An OLED monitor has been on my wishlist since forever, but I'm not about to drop $800–$1,000 on one anytime soon. Fortunately, I got to experience a fantastic OLED display that I recently received for testing. I spent a few days on the display, playing games and working a bit, just to see what having an OLED monitor felt like.

The Steam Deck OLED is a great way to play your games on the go, which is why it's important for it to have a strong and stable WiFi connection. I mean, sure, you can install a USB Ethernet port on it, but if you want to, say, play it online at a cafe, you're very much dependent on the Steam Deck's WiFi capabilities.

While we love our Raspberry Pi projects here at XDA, sometimes you don't want the whole Pi; sometimes you just want a slice. Such is the case for people who purchase the individual microcontrollers you can find on the Raspberry Pi for their bespoke projects. And while the microcontrollers are a fantastic way to add a teeny-tiny brain to whatever you like, it doesn't come with either a WiFi or Bluetooth adapter. Instead, you need to hook it up to an adapter yourself.

My working setup is somewhat odd. I use a MacBook Pro, plugged into a docking station that splits the video output into two separate monitors, and I use a SteelSeries Apex 7 keyboard. When I first built this configuration, I wanted a more clicky, larger keyboard than any of the Apple Magic Keyboards. There was just one problem: the Windows key was the Command key, but it's positioned one key over from where it would be on a Mac.

Most of us have clicked on a bit.ly or t.co link without a second thought, and these links are practically everywhere. These are known as link-shorteners, or URL-shorteners, and these services are baked into social platforms, email tools, and even print ads. Plenty of people use these for all kinds of things, but why use a publicly-hosted URL shortener when you can host your own instead?

AT&T believes its network wasn’t at fault for a conference call where President Donald Trump accused the company of being “totally unable to make their equipment work properly.” Instead, AT&T is blaming an unnamed “conference call platform.” Earlier on Monday, President Donald Trump complained on Truth Social about apparent issues with AT&T’s network during a “major conference call with faith leaders from all over the country” that had “tens of thousands of people on the line.” Trump said that “this is the second time it’s happened” and that if “the Boss of AT&T, whoever that may be” would get involved, “it would be good.”It appears AT&T initially wasn’t sure what was going on, as it replied to White House Press Secretary Karoline Leavitt on X to say that it had reached out to the White House ”to quickly understand and assess the situation.”At 6:53PM ET, the company said that “our initial analysis indicates the disruption was caused by an issue with the conference call platform, not our network. Unfortunately, this caused the delay, and we are working diligently to better understand the issue so we can prevent disruptions in the future.”View LinkAT&T didn’t say which conference call platform it believes is at fault, and didn’t immediately respond to The Verge’s question about that. The White House didn’t immediately reply to a request for comment.Shortly after his original post, Trump followed up to say that the call may be rescheduled and that “we’ll use another carrier next time.”

I always love it when someone builds a teeny-tiny version of a common device or machine. There's just something so cute about it, almost as if it were invented by some mad scientist trying to avoid doing the housework. Even if you have zero intent of actually making the device yourself, it's still a great way to get inspiration for your own Raspberry Pi and ESP32 projects.

Have you noticed that OneNote is a little blurrier than the other Office apps? If so, you've seen the app's lack of a DPI-changing feature. Microsoft Office apps usually consider your monitor's DPI, even if you move the window from one screen to another. Unfortunately, OneNote has been left behind, as it doesn't (or, more accurately, didn't) have this feature.

Tinder is trialing mandatory facial recognition security features in the US to verify profiles and crack down on impersonation and fake accounts. New users in California are now required to provide a biometric “Face Check” scan to confirm their face matches their profile photos for the dating service, Axios reported on Monday.The Face Check feature involves taking a short video selfie that’s used to match biometric indicators and prove that the Tinder user isn’t a bot using artificially generated images, providing them with a verified badge upon completion. The scan will also check if the user’s face is being used in multiple accounts, which could help to prevent users from being impersonated or having their likeness used by deceptive “catfish” profiles. Face Check is separate from Tinder’s ID Check feature, which uses government-issued ID to verify users’ age and identity, while Face Check seemingly only requires users to upload a selfie video. Tinder users have provided video selfies to verify their profiles since 2023, but verification wasn’t a mandatory requirement for creating a Tinder account. This change means that Californians will have to complete some version of verification if they want to use the platform at all. “We see this as one part of a set of identity assurance options that are available to users,” Match Group’s head of trust and safety, Yoel Roth, told Axios. “Face Check … is really meant to be about confirming that this person is a real, live person and not a bot or a spoofed account.”Tinder says the selfie video is deleted once verification is complete, but that the platform stores a “non-reversible, encrypted face map” to detect duplicate user accounts in the future. The Face Check feature has already been piloted in Colombia and Canada, with Roth telling Axios that those tests showed “promising” results in “improving perceptions of authenticity” and reducing reports of bad actors. Tinder will now monitor how users in California respond to the Face Check feature before deciding if it should be rolled out more broadly across the US, according to Roth.

I’ve been riding the self-hosting wave in my small home lab for the past few months. Thanks to my XDA colleagues and self-hosted communities, I no longer hesitate to deploy a Docker container to try out a new app. Managing and maintaining my favorite Docker containers became tough, so I employed Portainer. However, it felt limiting when I developed an affinity for Docker Compose, which allowed me to deploy containers and manageYAML files. I felt right at home when I tried Dockge as an alternative to container management for hosting Compose files.

Valve has added a new performance monitor to Steam that can help you understand why a game may or may not be running smoothly. Not only does it break out a game’s overall frame rate, it can tell you how many of those frames were generated by techniques  like Nvidia’s DLSS or AMD’s FSR, according to a post.The change is included as part of an update to the Steam Client that’s available now, though Valve notes that this “first version” focuses on “Windows users and on the most common GPU hardware.”The company says the new performance monitor currently offers up to four different levels of detail: a single FPS value, FPS details, CPU and GPU utilization, and “FPS, CPU, GPU & RAM Full Details.” The more you choose to show, the more of your screen will be taken up by the performance monitor.Steam previously offered a simple FPS counter, but separating out generated frames from the frames fully rendered by your graphics cores can help you better understand key differences between what you see and how a game feels. “Frame generation can’t help with things like input latency that matter to competitive gamers, but it can make things look visually smoother on today’s high refresh rate monitors,” Valve says in a detailed support document about the performance monitor. In practice, what that should mean is that you can see whether your game feels like it’s running at just 30 fps because it actually is running at 30 fps inside the game engine, even though you’re seeing a visually smoother image due to Nvidia and AMD’s added “fake frames.” (It’s a whole debate in the PC gaming community, and it appears Valve isn’t taking sides here.) Valve has already given handheld gamers a taste of these quick insights by building tools like MangoHud into the Steam Deck and SteamOS, which similarly let you monitor your CPU, graphics, RAM, and carefully ration out your battery life. But having a way to do so built into desktop Steam will make the insights much more accessible to many more gamers.In the future, Valve says that it has plans to “add some additional pieces of data to the performance overlay going forward, to detect certain common bad hardware performance scenarios, and to show a larger summary of your game’s performance in the overlay itself when you hit shift-tab.”

You can't go wrong with a portable SSD if you're looking for additional storage. Not only are you going to get great performance, but you're also getting something that's compact and durable. Of course, you're going to be paying the price with SSDs, being some of the more expensive options when it comes to storage media.