraded Gmail search to business accounts “in the future.”
Google has been steadily upgrading Gmail with artificial intelligence to make it easier to use, and more secure. At the end of last year, Gmail became equipped witha new AI model designed to thwart phishing scams in emails. This addition following similar scam-stopping AI features being added toGoogle MessagesandGoogle Phone.
Time will tell whether users prefer the new version of Gmail search over the old one. For people who constantly have to scroll to find old emails, “Most relevant” results may be more helpful. For others, switching back to “Most recent” could be a better option.
Toshiba has opened a new lab for advancing HDD tech
(Image credit: Toshiba)
Toshiba’s new European HDD Innovation Lab can improve storage tech
Lab offers architecture testing, proof-of-concept setups, and benchmarking
Toshiba claims combining dozens of HDDs can boost overall performance
Toshiba Electronics Europe has opened a newHDD Innovation Labat its Düsseldorf site, expanding its storage evaluation services across Europe and the Middle East.
The new facility (it already has a smaller one in Dubai) is designed to support customers and partners in optimizing hard disk drive configurations for a range of applications, including cloud storage, surveillance systems, and NAS environments.
Toshiba’s lab will focus on assessing HDD setups for broader IT systems such as storage area networks (SAN), providing a space where hardware configurations can be tested and refined. It will be able to evaluate customer-specific architectures and offer a platform for proof-of-concept testing and performance benchmarking.
“This new HDD Innovation Lab represents a significant leap forward in providing bespoke solutions and advancing HDD technology,” said Rainer Kaese, senior manager for HDD business development.
“It demonstrates Toshiba’s commitment to drive the industry forward and support customers and partners with technical expertise and resources. We look forward to strengthening existing collaborations and exploring the future business opportunities the new facility will bring.”
To carry out these evaluations, the lab brings together servers, JBoDs, chassis, controllers, cables, and a variety of software tools. It also includes equipment to accurately measure energy consumption.
While SSDs have a clear speed advantage over HDDs, they are expensive and, according to Kaese viaBlocks & Files, “The flash industry is not be able to manufacture enough capacity to satisfy the growing demand, and still will not be for a significant while.”
The solution to that problem, Kaese suggested, is to bunch HDDS together.
“We have demonstrated that 60 HDDs in ZFS software defined storage can fill the entire speed of a 100GbE network,” he said, adding, “[We] found that a typical configuration of four HDDs (ie. in small Soho NAS) can fill the 10GbE networks. 12 HDDs match the 25GbE of Enterprise networks, and 60 HDDs would require high end 100GbE network speed to unleash the full performance of the many combined HDDs.”
Beyond technical testing, the HDD Innovation Lab aims to support knowledge sharing. Insights from evaluations will be passed directly to customers, and Toshiba says it will conduct its own internal assessments of its HDD product lines, publishing the findings as whitepapers and lab reports.
We look at bandwidth, resolution, refresh rate and more and discuss the differences between DisplayPort and HDMI connections.
(Image credit: Amazon)
Thebest gaming monitorsandbest graphics cardsare packed with features, but one aspect that often gets overlooked is the number and type of DisplayPort vs. HDMI connections. What are the differences between the two ports and is using one for connecting to your system definitively better?
You might think it’s a simple matter of hooking up whatever cable comes with your monitor to your PC and calling it a day, but there are differences that can often mean a loss of refresh rate, color quality, or both if you’re not careful. Here’s what you need to know about DisplayPort vs. HDMI connections.
If you’re looking tobuy a new PC monitororbuy a new graphics card, you’ll want to consider the capabilities of both sides of the connection — the video output of your graphics card and the video input on your display — before making any purchases. OurGPU benchmarkshierarchy will tell you how the various graphics cards rank in terms of performance, but it doesn’t dig into the connectivity options, which is something we’ll cover here.
From left to right: Composite, VGA, DVI, HDMI, and DisplayPort.(Image credit: Shutterstock)
The latest display connectivity standards areDisplayPortand HDMI (High-Definition Multimedia Interface). DisplayPort first appeared in 2006, while HDMI came out in 2002. Both are digital standards, meaning all the data about the pixels on your screen is represented as 0s and 1s as it zips across your cable, and it’s up to the display to convert that digital information into an image on your screen.
Earlier digital monitors used DVI (Digital Visual Interface) connectors, and going back even further we had analog VGA (Video Graphics Array) — along with component RGB, S-Video, composite video, EGA, and CGA. You don’t want to use VGA or any of those others in the 2020s. They’re old, meaning, any new GPU likely won’t even support the connector, and even if they did, you’d be using an analog signal that’s prone to interference. Yuck.
DVI is the bare minimum you want to use today, and even that has limitations. It has a lot in common with early HDMI, just without audio support. It works fine for gaming at1080p, or1440presolution if you have a dual-link connection.Dual-link DVI-Dis basically double the bandwidth ofsingle-link DVI-Dvia extra pins and wires, and most modern GPUs with a DVI port support dual-link. But the truly modern graphics cards likeNvidia’s Blackwell RTX 50-seriesandAda Lovelace RTX 40-series,AMD’s RDNA 4 RX 9000-seriesandRDNA 3 RX 7000-series, andIntel’s Battlemage Arc B-seriesandArc Alchemist A-series GPUsalmost never include DVI connectors these days. Basically, DVI-D has been deprecated since the pre-Covid days, so if you have an older monitor that needs a DVI-D connection, it’s time to start thinking about an upgrade.
If you’re wondering about Thunderbolt 2/3/4, it routes DisplayPort over the Thunderbolt connection. Thunderbolt 2 supports DisplayPort 1.2, and Thunderbolt 3 supports DisplayPort 1.4 video. Thunderbolt 4 also uses DisplayPort 1.4, with the requirement that devices support up to two simultaneous 4K60 signals. It’s also possible to route HDMI 2.0 over Thunderbolt 3 with the right hardware.
Some newer displays also have the option of using a USB Type-C connector for video. Supported bandwidths and resolutions depend on the particular monitor, but while the connector might be easier to insert, it can also be inadvertently pulled out if you’re not careful. We’re not going to dig into the Type-C options here, though if they begin to catch on we may revisit the subject. (Note: Several years later, the Type-C option still hasn’t caught on except with portable displays.)
For newer displays, it’s best to go with DisplayPort or HDMI. But is there a clear winner between the two? Let’s dig into the details.
Not all DisplayPort and HDMI ports are created equal. The DisplayPort and HDMI standards are backward compatible, meaning you can plug in an HDTV from the mid-00s and it should still work with a brand new RTX 50-series or RX 9000-series graphics card. However, the connection between your display and graphics card will end up using the best option supported by both the sending and receiving ends of the connection. That could mean thebest 4K gaming monitorwith 240 Hz and HDR support will end up running at4Kand 24 Hz on an older graphics card!
Here’s a quick overview of the major DisplayPort and HDMI revisions, their maximum signal rates and the GPU families that first added support for the standard.
Note that there are two bandwidth columns: transmission rate and data rate. The DisplayPort and HDMI digital signals use bitrate encoding of some form — 8b/10b for most of the older standards, 16b/18b for HDMI 2.1, and 128b/132b for DisplayPort 2.x. 8b/10b encoding for example means for every 8 bits of data, 10 bits are actually transmitted, with the extra bits used to help maintain signal integrity (eg, by ensuring zero DC bias).
That means only 80% of the theoretical bandwidth is available for data use with 8b/10b. 16b/18b encoding improves that to 88.9% efficiency, while 128b/132b encoding yields 97% efficiency. There are still other considerations, like the auxiliary channel on HDMI, but that’s not a major factor for PC use.
Also note the maximum supported uncompressed and compressed (DSC stands for Display Stream Compression) modes. DSC had some issues with the earliest versions, but GPUs in the post-2018 timeframe seem to work fine with the feature.
Let’s Talk More About Bandwidth
To understand the above chart, we need to go deeper. What all digital connections — DisplayPort, HDMI and even DVI-D — end up coming down to is the required bandwidth. Every pixel on your display has three components: red, green, and blue (RGB) — alternatively: luma, blue chroma difference, and red chroma difference (YCbCr/YPbPr) can be used. Whatever your GPU renders internally (typically 16-bit floating point RGBA, where A is the alpha/transparency information), that data gets converted into a signal for your display.
The standard in the past has been 24-bit color, or eight bits each for the red, green and blue color components.HDRand high color depth displays have bumped that to10-bitcolor, with 12-bit and 16-bit options as well, though the latter two are mostly in the professional space. Generally speaking, display signals use either 24 bits per pixel (bpp) or 30 bpp, with thebest HDR monitorsopting for 30 bpp. Multiply the color depth by the number of pixels and the screen refresh rate and you get the minimum required bandwidth. We say ‘minimum’ because there are a bunch of other factors as well.
Display timings are relatively complex calculations. The VESA governing body defines the standards, and there’seven a handy spreadsheetthat spits out the actual timings for a given resolution. A 1920×1080 monitor at a 60 Hz refresh rate, for example, uses 2,000 pixels per horizontal line and 1,111 lines once all the timing stuff is added. That’s because display blanking intervals need to be factored in. (These blanking intervals are partly a holdover from the analog CRT screen days, but the standards still include it even with digital displays.)
Using the VESA spreadsheet and running the calculations gives the following bandwidth requirements. Look at the following table and compare it with the first table; if the required data bandwidth is less than the max data rate that a standard supports, then the resolution can be used.
Common Resolution Bandwidth Requirementsb
Color Depth
Refresh Rate (Hz)
Required Data Bandwidth
1920 x 1080
8-bit
60
3.20 Gbps
1920 x 1080
10-bit
60
4.00 Gbps
1920 x 1080
8-bit
144
8.00 Gbps
1920 x 1080
10-bit
144
10.00 Gbps
2560 x 1440
8-bit
60
5.63 Gbps
2560 x 1440
10-bit
60
7.04 Gbps
2560 x 1440
8-bit
144
14.08 Gbps
2560 x 1440
10-bit
144
17.60 Gbps
3840 x 2160
8-bit
60
12.54 Gbps
3840 x 2160
10-bit
60
15.68 Gbps
3840 x 2160
8-bit
144
31.35 Gbps
3840 x 2160
10-bit
144
39.19 Gbps
3840 x 2160
8-bit
240
56.45 Gbps (~19 DSC)
3840 x 2160
10-bit
240
70.56 Gbps (~24 DSC)
7680 x 4320
8-bit
60
49.99 Gbps (~17 DSC)
7680 x 4320
10-bit
60
62.49Gbps (~21 DSC)
7680 x 4320
8-bit
120
103.62 Gbps (~35 DSC)
7680 x 4320
10-bit
120
129.53 Gbps (~43 DSC)
7680 x 4320
8-bit
240
223.48 Gbps (~75 DSC)
7680 x 4320
10-bit
240
279.35 Gbps (~93 DSC)
The above figures are foruncompressedsignals, however, and DisplayPort 1.4 added the option ofDisplay Stream Compression1.2a (DSC), which is also part of HDMI 2.1. In short, DSC helps overcome bandwidth limitations, which are becoming increasingly problematic as resolutions and refresh rates increase. For example, basic 24 bpp at 8K and 60 Hz needs 49.65 Gbps of data bandwidth, or 62.06 Gbps for 10 bpp HDR color — the former could be supported by DisplayPort 2.1 UHBR13.5, while the latter would require UHBR20 (which few monitors currently support, and which is only available on RTX 50-series and one port for AMD’s professional W7900 GPU).
DSC can provide up to a 3:1 compression ratio by converting to YCgCo and using delta PCM encoding. It provides a “visually lossless” (and sometimes even truly lossless, depending on what you’re viewing) result. Using DSC, 8K 120 Hz HDR is suddenly viable, with a bandwidth requirement of ‘only’ 42.58 Gbps. DisplayPort 1.4 can also run 4K at 240 Hz using DSC.
There’s a catch with DSC, however: Support can be a bit hit and miss, particularly on older GPUs. We’ve tested a bunch of graphics cards using a Samsung Odyssey Neo G8 32, which supports up to 4K at 240 Hz over DisplayPort 1.4 or HDMI 2.1. On DisplayPort connections, most of the latest GPUs are fine, but cards from 2016 and earlier may not even allow the use of 240 Hz. We’ve also seen video signal corruption on occasion, where dropping to 120 Hz (still with DSC) often fixes the problem. In short, cable quality and the DSC hardware implementation still factor into the equation.
Both HDMI and DisplayPort can also carry audio data, which requires bandwidth as well, though it’s a minuscule amount compared to the video data. DisplayPort and HDMI currently use a maximum of 36.86 Mbps for audio, or 0.037 Gbps if we keep things in the same units as video. Earlier versions of each standard can use even less data for audio. One important note is that HDMI supports audio pass through, while DisplayPort does not. If you’re planning to hook up your GPU to an amplifier, HDMI provides a better solution.
That’s a lengthy introduction to a complex subject, but if you’ve ever wondered why the simple math (resolution * refresh rate * color depth) doesn’t match published specs, it’s because of all the timing standards, encoding, audio, and more. Bandwidth isn’t the only factor, but in general, the standard with a higher maximum bandwidth is ‘better.’
DisplayPort: The PC Choice
Currently DisplayPort 2.1 is the most capable version of the DisplayPort standard. The DisplayPort 2.0 spec came out in June 2019, and Intel’s Arc Alchemist GPUs along with AMD’s RDNA 3 GPUs supported the standard. It was later revised to DisplayPort 2.1, but all DP2.0 capable hardware should still be compatible. Nvidia finally added DisplayPort 2.1b with its Blackwell RTX 50-series GPUs.
So, there are now cards with DisplayPort 2.1 support, but they’re still of different levels. Intel’s Arc GPUs support UHBR10 (Ultra-High Bitrate 10 Gbps per lane), for a 40 Gbps maximum connection speed (not including 128b/132b encoding). AMD opted for the faster UHBR13.5 (54 Gbps total), but neither company supports the potential 20 Gbps per lane variant — except for AMD allowing UHBR20 on a single port for its professional Radeon Pro W7900 graphics card. Nvidia went with full UHBR20 (80 Gbps) support on all outputs for it’s 50-series and future solutions. But perhaps the bigger issue now isn’t GPU support.
There still aren’t many displays that support DisplayPort 2.1. Those are starting to appear, but it’s the old chicken and egg scenario. GPUs with DisplayPort 2.1 support have now been around for three years, monitors that can use DP2.1 have been lagging behind. Probably that’s because DisplayPort 1.4 remains sufficient for up to 4K 240 Hz and 8K 60Hz with DSC, and HDMI 2.1 support is there for people that need up to 48 Gbps.
There are DP2.1 monitors now, supposedly with UHBR20 support. Curiously, we’ve seen 4K 240 Hz OLED monitors that will still enable DSC if you want to use the full capabilities of the display. That should be possible without DSC, but at least one monitor we’ve tested (MSI MPG272UX OLED) drops the maximum 4K refresh rate to 120 Hz if we disable DSC in the monitor OSD (on-screen display).
One advantage of DisplayPort is that variable refresh rates (VRR) have been part of the standard since DisplayPort 1.2a. We also like the robust DisplayPort connector (but not mini-DisplayPort), which has hooks that latch into place to keep cables secure. It’s a small thing, but we’ve definitely pulled loose more than a few HDMI cables by accident. DisplayPort can also connect multiple screens to a single port via Multi-Stream Transport (MST), and the DisplayPort signal can be piped over a USB Type-C connector that also supports MST.
One area where there has been some confusion is in regards to licensing and royalties. DisplayPort was supposed to be a less expensive standard, but both HDMI and DisplayPort have various associated brands, trademarks, and patents that have to be licensed. With technologies like HDCP (High-bandwidth Digital Content Protection), DSC, and more, companies have to pay a royalty for DP just like HDMI. The current rate appears to be $0.20 per product with a DisplayPort interface, with a cap of $7 million per year. HDMI charges $0.15 per product, or $0.05 if the HDMI logo is used in promotional materials.
Because the standard has evolved over the years, not all DisplayPort cables will work properly at the latest speeds. The original Display 1.0-1.1a spec allowed for RBR (reduced bit rate) and HBR (high bit rate) cables, capable of 5.18 Gbps and 8.64 Gbps of data bandwidth, respectively. DisplayPort 1.2 introduced HBR2, doubled the maximum data bit rate to 17.28 Gbps and is compatible with standard HBR DisplayPort cables. HBR3 with DisplayPort 1.3-1.4a increased things again to 25.92 Gbps, and added the requirement of DP8K DisplayPort certified cables.
Finally, with DisplayPort 2.1 there are three new transmission modes: UHBR10 (ultra high bit rate), UHBR13.5 and UHBR20. The number refers to the bandwidth of each lane, and DisplayPort uses four lanes, so UHBR10 offers up to 40 Gbps of transmission rate, UHBR13.5 can do 54 Gbps and UHBR20 peaks at 80 Gbps. DP2.1 uses 128b/132b encoding, meaning data bit rates of 38.69 Gbs, 52.22 Gbps, and 77.37 Gbps. And now there are new cables to meet those standards.
Officially, the maximum length of a DisplayPort cable is up to 3m (9.8 feet), which is one of the potential drawbacks, particularly for consumer electronics use. Newer versions with higher bandwidths can cut that length even more. As an example, checking the official DisplayPort certification list, the longest DP80 certified cable right now is only 1.2m (3.94 ft) long, and many are only 0.8 or 1.0 meters. Getting a cable that’s only 3 feet or less in length generally means you have to have your PC on top of your desk.
With a maximum data rate of 25.92 Gbps, DisplayPort 1.4 can handle 4K resolution 24-bit color at 98 Hz, and dropping to 4:2:2 YCbCr gets it to 144 Hz with HDR. Alternatively, DSC allows up to 4K and 240 Hz, even with HDR. Keep in mind that 4K HDR monitors running at 144 Hz or more carry premium pricing, so gamers will more likely be looking at something like a 144Hz display at 1440p. That only requires 14.08 Gbps for 24-bit color or 17.60 Gbps for 30-bit HDR, which DP 1.4 can easily handle.
If you’re wondering about 8K content in the future, the reality is that even though it’s doable right now via DSC and DisplayPort 1.4a or HDMI 2.1b, the displays and PC hardware needed to drive such displays aren’t generally within reach of consumer budgets. Top-tier GPUs like the GeForce RTX 4090 and GeForce RTX 5090 sort of overcome that limitation, but 8K pixel densities often outstrip modest human eyesight. By the time 8K becomes a viable resolution, both in price and in the GPU performance required to run it adequately, we’ll likely have gone through another generation or three of GPU hardware.
HDMI: Ubiquitous Consumer Electronics
(Image credit: HDMI.org)
Updates to HDMI have kept the standard relevant for over 20 years. The earliest versions of HDMI have become outdated, but later versions have increased bandwidth and features.
HDMI 2.0b and earlier are ‘worse’ in some ways compared to DisplayPort 1.4, but if you’re not trying to run at extremely high resolutions or refresh rates, you probably won’t notice the difference. Full 24-bit RGB color at 4K 60 Hz has been available since HDMI 2.0 released in 2013, and higher resolutions and/or refresh rates are possible with 4:2:0 YCbCr output — though you generally don’t want to use that with PC text, as it can make the edges look fuzzy.
For AMD FreeSync users, HDMI has also supported VRR via an AMD extension since 2.0b, but HDMI 2.1 is where VRR became part of the official standard. AMD and Nvidia all support HDMI 2.1 with VRR, starting with Turing and RDNA 2. (Intel has also supported the standard since its first Alchemist GPUs.) Nvidia opted to call its HDMI 2.1 VRR solution “G-Sync Compatible,” and you can find a list of all the officially tested andsupported displays on Nvidia’s site.
One major advantage of HDMI is that it’s ubiquitous. Millions of devices with HDMI shipped in 2004 when the standard was young, and it’s now found everywhere. These days, consumer electronics devices like TVs often include support for three or more HDMI ports. TVs and consumer electronics hardware have been shipping HDMI 2.1 devices for a while, before PCs even had support. Finding a TV with a DisplayPort input, by contrast, remains very uncommon.
HDMI cable requirements have changed over time, just like DisplayPort. One of the big advantages is that high quality HDMI cables can be up to 15m (49.2 feet) in length — five times longer than DisplayPort. That may not be important for a display sitting on your desk, but it can definitely matter for home theater use. Originally, HDMI had two categories of cables: category 1 or standard HDMI cables are intended for lower resolutions and/or shorter runs, and category 2 or “High Speed” HDMI cables are capable of 1080p at 60 Hz and 4K at 30 Hz with lengths of up to 15m.
More recently, HDMI 2.0 introduced “Premium High Speed” cables certified to meet the 18 Gbps bit rate, and HDMI 2.1 has created a fourth class of cable, “Ultra High Speed” HDMI that can handle up to 48 Gbps. HDMI also provides for routing Ethernet signals over the HDMI cable, though this is rarely used in the PC space.
We mentioned licensing fees earlier, and while HDMI Technology doesn’t explicitly state the cost, this website details the variousHDMI licencing feesas of 2014. The short summary: for a high volume business making a lot of cables or devices, it’s $10,000 annually, and $0.05 per HDMI port provided HDCP (High Definition Content Protection) is used and the HDMI logo is displayed in marketing material. In other words, the cost to end users is easily absorbed in most cases — unless some bean counter comes down with a case of extreme penny pinching.
Like DisplayPort, HDMI also supports HDCP to protect the content from being copied. That’s a separate licensing fee, naturally (though it reduces the HDMI fee). HDMI has supported HDCP since the beginning, starting at HDCP 1.1 and reaching HDCP 2.2 with HDMI 2.0. HDCP can cause issues with longer cables, and ultimately it appears to annoy consumers more than the pirates. At present, known hacks / workarounds to strip HDCP 2.2 from video signals can be found.
HDMI 2.1 allows for up to 48 Gbps signaling rates, and it also supports DSC. Theoretically, that means resolutions and refresh rates of up to 4K at 480 Hz or 8K at 120 Hz are supported over a single connection and cable. We’re not aware of any 4K 480 Hz displays yet, though there are prototype 8K 120 Hz TVs that have been shown atCES.
DisplayPort vs. HDMI: The Bottom Line for Gamers
We’ve covered the technical details of DisplayPort and HDMI, but which one is actually better for gaming? Some of that will depend on the hardware you already own or intend to purchase. Both standards are capable of delivering a good gaming experience, but if you want a great gaming experience, right now DisplayPort 1.4 is generally better than HDMI 2.0, HDMI 2.1 technically beats DP 1.4, and DisplayPort 2.1 trumps HDMI 2.1 (provided you have UHBR13.5 or higher support). The problem is that you’ll need support for the desired standard from both your graphics card and your display for things to work right.
For most Nvidia gamers, your best option right now is a DisplayPort 1.4 connection to a G-Sync certified (compatible or official) display. Alternatively, HDMI 2.1 with a newer display works as well. Both the RTX 30-series and 40-series cards support the same connection standards, for better or worse. Most graphics cards will come with three DisplayPort connections and a single HDMI output, though you can find models with two HDMI and two (or three) DisplayPort connections as well — only four active outputs at a time are supported. RTX 50-series GPUs meanwhile can benefit from DisplayPort 2.1 monitors, so if you’re planning on picking up a 4K 240 Hz display with an RTX 5080 or 5090, that’s a potent combination.
AMD gamers have a few more options. You can find DisplayPort 2.1 monitors and TVs, if you look hard enough. The Asus ROG Swift PG32UXQR for example supports DisplayPort 2.1. HDMI 2.1 connectivity is also sufficient, and there are more displays available. Keep in mind that maximum bandwidth of both the RDNA 3 and RDNA 4 GPUs is 54 Gbps over DisplayPort 2.1, or 48 Gbps over HDMI 2.1, so it’s not a huge difference. Most AMD RX 7900-series cards that we’ve seen include two DisplayPort 2.1 ports, and either two HDMI 2.1 or a single HDMI 2.1 alongside a USB Type-C connection. The newer RX 9070-series GPUs typically have triple DP2.1 and a single HDMI port.
Intel’s GPUs support DP2.1 UHBR10, with the Battlemage B-series parts adding a single UHBR13.5 port. VRR is also supported, if you have an appropriate Adaptive Sync monitor. Basically, you don’t want to try using a G-Sync display that isn’t Adaptive Sync compatible with either AMD or Intel GPUs.
What if you already have a monitor that isn’t running at higher refresh rates or doesn’t have G-Sync or FreeSync capability, and it has both HDMI and DisplayPort inputs? Assuming your graphics card also supports both connections (and it probably does if it’s a card made in the past eight years), in many instances the choice of connection won’t really matter.
2560×1440 at a fixed 144 Hz refresh rate and 24-bit color works just fine on DisplayPort 1.2 or higher, as well as HDMI 2.0 or higher. Anything lower than that will also work without trouble on either connection type. About the only caveat is that sometimes HDMI connections on a monitor will default to a limited RGB range, but you can correct that in the AMD, Intel, or Nvidia display options. (This is because old TV standards used a limited color range, and some modern displays still think that’s a good idea. News flash: it’s not.)
Other use cases might push you toward DisplayPort as well, like if you want to use MST to have multiple displays daisy chained from a single port. That’s not a very common scenario, but DisplayPort does make it possible. Home theater use on the other hand continues to prefer HDMI, and the auxiliary channel can improve universal remote compatibility. If you’re hooking up your PC to a TV, HDMI is usually required, as there aren’t many TVs that have a DisplayPort input.
You can do 4K at 60 Hz on both standards without DSC, so it’s only 8K or 4K at refresh rates above 60 Hz where you actually run into limitations on recent GPUs. We’ve used AMD, Intel, and Nvidia GPUs at 4K and 98 Hz (8-bit RGB) with most models going back to the 2016 era, and 4:2:2 chroma can push even higher refresh rates if needed. Modern gaming monitors like the Samsung Odyssey Neo G8 32 with 4K and up to 240 Hz are also available, with DisplayPort 1.4 and HDMI 2.1 connectivity.
Ultimately, while there are certain specs advantages to DisplayPort and some features on HDMI that can make it a better choice for consumer electronics use, the two standards end up overlapping in many areas. The VESA standards group in charge of DisplayPort has its eyes on PC adoption growth, whereas HDMI is defined by a consumer electronics consortium and thinks about TVs first. But DisplayPort and HDMI end up with similar capabilities.
Better GPU performance for Tensor and no satellites for the 9a
The Google Pixel 9(Image credit: Philip Berne / Future)
Pixel models from the Pixel 6 onwards are achieving improved benchmark results
This is likely down do newer GPU drivers being added in recent software updates
In less positive news, the Pixel 6a lacks satellite communication features
There’s both good and bad Pixel news today, but the good news will affect more people than the bad, so let’s start there.
Reddit usersare finding that Pixel phones with Tensor chipsets (meaning everything from theGoogle Pixel 6onwards) are achieving much higher GPU scores on Geekbench 6 than they did at launch. This is widely being attributed to the Android 16 beta, butAndroid Authorityreports seeing similarly upgraded performance onAndroid 15.
So chances are you don’t need to grab a beta version of Android to see improvements, but rather that recent stable software updates have massively boosted GPU performance.
The exact boost varies depending on model, but Android Authority claims itsPixel 6aunit saw a nearly 23% GPU performance increase, while elsewhere there are reports of a 62% improvement for thePixel 7a, a 31% improvement for thePixel 8, and even a 32% improvement for the recentGoogle Pixel 9.
Android Authority speculates thatGoogleachieved this through including newer GPU drivers in recent Android updates, as while all recent Pixels use anArmMali GPU, they don’t always ship with the latest available GPU driver version.
How much impact these performance improvements will have in the real world remains to be seen, but they’re nice to see, and could help extend the lifespan of older Pixel models.
Now for the bad news, and this relates specifically to the newGoogle Pixel 9a, which we’ve learned doesn’t support Satellite SOS. Google confirmed as much toAndroid Authority, and this is a feature found on other Google Pixel 9 models which allows you to contact emergency services in areas without Wi-Fi or cell signal.
So it’s a potentially life-saving tool, and while Google didn’t say why it’s absent here, it’s likely because the Pixel 9a uses an older Exynos Modem 5300, rather than the 5400 used by the rest of the Pixel 9 series.
While this is a feature that you’ll hopefully never need to use, it would be reassuring to have, and this isn’t the only omission in the Pixel 9a, as we also recently learned thatit lacks several AI toolsoffered by the rest of the Pixel 9 line.
In fact, this phone has had a slightly troubled launch, with not just these omissions emerging, but also a delay in sales of the phone while Google investigatesa “component quality issue”.
Still, the silver lining there is that this delay allowed time for these omissions to be uncovered, so you might think twice about buying the Google Pixel 9a. Certainly, we’d wait until we’ve had a chance to put it through a full review before purchasing one.
