Besides those, AMD also expanded its Ryzen AI 300 series APU lineup today that so far comprised of the Ryzen AI 9 HX 375, 370, and 365 and their corresponding PRO counterparts. On one hand, the company is bringing more affordable AI chips with the new Ryzen AI 7 (PRO) 350 and Ryzen AI 5 (PRO) 340. On the other hand, it is also debuting more powerful Ryzen AI Max and Max+ SKUs.

In case you have not been following, an NPU needs to do over 40 TOPS which is the threshold for qualifying Microsoft's Copilot+ PC requirement. Thus the newly released Ryzen 200 series for example does not qualify since they top out at just 16 TOPS.

Going forward, the company is dividing its AI PC chips into three classes: Halo, Premium, and Advanced. The new Max and Max+ SKUs will comprise the Halo segment, the Ryzen AI 9 will be the Premium tier, and finally, the Ryzen AI 7/5 will belong to the Advanced class.

The new Max+ parts are being touted very highly as AMD claims performance leadership over the likes of Nvidia RTX 4090, Apple M4 Pro, and Intel Ultra 9 288V across various tasks. For example, the company says that its new Ryzen AI Max+ 395 APU is on average 2.6 times faster than Intel 288V in rendering and around 1.4 times faster in (synthetic) gaming.

Compared against Apple's 14-core M4 Pro, AMD expects better performance on the Max+ 395 in rendering with Blender, Corona, and V-Ray, with V-Ray showing the most favorable result.

Finally, the Max+ 395 has been compared against Nvidia RTX 4090 where AMD once again claims victory as the Max+ is said to be 2.2 times faster in AI processing (in tokens/second) on LM Studio. The company boasts about the 395 here claiming it as "the World's first Copilot PC+ processor to run 70B LLM".

AMD added that its Max APU can allocate up to 96 GB of memory to the integrated 40 CU RDNA 3.5 graphics portion out of the total 128 GB of memory that is available to it.

If you are wondering how the new Ryzen AI Max+ APUs are so powerful, that's because it has 16 Zen 5 cores on the CPU side, just like the Ryzen 9955HX(3D), and it also packs 40 RDNA 3.5 comput units (CUs), and finally a 50 TOPS capable XDNA 2 NPU.

What is interesting with the Halo Copilot+ (Max and Max+) class of AMD Ryzen AI processors is that these SKUs feature a "new memory interface" which the company refers to as "Unified Coherent Memory Architecture." Thanks to this new interface, the Ryzen AI Max and Max+ APUs enjoy a memory bandwidth of up to 256 GB/s.

The image above shows the full Ryzen AI 300 Max and Max+ lineup as well as their availabilities.

Aside from performance data of the Max+ APU, AMD has also shared how it expects the Ryzen AI 7 350 to compete against Intel and Qualcomm. According to the provided figures, the company expects the 350 APU to be up to 49% better than the Snapdragon X Plus X1P-42-100, and up to 78% faster than Intel Core Ultra 7 258V.

You can check the full specs and availability of the Ryzen AI 350 and 340 in the image above.

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every day for many years.

News posts... 2023: 5,800+ | 2024: 5,700+

RIP Matrix | Farewell my friend  

AMD's X3D chips incorporate a special sauce called the 3D V-cache (vertical cache) that stacks a ton of cache vertically on top of each other to drastically increase the amount of last-level cache (LLC) or L3 in a rather small footprint as workloads like games absolutely love it. AMD launched its 2nd Gen 3D V-cache last year with the Ryzen 7 9800X3D.

The company has provided a performance comparison of the new Ryzen 9 9950X3D against the 7950X3D and the Intel Ultra 9 285K.

Against its Intel competition, AMD claims the 9950X3D can offer up to 64% faster gaming performance in case of titles like Watch Dogs Legion. Across 40 games, the company says its new 16 core X3D part is on average 20% faster than Intel's 285K. AMD tested these at 1080p to eliminate GPU bottlenecks (read our dedicated article here to understand why it's important) on Windows 11 24H2 with VBS enabled.

AMD has also provided gaming data against the 7950X3D. It looks like folks will have a harder time justifying this upgrade as the company shows an 8% improvement across 40 titles. However, in some games, users can expect to see massive gains like in the case of Counter Strike 2 where 58% bump is claimed.

While AMD did not directly compare the Ryzen 9950X3D with the 9800X3D today, the firm has in the past shared gaming performance data of the latter vs the Intel 285K.

The table below shows the performance of the 9800X3D vs 9950X3D using the Intel 285K as baseline since it's at 100% in both the charts:

It looks like the 9800X3D remains the fastest gaming CPU overall although the 9950X3D is also close behind. There are exceptions to this and we have bolded all such interesting cases.

In terms of content creation performance, AMD claims that the new 9950X3D is 10% faster than the 285K and 13% faster than the 7950X3D respectively across 20 applications. The new AMD X3D CPU is especially better in rendering apps (like Corona and Blender) vs its 7000 series predecessor. meanwhile against the Intel 285K, the largest differentials are seen in Adobe Photoshop and 7-Zip.

Aside from the 16 core 32 thread 9950X3D, AMD also unveiled the 12 core 24 thread 9900X3D. Spec-wise it is nearly identical to the 7900X3D though the boost clock is down by 100 MHz at 5.5 GHz. Thus the Ryzen 9 9900X3D also has 140 MB of total cache with 128 MB of L3 (3D V-cache).

In terms of availability, AMD expects the Ryzen 9950X3D and 9900X3D to be out on shelves in Q1 2025 although no particular dates were disclosed.

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every day for many years.

News posts... 2023: 5,800+ | 2024: 5,700+

RIP Matrix | Farewell my friend  

Thus the new Ryzen 9 9955HX3D is AMD's second X3D CPU for laptops and notebooks. Specwise the new part remains identical to the 7945HX3D with 16 cores 32 threads 144 MB of total cache, but the CPU architecture has gone from Zen 4 in the 7945HX3D to Zen 5 in 9955HX3D.

AMD has not detailed the performance of the 9955HX3D but simply is calling it "the world's best gaming and content creation mobile processor."

Aside from the 9955HX3D, AMD has also announced a couple of non-X3D parts in the form of 16 core 9955HX and 12 core 9850HX. Since these do not have the 3D V-cache, the total cache on these two are 80 MB and 76 MB respectively. The TDP of all three CPUs is set to 54 watts.

Besides high-core count mobile workstation parts, AMD is also announcing new Ryzen 200 series mainstream processors for notebooks comprising four, six and eight core SKUs. These are different from the Ryzen AI Max/+ and 300 series APUs meant for Copilot+ PCs.

At the top of the lineup we have the Ryzen 9 270, Ryzen 7 260, and 250, which are the octa-core parts. Following that, we have Ryzen 5 240, 230, and 220, which are hexa-cores. There is a single quad-core SKU in this lineup in the form of the Ryzen 3 210.

The TDP of these chips is configurable and the PRO variants (as well as their non-PRO counterparts) are the more efficient ones with cTDP of 15 to 30 watts. Meanwhile, the rest of the lineup features a cTDP of 35 to 54 watts. Except the Ryzen 220 and 210, all the other models have an NPU that is rated at 16 peak NPU TOPS.

AMD says the Ryzen 200 series mobile will be available in Q2 of 2025. Meanwhile, the high-core count 9000 series mobile parts will be launched sometime in this half of the year (H1 2025).

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every day for many years.

News posts... 2023: 5,800+ | 2024: 5,700+

RIP Matrix | Farewell my friend  

Transform your PC experience with the processor tailored for today's dynamic on-the-go lifestyles-on a budget. Delivering incredible performance and all-day battery life with a responsive and efficient CPU, integrated GPU, and groundbreaking NPU. It all adds up to a quantum leap forward in computing, with built-in Al to help you achieve more and ease of use to simplify tasks with ease. Get the most out of your Copilot+ PC wherever your day takes you.

The new Snapdragon X1-26-100 (just "Snapdragon X") joins other Snapdragon X Plus and Elite models in the lineup. In a nutshell, it is the same 8-core Snapdragon X Plus processor (X1P-42-100), just with lower clocks—3.0GHz instead of 3.4GHz. GPU, NPU, memory, cache, and cores are the same.

By keeping the same NPU, Qualcomm ensures users can have the same AI-powered experience without performance compromises. Also, having the same GPU enables better display output, with the processor being able to power three external 4K 60Hz monitors or two 5K 60Hz/4K 120Hz displays. The chip also supports Wi-Fi 7 and optional 5G, AV1 encoding and decoding, and 4K HDR streaming.

The first devices with the new Snapdragon X processors will show up at CES 2025. Also, we are finally about to see Qualcomm delivering its promise to expand its ARM chips into other form factors. The company says the first mini PC with Snapdragon X Series chips will be unveiled at CES 2025.

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every day for many years.

News posts... 2023: 5,800+ | 2024: 5,700+

RIP Matrix | Farewell my friend  

We've all had a good seven years to figure out why our interconnected devices refused to work properly with the HDMI 2.1 specification. The HDMI Forum announced at CES today that it's time to start considering new headaches. HDMI 2.2 will require new cables for full compatibility, but it has the same physical connectors. Tiny QR codes are suggested to help with that, however.

The new specification is named HDMI 2.2, but compatible cables will carry an "Ultra96" marker to indicate that they can carry 96GBps, double the 48 of HDMI 2.1b. The Forum anticipates this will result in higher resolutions and refresh rates and a "next-gen HDMI Fixed Rate Link." The Forum cited "AR/VR/MR, spatial reality, and light field displays" as benefiting from increased bandwidth, along with medical imaging and machine vision.

Examples of how HDMI 2.2's synchronization abilities will benefit home theaters.

 

A visualization of how far HDMI has come in bandwidth, from 1.0 to 2.2.

 

A bit closer to home, the HDMI 2.2 specification also includes "Latency Indication Protocol" (LIP), which can help improve audio and video synchronization. This should matter most in "multi-hop" systems, such as home theater setups with soundbars or receivers. Illustrations offered by the Forum show LIP working to correct delays on headphones, soundbars connected through ARC or eARC, and mixed systems where some components may be connected to a TV, while others go straight into the receiver.

HDMI 2.2 is not a real thing you can buy into yet, as both cable and device manufacturers have not received the full specification yet and need time to build supported components.

As to whether you need to buy into HDMI 2.2, that's another issue. You can already get uncompressed 8K, 60HZ video signals with 8-bit color depth or compressed 10K at 120Hz with 12-bit color depth through a top-flight HDMI-2.1b-compliant cable. It's the rare home system that needs more than that right now, let alone one that has any content to actually play at those rates.

The Forum, perhaps knowing the deep grief that is trying to buy a "high speed" cable in modern times, wants people to look for an "Ultra96" badge on cable packaging. In addition, one can scan a QR code and see that a cable has been certified and is not counterfeit.

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every day for many years.

News posts... 2023: 5,800+ | 2024: 5,700+

RIP Matrix | Farewell my friend  

For starters, Gemini AI is an artificial intelligence model developed by Google that can understand and respond to voice commands in a more conversational manner. This means that users can ask their TVs questions or give commands using everyday language, making interactions feel less robotic and more like a conversation with a friend.

Users will be able to interact with their Google TVs using simpler and more casual phrases. For instance, you could simply ask, "What are the newest movies from Disney?" and the AI would understand your request and provide relevant results.

Newer Google TV devices, such as the Google TV Streamer also come equipped with far-field microphones that would allow users to communicate with the TV from across the room without needing a remote control, or saying "Hey Google" every time. Google also gave a preview of a new ambient experience that makes the use of proximity sensors, the same ones on your phone that would turn off the screen whenever you'd pick up your phone and hold it against your ear. The ambient experience would include personalised widgets like weather updates and news when not actively being used.

The TV would also show AI-generated screensavers and artwork when it is left idle, part of the new "always-on" mode.

Gemini will also significantly improve content recommendations on Google TV since it can analyze user preferences and viewing habits. Using this data, it will then suggest shows and movies that align with users' tastes.

Google says that Gemini will be rolling out later this year on select Google TV devices, including those from Sony, Hisense, TCL, and more.

Keep watching out the space for more CES 2025 coverage.

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every day for many years.

News posts... 2023: 5,800+ | 2024: 5,700+

RIP Matrix | Farewell my friend  

In a newly published blog post, Microsoft says 2025 is "the year of Windows 11 PC refresh." The company expects millions of users to migrate to Windows 11 and the newly launched Copilot+ PCs with AMD, Intel, and Qualcomm processors. However, the biggest push to Windows 11 is, of course, the end of Windows 10 support, which is still on schedule for October 14, 2025.

Whether the current PC needs a refresh, or it has security vulnerabilities that require the latest hardware-backed protection, now is the time to move forward with a new Windows 11 PC.

While some businesses and customers will remain on Windows 10 and use the Extended Security Program to get more support for the aging OS, Microsoft and analysts anticipate a big wave of updates to Windows 11. The software giant cites IDC's June 2024 report predicting that nearly 80% of businesses will update their PCs by the end of 2025 and that 70% of customers will update their PCs in the next two years.

In 2025, Microsoft promises to "continue to advance the category" and make AI accessible to more users. Microsoft says users can expect more AI-powered features (for example, Recall and Click to Do), security improvements, and other enhancements to the functionality of the operating system. Chip manufacturers, on the other hand, will deliver better performance and improved battery life. Finally, users will get to choose from a wide range of devices with "a price for everyone."

Are you planning to purchase a Copilot+ PC this year?

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every day for many years.

News posts... 2023: 5,800+ | 2024: 5,700+

RIP Matrix | Farewell my friend  

Pavan Davuluri, CVP of Windows + Devices at Microsoft, said the following about the Intel 200V series:

Copilot+ PCs offer exceptional performance, battery life, enhanced AI experiences, and
are all Secured-core PCs with the Microsoft Pluton security processor. Copilot+ PCs
powered by Intel Core Ultra 200V series deliver on all these fronts, and we are excited to
partner with Intel to bring a broad set of Copilot+ PCs to commercial audiences.

 

Intel Core Ultra 200V series Copilot+ PCs are an excellent choice for commercial customers looking to upgrade their existing Windows 10 PCs to Windows 11.

Today, the rest of the 200 series lineup, ie, the non-Copilot+ chips, has been revealed and they span across the premium enthusiast segment with the Core Ultra 200HX, the mainstream with the Ultra 200H, and the entry-level and efficient segment with the 200U series.

Intel summarised the major highlights of the 200H and 200HX series, including the features and performance (tested on Windows 11 24H2):

• Up to 24 cores for HX-series, including eight P-cores and 16 E-cores.
• Up to 16 cores for H-series, including six P-cores, eight E-cores, and two LP (low power) E-cores.
• Up to 41% better multi-thread (MT) performance for Intel Core Ultra 200HX series compared to last gen HX-series and 20% better MT vs Raptor Lake-H Refresh
• The Intel Core Ultra 200H series features Intel Arc graphics with up to eight Intel Xe cores and Intel Xe Matrix extensions (XMX) for AI acceleration.
• Up to 22% better gaming performance compared to last gen Meteor Lake-H, and delivering up to 99 combined TOPS (2.5 times better than Meteor Lake-H) when using GPU, CPU, and NPU.
• Intel Core Ultra 200HX series processor is Intel’s first mobile enthusiast AI PC with a built-in NPU, providing 13 TOPS.
• Provides up to 48 total PCIe lanes (including PCIe 4.0 and 5.0)
• Packaging improvements result in a 33% smaller processor package
• Select Intel Core Ultra 200HX SKUs offer overclocking (OC) options:
  • Compute OC for P-cores and E-cores.
  • Intel XMP support for DDR5 SODIMM OC.
  • New overclocking interfaces including die-to-die and fabric and controls like 16.6 MHz ratios.
  • Intel XTU one-click OC using the Intel Speed Optimizer feature.

Platform features are as follows:

• Greater networking speeds, responsiveness and reliability with Intel ® Wi-Fi 7 (5
  Gig) support, along with the fastest, simplest and most reliable USB-C connection
  with Thunderbolt ™ 5 or Thunderbolt™ 4 technology.
• Continuous AI-based network connection optimization with Intel ® Connectivity
  Performance Suite software.
• High-fidelity, low-power Intel ® Bluetooth® LE Audio.
• Ability to share screens, keyboard, mouse, storage and files with two PCs using
  Thunderbolt™ Share.

The detailed specs and platform features of each of the Core Ultra lineups are given in the images below (click to open):

Ultra 200V:

Ultra 200HX:

Ultra 200H:

Ultra 200U:

In terms of availability, Intel says that the 200V and 200H-based notebooks and laptops will be available starting this quarter (Q1 2025), while the 200HX gaming notebooks with discrete GPUs will land in late Q1.

Intel also added that systems with the Ultra 200S desktop Bartlett Lake S parts are also coming this Q1. Bartlett Lake S is different from Arrow Lake S (Ultra 200S) since the former is new and will work with LGA1700 sockets, unlike the existing Ultra 200S Arrow Lake S parts like the 285K that are LGA1851 compatible.

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every day for many years.

News posts... 2023: 5,800+ | 2024: 5,700+

RIP Matrix | Farewell my friend  

With its new lineup of TVs, Samsung is making a decision that I think might prove somewhat divisive. The company is bringing the matte, glare-free display technology that debuted on last year’s S95D OLED to several more models — including its flagship Mini LED sets. Here at CES 2025 in Las Vegas, Samsung is showcasing all of its latest TVs at its annual First Look event.

 

The new S95F QD-OLED gets blazingly bright, likely using a just-announced Samsung Display panel that’s technically capable of reaching 4,000 nits. According to the company, the glare-free coating has been improved to further cut down on reflections from overhead lighting, floor lamps, and sunlight. And similar to LG’s top-tier 2025 OLEDs announced earlier today, the S95F is capable of hitting a maximum refresh rate of 165Hz. PC gamers, rejoice.

 

Even Samsung’s First Look show floor, with bright lights everywhere, poses no issue for the glare-free screen.

 

It’s really quite impressive. And now Samsung is bringing it to more models. But not everyone likes the perceived tradeoffs.

 

If you’re wondering what’s so controversial about Samsung’s glare-free screen, some people insist it results in a worse overall picture than glossy coatings and that the perfect blacks of OLED aren’t so inky black in all lighting conditions anymore. This issue has been debated at length on Reddit, in YouTube videos (hey, Caleb), and all over AVS Forum.

 

But clearly Samsung remains undeterred by the haters because now the glare-free display is also coming to the company’s Mini LED “Neo QLED” TVs for the first time. That includes the flagship 4K QN90F and both of this year’s 8K models. (Buying an 8K TV is very silly; I still very much recommend against doing that.) Samsung’s other 2025 4K TVs will stick with a glossy treatment, so at least there are options if you refuse to go glare-free.

 

These latest Mini LED TVs are also available in some truly enormous sizes: the QN90F tops out at 115 inches, while the (glossy) QN80F can be had at up to 100 inches. Samsung says the wonderfully named “Supersize Picture Enhancer” will help keep 4K content looking crisp even on that giant QN90F. You still won’t find Dolby Vision on any of these TVs no matter the size; clearly, that’s a philosophical choice at this point.

 

The company’s 115-inch 4K Neo QLED TV uses a “Supersize Picture Enhancer” to boost clarity at this enormous size.

Vision AI

For 2025, Samsung is pulling all of its AI-powered TV features under new branding called Vision AI. These include the usual suspects like AI Upscaling, Auto HDR Remastering, and Adaptive Sound Pro. But there’s a new Click to Search feature that can identify actors on-screen, the location of a shot, or what clothes are featured in a scene “with just one click of the new AI button on your SolarCell remote.” Yes, there’s now a dedicated AI remote button.

 

Another new AI trick is Samsung Food, which “recognizes the food on your screen and provides recipes for bringing it to life.” I’m mildly curious about this and can’t wait to see how accurate or off the mark it is. Live Translate is a much more helpful addition: it can “instantly translate closed captions on live broadcasts in up to seven languages.”

 

The company is also using AI to provide more robust home security features. From tonight’s press release:

 

Samsung AI Home Security transforms your TV into a smart security hub. It analyzes video feeds from your connected cameras and audio from your TV’s microphone to provide comprehensive home monitoring.  

 

It can detect unusual sounds and movements, such as falls or break-ins, to give you more peace of mind whether you’re at home, or away.  

 

You’ll receive alerts and notifications on your phone or directly on your TV screen, helping you stay connected to your home while ensuring the safety and well-being of your loved ones.

Samsung is even leveling up Bixby, which isn’t something we’ve said in a long time. The company’s voice assistant can now “better understand context and assist with multiple actions — like changing the channel and raising the volume at the same time.” You can also now control your Samsung TV with the Galaxy Watch on your wrist. That’s got absolutely nothing to do with AI, but it might be convenient at times.

 

As for its lifestyle TVs, Samsung is announcing The Frame Pro, which you can read all about here. Pricing for all of these 2025 TVs will be announced over the next few months, and they’ll begin to ship this spring.

 

Photography by Chris Welch / The Verge

 

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every day for many years.

News posts... 2023: 5,800+ | 2024: 5,700+

RIP Matrix | Farewell my friend  

The Frame Pro uses Mini LEDs, but they’re at the bottom of the panel — not behind it.

There’s no more wire running from Samsung’s breakout box to the actual TV.

 

You plug your game consoles, streaming boxes, and other devices into the Wireless One Connect Box.

The Wireless One Connect Box can be placed up to 10 meters away.

 

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every day for many years.

News posts... 2023: 5,800+ | 2024: 5,700+

RIP Matrix | Farewell my friend  

Gamers Nexus (2.38M subscribers)

January 6, 2025

Video length: 29m 31s

00:00 - Recapping the Week

02:06 - Disappointment Build & The End of 2024

05:30 - NZXT Class Action Lawsuits

09:06 - Final RTX 50 Series Leaks and Rumors

13:40 - Final AMD Radeon 9070 Leaks

16:42 - OneXPlayer G1 Convertible Handheld

19:03 - Meta Plans to Unleash AI Users

24:31 - AI AI AIAIAIAI

25:38 - TrendForce Reports That DRAM Prices Dropping

27:01 - Half-Life Shows Half-Signs of Half Life

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every day for many years.

News posts... 2023: 5,800+ | 2024: 5,700+

RIP Matrix | Farewell my friend  

ExplainingComputers (1.07M subscribers)

January 6, 2025

Video length: 16m 47s

00:00 Titles & Intro

00:48 Card Sizes & Capacities

02:54 Speed Classes

05:15 Bus Interfaces

09:34 QuickFlow & UHS-II Tests

14:19 Reality Check

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every day for many years.

News posts... 2023: 5,800+ | 2024: 5,700+

RIP Matrix | Farewell my friend  

Table of contents:

1. Windows 10 and 11 news
2. Windows Insider Program
3. Updates are available
4. Reviews are in
5. Gaming news

Windows 11 and 10

Here we talk about everything happening around Microsoft's latest operating system in the Stable channel and preview builds: new features, removed features, controversies, bugs, interesting findings, and more. And of course, you may find a word or two about older but still supported versions.

Let us start the first Weekly of 2025 with a recap of some cool features Windows 11 received in 2024. From faster Windows updates to better context menus, here are ten nice things Microsoft implemented in its operating system over the course of the last year.

Now, moving to fresh stats. Statcounter reports that in December 2024, Windows 11's global market share dropped again and reached 34.1%. Windows 10, on the other hand, increased the number of its customers (check out the latest browser report here). On the gaming side, though, Windows 11 is doing much better. Valve says the operating system is installed on 54.96% of all Windows systems accessing Steam.

ESET, a popular AV provider, has some recommendations for those who cannot update from Windows 10 before its end of support later this year. The recommendation is simple: move to Linux.

¯\_(ツ)_/¯

However, if you indeed want to jump-ship out of curiosity or necessity, this article might be of help. This week, we published a list of ten apps switchers to Linux will find useful when changing operating systems.

Microsoft has also updated the official guidance on the "Operation is not supported" error when using MSU packages. Speaking of updates, this week, we posted a lengthy editorial discussing the need for slower development cycles to avoid frustration and bugs in modern software like browsers and operating systems.

Windows Insider Program

Microsoft is back to releasing weekly Windows 11 preview builds. This time, the company pushed two new builds: one for Canary users and one for Beta users. Both contain mostly changes, but the one for the Beta Channel backports the updated context menus from Windows 11 version 24H2 to version 23H2.

To finish this week's Windows section, here are some nice unreleased wallpapers that Microsoft planned to implement in Windows 11. Not only that but some wallpapers were supposed to be dynamic and animated to showcase some new parts of the operating system, like the updated Start menu. Sadly, the project was nuked, but wallpapers are still out there.

Updates are available

This section covers hardware, software, firmware, and other notable updates (released and coming soon) delivering new features, security fixes, improvements, patches, and more from Microsoft and third parties.

Incase is finally bringing back some of Microsoft-designed PC accessories. Earlier this week, the company announced the revival of the Microsoft Ergonomic Compact keyboard. The new version costs $119.99, and it offers some upgrades like a dedicated Copilot key if you are into that. Incase says the keyboard will be available later this year.

And here are the newest drivers released this week:

• Intel 32.0101.6449 WHQL and 32.0101.6256 WHQL with support for the upcoming Core N processors.

Reviews are in

Here is hardware and software we reviewed this week

Robbie Khan dropped a review of the Edifier M60: a pair of compact speakers with some seriously impressive sound for their size and price. There are some minor omissions, but the M60 overall is a fantastic set for those who want great sound at a solid price.

On the gaming side

Learn about upcoming game releases, Xbox rumors, new hardware, software updates, freebies, deals, discounts and more.

Minecraft 2, anyone? Notch, the maker of the original game that Microsoft bought in 2014, pitched the idea of a "spiritual successor" to Minecraft on his X and even said he was willing to partner with Microsoft to make it. Of course, at this point, these are just talks, but it is still interesting to see Notch wanting to return to the legendary blocky game.

If you wonder why there is no Black Myth: Wukong on Xbox, here is your answer: the game is delayed due to memory limitations of Xbox Series S. The game's director revealed on social media that the console's 10GB of shared memory was a bridge too far for developers. With the studio not having years of experience in optimizations, releasing the game just was not possible on Xbox Series S.

Nvidia kicked off 2025 with 14 new games for the GeForce NOW streaming service. You can now play Mirthwood and Slime Rancher 2, and 12 more games will be available later this month.

Finally, Valve revealed the Steam Awards 2024 winners in eleven categories like "Game of the Year," "Labor of Love," "Better With Friends," "Best Soundtrack," and more.

Deals and freebies

This week's Weekend PC Game Deals article is full of bundles and big discounts on popular and not-so-popular titles that are part of Steam's Winter Sale and Epic Games Store's holiday sale.

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every day for many years.

News posts... 2023: 5,800+ | 2024: 5,700+

RIP Matrix | Farewell my friend  

Nvidia’s RTX 5090 has leaked today in the form of a marketing image of the unannounced next-gen GPU. VideoCardz has obtained a box shot of the RTX 5090, which suggests that the rumors of 32GB of GDDR7 memory are true.

 

While the packaging of the unannounced Inno3D RTX 5090 iChill X3 doesn’t reveal more specs about Nvidia’s flagship next-gen GPU, it does suggest that this particular model will ship with a 3.5-slot cooler.

 

Inno3D’s RTX 5090 packaging.

Image: VideoCardz

 

The RTX 5090 is expected to have double the VRAM of the RTX 5080, which is rumored to include 16GB of GDDR7 memory. It’s also rumored to include 21,760 CUDA cores, nearly 1.8TB/s of memory bandwidth, and a TDP of 575 watts — 125 watts more than the RTX 4090.

 

We shouldn’t have long to wait until details about the RTX 50-series of GPUs are official. Nvidia is hosting a CES keynote tomorrow night, where the GPU maker is widely expected to announce its RTX 5090, RTX 5080, RTX 5070 Ti, RTX 5070, and even an RTX 5090D model for China. Rumors suggest the RTX 5080 could debut first on January 21st, followed by Nvidia’s other RTX 50-series cards.

 

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every day for many years.

News posts... 2023: 5,800+ | 2024: 5,700+

RIP Matrix | Farewell my friend  

The new year is a time for reflection, renewal, and rampant speculation about what wonders (or fresh hell) the future might hold. No place does this mix of anxiety and forward-looking techno-evangelism spring forth more profusely than at CES.

The giant consumer tech showcase is barreling down on Las Vegas starting January 7, bringing with it a whirlwind of fuss about the newest gadgets and devices. And yes, you bet all these things are going to be packed full of AI features. You’re probably going to be asked to wear many of them.

AI has dominated CES—and seemingly every facet of our culture—for the past couple of years. A tsunami of AI tech broke on the shores of CES 2024, and while the floodwaters may have receded a bit since then, the storm of hype around AI’s promise will keep the surge coming.

“You will certainly see AI wearables from the slew of vendors at CES,” says Jitesh Ubrani, research manager at the analyst firm IDC. “I don't think it's going to be similar to what we saw last year, where you had lots of dedicated hardware for AI. More so it will be AI being incorporated into existing devices or it being an additional feature of a device as opposed to the only thing that it’s good for.”

AI Wrapped

With most of the spoils of the current AI boom going to the industry's apex predators—OpenAI, Google, Meta—and their mature, well trained language models, startups who want to compete are focusing more on the physical layer of the user experience.

“You're not going to add any value by creating your own AI model,” says Anshel Sag, principal analyst at Moor Insights and Strategy. “So the next step is implementing the AI. The easiest way to do that is with some kind of hardware.”

We saw a lot of these AI-powered devices in 2024. Together they showed a variety of use cases, but they were mostly still just vessels for utilizing AI on something that wasn’t your phone or computer. Some of them licensed models like ChatGPT while others used bespoke software to power their interactions, but both paths have been rocky. Devices like Humane AI pin and Rabbit R1 kicked off the trend, though they were ultimately underwhelming. The wearable Friend necklace, which you interact with through a mobile app, garnered controversy for being a device with an always-listening microphone. Others have hinted at greater ambitions, like the Plaud.AI pin, which for now summarizes your meetings but could eventually be instructed to attend the meeting for you.

There are still many more dedicated AI devices to come. Some could be useful, and some are leaning on AI interactions just to appear forward-looking or drum up interest. Whatever the flavor of AI gadget, CES will be the place to stick, strap, slip, and clip them on. We will see the debut of new necklaces, eyewear, pins, and of course headphones. I've lost count of how many press pitches I have received ahead of the event from companies wanting to show off their earbuds built specifically for AI that let you interact with Her-esque chatbots. There are even sex toys. (I will spare you the content of those emails.)

Whether any of them will implement chatbots and agents well or in new and exciting ways is much harder to say. While the addition of AI may have been enough to garner the investment needed to build a device, it may not be enough to get people to actually buy the thing. Chatbots and AI agents don't yet provide enough of a use case to justify people pinning them to their shirts en masse. We’re also at a point of AI saturation where the tech is in everything. So then, what makes your AI earbuds special?

“That's the problem a lot of these startups have; if AI is their differentiator, then what happens when everybody has it?” Sag says. “It’s now table stakes.”

Wearables and devices built specifically to provide some AI-powered services may have seemed like the logical next step in the AI evolution, but so far the utility we’re getting from them doesn’t push any boundaries.

“The reality is we don't need dedicated hardware for the kind of features or use cases that they're showing off,” Ubrani says. “Your phone can do most of those things.”

In the space of a year, AI has gone from being a selling point on its own to something akin to a slightly more potent form of vanilla.

Making a Dent

There are AI hardware success stories, of course, such as the Ray-Ban Meta smart glasses, which have done well by incorporating AI as one of many features in a device that offers use cases—taking pictures, listening to music—well beyond what AI can do on its own. (This will certainly be a year filled with smart glasses, and CES is bound to brimming with them too.)

Meta, of course, is one of those giant companies with resources to put into incorporating AI into its services. Smaller manufacturers may not have the financial stamina to compete, but they're feeling the pressure to get in on the game all the same.

“It’s going to be difficult to see how those smaller startups survive,” Sag says.

Sag says there are ways to stand out from the big devices and glut of other AI gadgets in the mix. Privacy, for instance. Meta may have the most successful smart glasses right now, but the company’s platform is a data vacuum that sucks up almost every bit of information about its users that it can. Sag points toward competitors like Even Realities or Looktech.AI, which make smart glasses that allow broad user controls over privacy settings and don’t necessarily just send every bit of data back to the mothership. He says startups like those can use the more secure approach to differentiate their products, offering users an alternative to the big, data mining platforms.

No matter how safe and secure the tech is, people are still going to want something that fundamentally does something beneficial for them.

“The next kind of wave of this is like, well, what is AI doing for me right now other than telling me that I have AI?” Sag says. “A lot of AI isn't necessarily driving sales, because it's not really changing people’s lives.”

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every day for many years.

News posts... 2023: 5,800+ | 2024: 5,700+

RIP Matrix | Farewell my friend  

LG’s wireless TV tech is expanding to its QNED models this year.

Image: LG

The Honda Zero concept from CES 2024.

Image: Honda

A leaked image appears to show Lenovo’s SteamOS-powered handheld.

Image: Evan Blass (X)

The Xreal One takes a simple approach to AR by just mirroring another screen right in front of your eyes.

Image: Xreal

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every day for many years.

News posts... 2023: 5,800+ | 2024: 5,700+

RIP Matrix | Farewell my friend  

Fast forward to now, and Microsoft’s ready to invest a jaw-dropping $80 billion in AI-driven data centers in fiscal year 2025. As Brad Smith, Microsoft's Vice Chair & President, puts it:

As we look into the future, it’s clear that artificial intelligence is poised to become a world-changing GPT. AI promises to drive innovation and boost productivity in every sector of the economy. The United States is poised to stand at the forefront of this new technology wave, especially if it doubles down on its strengths and effectively partners internationally... In FY 2025, Microsoft is on track to invest approximately $80 billion to build out AI-enabled datacenters to train AI models and deploy AI and cloud-based applications around the world. More than half of this total investment will be in the United States, reflecting our commitment to this country and our confidence in the American economy.

This massive investment is aimed at keeping up with the explosive demand for AI, especially after ChatGPT launched in 2022. With companies scrambling to add AI to their products, the need for powerful computing resources is off the charts.

But Microsoft’s not just putting money into infrastructure. In November 2024, they launched an AI service that processes multimodal data, including text and images, all at once. It's a game-changer, giving AI a "supercharged brain" to generate and understand content across formats.

On top of that, Microsoft’s global strategy is expanding too. Last April, they invested $1.7 billion in AI and cloud development in Indonesia, building data centers and supporting local talent to tap into Southeast Asia’s growing tech scene.

Meanwhile, OpenAI is also on a roll. A while back, they wrapped up a $6.6 billion funding round, bringing their valuation to $157 billion. This will fuel even more cutting-edge AI research. With Microsoft’s strong ties to OpenAI, you can expect some serious AI advancements ahead.

That said, the AI rush isn’t without its challenges. Late last year, a study highlighted the significant strain AI data centers were placing on local electricity grids, sometimes even causing damage to nearby homes.

Image via Depositphotos.com

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every day for many years.

News posts... 2023: 5,800+ | 2024: 5,700+

RIP Matrix | Farewell my friend  

Nvidia is reportedly gearing up to launch the first few cards in its RTX 50-series at CES next week, including an RTX 5090, RTX 5080, RTX 5070 Ti, and RTX 5070. The 5090 will be of particular interest to performance-obsessed, money-is-no-object PC gaming fanatics since it's the first new GPU in over two years that can beat the performance of 2022's RTX 4090.

But boosted performance and slower advancements in chip manufacturing technology mean that the 5090's maximum power draw will far outstrip the 4090's, according to leakers. VideoCardz reports that the 5090's thermal design power (TDP) will be set at 575 W, up from 450 W for the already power-hungry RTX 4090. The RTX 5080's TDP is also increasing to 360 W, up from 320 W for the RTX 4080 Super.

That also puts the RTX 5090 close to the maximum power draw available over a single 12VHPWR connector, which is capable of delivering up to 600 W of power (though once you include the 75 W available via the PCI Express slot on your motherboard, the actual maximum possible power draw for a GPU with a single 12VHPWR connector is a slightly higher 675 W).

Higher peak power consumption doesn't necessarily mean that these cards will always draw more power during actual gaming than their 40-series counterparts. And their performance could be good enough that they could still be very efficient cards in terms of performance per watt.

But if you're considering an upgrade to an RTX 5090 and these power specs are accurate, you may need to consider an upgraded power supply along with your new graphics card. Nvidia recommends at least an 850 W power supply for the RTX 4090 to accommodate what the GPU needs while leaving enough power left over for the rest of the system. An additional 125 W bump suggests that Nvidia will recommend a 1,000 W power supply as the minimum for the 5090.

We'll probably know more about Nvidia's next-gen cards after its CES keynote, currently scheduled for 9:30 pm Eastern/6:30 pm Pacific on Monday, January 6.

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every day for many years.

News posts... 2023: 5,800+ | 2024: 5,700+

RIP Matrix | Farewell my friend  

According to the company, traditional monitors with 144 Hz, 240 Hz, and 480 Hz simply don't cut it anymore for esports players. With every frame mattering in fast-paced first-person shooters and multiplayer online battle arenas, there is a growing demand for monitors with even faster refresh rates, which KOORUI hopes to meet.

The monitor with this high refresh rate will be called the KOORUI G7, here are some of its core specifications shared by the company:

• Display Type: TN
• Screen Size: 24.5-inch
• Resolution: FHD
• Color Gamut: DCI-P3 95%
• Refresh Rate (Max.): 750Hz
• Response Time: 0.5ms
• HDR: HDR 400

TN panels like the one that comes on the G7 have a few notable flaws. They can suffer from limited viewing angles, have poor color accuracy, and typically have lower contrast. To compensate for the color defects, the G7 is equipped with the latest QD film with a wide color gamut solution, enabling color gamut of up to DCI-P3 95%.

Pricing and availability have not been shared by KOORUI, but we could hear more details at CES 2025 this month.

CES 2025 is due to kick off next Tuesday on January 7. It will run over four days, ending on January 10. The world will get to have a closer look at what technology companies from around the world have been working on before coming to market for sale.

Other companies prepping for CES 2025 are Samsung and LG. Samsung recently announced a lineup of monitors while LG unveiled a bendable monitor.

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every day for many years.

News posts... 2023: 5,800+ | 2024: 5,700+

RIP Matrix | Farewell my friend  

It seems like Nvidia might be about to reveal its RTX 50 series GPUs, and now Nvidia itself is teasing that the new graphics cards are nearly here. As spotted by VideoCardz, a video for Nvidia’s upcoming LAN party includes brief look at a shadowy PC that will be given as a prize, and if you brighten up the image, the GPU in the computer appears to be one that’s a new design.

 

See for yourself — I’ve included an image of the shadowed version at the top of this post, but here’s our take at tweaking the image to reveal what’s hidden:

 

Image: Tom Warren / The Verge

 

I’ll admit I am not an expert in GPU designs, but VideoCardz says the design is “unfamiliar” and that an RTX 5080 or 5070 Ti could be “among possible candidates” of what’s being shown.

 

Leaks last month from Zotac and Acer indicated that Nvidia may reveal as many as five RTX 50 series GPUs very soon, including an RTX 5090, an RTX 5080, and an 5070 Ti. VideoCardz also reported today on a leak of an RTX 5080 from MSI featuring 16GB of GDDR7 memory.

 

And Nvidia isn’t being particularly secretive about the potentially imminent launch of the RTX 50-series GPUs. The LAN party, which runs from January 4th through 6th, is called the “GeForce LAN 50”, and Nvidia CEO Jensen Huang is set to deliver a CES keynote on January 6th at 9:30PM ET.

 

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every day for many years.

News posts... 2023: 5,800+ | 2024: 5,700+

RIP Matrix | Farewell my friend  

The more than two decades since Half-Life 2's release have been filled with plenty of rumors and hints about Half-Life 3, ranging from the official-ish to the thin to the downright misleading. As we head into 2025, though, we're approaching something close to a critical mass of rumors and leaks suggesting that Half-Life 3 is really in the works this time, and could be officially announced in the coming months.

The latest tease came just before the end of 2024 via a New Year's Eve social media video from G-Man voice actor Mike Shapiro. In the voice of the mysterious in-game bureaucrat, Shapiro expresses hopes that "the next quarter century [will] deliver as many unexpected surprises as did the millennium's first (emphasis added)... See you in the new year."

The post is all the more notable because it's Shapiro's first in over four years, when he concluded a flurry of promotional posts surrounding the release of Half-Life: Alyx (many of which were in-character as G-Man). And in 2020, just after Alyx's release, Shapiro told USGamer that he had recently worked on a "blast from the past" project that he would "announce... on my Twitter feed when I'm allowed to" (no such announcement has been forthcoming for any other game).

"I was working on that game for quite a while before I knew [what it was]," Shapiro said at the time of the unannounced project. "There was a rehearsal and some recordings, and after one of the recording sessions I was having a drink with the director. He told me what the game was, and nobody knows that this is coming.

"This is going to be such a mindblowing re-up from what people have come to know," Shapiro continued in 2020. "It's going to really... it's going to make people have a complete re-understanding of what they thought they knew about the story in the game prior to it, and I don't even know if people are expecting it."

Raised HLX-pectations

On its own, a single in-character post from a voice actor would probably be a bit too cryptic to excite Half-Life fans who have seen their sequel hopes dashed so often over the last two decades. But the unexpected tease comes amid a wave of leaks and rumors surrounding "HLX," an internal Valve project that has been referenced in a number of other Source 2 engine game files recently.

Those HLX references got a lot more attention about five months ago when noted Valve watcher Tyler McVicker posted a video analyzing the Valve code and concluding that the HLX project is "a fully-fledged non-VR Half-Life game." In subsequent videos, McVicker has gone into more detail on datamined code for everything from voxel-based deformation systems to zero-g and underwater navigation systems to "so much flammability stuff," all seemingly in service of "HLX."

"It's obvious that Valve wants to be able to push the envelope in the same way that Half-Life 1 and Half-Life 2 did in their day," McVicker said in November in a video bluntly entitled "Valve Isn't Trying to Hide HL3 Anymore." McVicker now believes that "instead of pushing graphical fidelity like everybody else, it seems like Valve is pushing for physical fidelity" with its in-game systems and objects.

Beyond those data-mined details, Valve-watching YouTuber "Gabe Follower" said in a video this week that "according to my information, in the last few months Valve has started active HLX playtests, now not only with people from other teams, but also friends and family members." That suggestion gains at least a little bit of extra credence given Gabe Follower's history of leaking early Valve content, including screenshots and information about Deadlock (originally code-named Citadel) well before it was officially revealed.

"It is a quite important stage for the project," Gabe Follower says of the reported friends and family playtesting. "If it survives the New Year holidays and there are no serious problems during the playtests with a close circle of people, then the development will continue at the right pace and the announcement in 2025 may turn out to be quite possible."

Add that to leaks regarding a mysterious "Project White Sands" that Valve seems to be using in public hiring announcements, and you have enough smoke to suggest there might actually be some real Half-Life 3 fire burning at Valve this time around. We know it's easy to be gun-shy after so many years of false starts, but we hope you'll forgive us for still wanting to believe.

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every day for many years.

News posts... 2023: 5,800+ | 2024: 5,700+

RIP Matrix | Farewell my friend  

Silicon chip manufacturers like Intel and TSMC are constantly outdoing themselves to make ever smaller features, but they are getting closer to the physical limits of silicon.

“We already have very, very high density in silicon-based architectures where silicon performance degrades sharply,” said Ki Seok Kim, a scientist working at the Massachusetts Institute of Technology’s Research Laboratory of Electronics.

One way around this problem is to replace silicon with graphene-like 2D materials that maintain their semiconducting properties even at a single-atom scale. Another way is building 3D chips, which squeeze more transistors into the same area without making transistors smaller. Kim’s team did both, building a 3D chip out of vertically stacked 2D semiconductors.

Coming in hot

Graphene, a single-atom-thin sheet of carbon, is probably the most famous 2D material, but it's not a semiconductor. There are 2D materials that are good semiconductors, though, like molybdenum disulfide or tungsten diselenide. “They offer very stable electrical performance even below one nanometer,” Kim said. Both molybdenum disulphide and tungsten diselenide belong to a group called transition metal dichalcogenides (TMDs).

These materials are usually grown via a process called chemical vapor deposition, where vaporized materials are sprayed over a substrate and form single-atom-thin crystalline structures on its surface. To make decently performing transistors, those TMDs had to be deposited at temperatures reaching 900º Celsius. That was fine if you wanted just one layer of transistors, but it isn’t compatible with normal silicon fabrication techniques.

The chip manufacturing process starts with a round silicon wafer. Transistors are manufactured on those wafers directly and form the bottom layer of a chip in a part of the chip fabrication process called the front end of line. The first layer of metal wiring, or interconnects, is then added on top of the transistor layer; this is called the back end of line.

If you wanted to add a layer of TMD transistors on top of the back end of line to build a 3D stacked chip, you’d need to heat the whole thing to 900° Celsius, which would basically fry the circuitry. Researchers looking into building 3D chips proposed various solutions to this circuitry-frying problem, but none seemed to work particularly well.

The most popular approach was a process called the through-silicon-via (TSV), which was used to manufacture layers of transistors on separate silicon wafers with drilled microscopic holes to connect the upper wafer with the bottom wafer. But the process was very expensive, and aligning the wafers carefully enough to connect nanometer-scale devices was a challenge. An alternative was to transfer transistors grown on a separate wafer, which solved the hole-drilling issue but left the alignment problem.

Scientists also tried to do chemical vapor deposition of TMDs at temperatures below 400° Celsius, which was considered safe for the metal circuits. But at these low temperatures, TMDs formed poly-crystalline materials instead of single-crystal structures, which significantly degraded their electrical performance.

“The goal was to deposit single-crystalline TMD transistors directly on top of the back end of line in temperatures below 400° Celsius,” Kim said. “This is exactly what we have achieved.”

Nanoscale metallurgy

The solution proposed by Kim’s team was inspired by metallurgy. When a molten metal is poured into a mold, it slowly forms what are called nuclei: grains of solid material that merge to create regular crystal patterns that further harden into solid form. Kim and his colleagues noticed that this nucleation process usually started at the edges of the mold. “Nucleating at the edges requires less energy and heat,” Kim said. So the researchers borrowed this concept and used it to create single-crystalline TMD transistors. They called this technique “geometric confinement.”

Just like with standard chip manufacturing, the process started with a silicon wafer, which was then covered with a thin insulator layer of hafnium oxide. On top of that layer, the team added a polymer coating. That polymer layer was then shaped into trenches—rectangular pockets that were then covered with silicon dioxide. Tungsten diselenide vapor was sprayed over this array of trenches at 485° C as part of the chemical vapor deposition process. This is where the magic happened.

Normally, crystallization forms a polycrystalline material. In a nucleation event, a single-crystalline structure is formed, but as the process goes on, a second nucleation site will rapidly form, and then another, until the whole bulk of material becomes crystallized. The crystalline structures formed in each of those nucleation events are separated by grain boundaries, the signature of a polycrystalline material.

Kim’s team fine-tuned the size of the trenches so that the entire trench could be filled with a crystalline structure formed in a single nucleation event. The crystallization in each trench started with a nucleation event at its edge, just as it would in metal molds. The crystalline structure then filled the trench and ran out of available space just before the second nucleation event could happen.

This formed the first single-crystalline 2D semiconducting layer, which was then turned into transistors by depositing platinum source and drain regions topped with the gate and isolated with another layer of hafnium oxide. That completed the first back end of line.

The second layer of transistors was grown directly on top of the first one using the same process, with two differences. The semiconducting TMD was molybdenum disulfide, which could be deposited at 385° C, and the material used for source and drain regions was chromium. “The electrical performance we achieved was excellent, similar to high-temperature grown TMDs. We demonstrated monolithic, vertically stacked CMOS made with single-crystalline TMDs at below 400° C temperature for the first time,” Kim said.

Getting the temperature of TMDs growth down enough to make large-scale manufacturing feasible could be a huge leap toward making 3D stacked CMOS with 2D semiconductors in actual products people can use. Five members of Kim’s team worked at Samsung’s Device Research Center in South Korea, which might make it look like we’re on the verge of getting this technology to market. The problem is that we need at least one more similar leap before we say goodbye to silicon chips.

Doping problem

Currently, we don’t know how to connect TMD semiconductors with other devices. In today’s chips, connections are achieved by doping: injecting the silicon with impurities to increase its conductivity precisely at points where it interfaces with metal wiring. Those injected impurities smooth out the conductivity drop when the signal is moving between semiconductors and highly conductive materials like chromium, copper, or platinum.

At an atomic scale, doping substitutes atoms of a host metal with foreign atoms. But how do you do that when your host material is one atom thin?

There have been attempts to achieve effects somewhat equivalent to doping in 2D molybdenum disulphide, but the process was difficult to control. In Kim’s paper, the team wrote that developing a doping process for TMDs that can be done below 400° Celsius is necessary to build high-performing chips based on 2D semiconductors. The team thinks its technology may help in growing doped TMDs, but so far, we have no way to do so.

Another looming issue is cooling. Getting the heat out of a super densely packed chip is hard enough with a single layer of transistors. Stacking multiple such layers on top of each other should make the issue considerably worse. “A heat sink area is needed in devices like that. This is also something we plan to do in the future—developing a new cooling scheme for such chips,” Kim said.

But Kim argues that these challenges are worth tackling, as he feels 3D stacked chips based on 2D semiconductors will dramatically increase the performance we can get out of the same chip area and significantly lower the power consumption compared to standard CMOS electronics. And all that will be needed for powering future AI systems.

“We will achieve very, very high-density AI chips,” Kim said.

Nature, 2024.  DOI: 10.1038/s41586-024-08236-9

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every day for many years.

News posts... 2023: 5,800+ | 2024: 5,700+

RIP Matrix | Farewell my friend  

27-inch 4K OLED 240Hz monitors seem to be like buses: you wait ages for one and then three turn up at once. Asus, Samsung, and MSI are all announcing the industry’s next-generation QD-OLED gaming monitors that offer the benefits of 4K OLED 240Hz panels at the smaller 27-inch size instead of 32 inches.

 

All three appear to be using the same fourth generation QD-OLED panel from Samsung Display, which Asus says offers “a longer lifespan over previous-gen OLEDs.” Both the Asus ROG Swift OLED PG27UCDM and the MSI MPG 272URX QD-OLED (who names these things?) include DisplayPort 2.1a (UHBR20), which offers 80Gbps of bandwidth to support 4K at 240Hz without the need for Display Stream Compression (DSC). Samsung’s press release about its Odyssey OLED G8 (G81SF) doesn’t mention DisplayPort 2.1a compatibility, but it’s reasonable to assume it’s part of the spec list.

 

Asus’ latest OLED monitor has a 26.5-inch viewable display.

Image: Asus

 

Samsung’s Odyssey G8 now comes in a 27-inch OLED 4K 240Hz variant.

Image: Samsung

 

MSI and Asus’ models both support DisplayHDR True Black 400, and Asus also supports Dolby Vision HDR. Both MSI and Asus are offering a three-year warranty that includes burn-in protection, but Samsung hasn’t confirmed its warranty situation for its latest G8 model. Samsung also hasn’t fully detailed the specs of its latest G8 OLED model, but it’s reasonable to assume it will support DisplayHDR True Black 400 at the minimum.

 

Interestingly, Asus’ model only has a 26.5-inch viewable display, but both MSI and Samsung are marketing their monitors as 27-inch ones. With all three offering the 0.03ms response times of OLED, 240Hz refresh rates, and above 160PPI for improved text clarity, the choice will really come down to design, features, and pricing. Unfortunately, Samsung, Asus, and MSI haven’t announced release dates or pricing yet.

 

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every day for many years.

News posts... 2023: 5,800+ | 2024: 5,700+

RIP Matrix | Farewell my friend  

Just a few days after LG announced its CES 2025 lineup of monitors, Samsung is doing the same. The company just introduced several new models, and perhaps the most impressive among them is the Odyssey OLED G81SF. It’s a 27-inch 4K monitor with a maximum refresh rate of 240Hz, 0.03ms response time, a glare-free display, and rear-core lighting with 52 color options. Samsung is also including a number of burn-in protection measures to ensure that the Odyssey G81SF’s screen looks pristine for years to come. The G8 is likely to be using the latest and greatest OLED panel from Samsung Display.

 

Next up is the Odyssey OLED G60SF — also a 27-inch OLED monitor — with a QHD resolution and an impressive 500Hz refresh rate that Samsung says effectively “eliminates lag and motion blur for ultra-smooth gameplay during critical moments.”

 

In 2025, Samsung is also finally shipping the Odyssey 3D monitor it first teased at least year’s CES. It lets you experience 3D visuals without any special glasses by using a lenticular lens on the front panel. “Eye tracking monitors the movement of both eyes using a built-in stereo camera, while view mapping continuously adjusts the image to enhance depth perception,” Samsung said in its press release. The Odyssey 3D only comes in a 27-inch 4K size; the larger 37-inch display mentioned last year has apparently been scrapped. The company says it offers “a rapid 1ms gray-to-gray response time, and a 165Hz refresh rate.”

 

Samsung’s Smart Monitor series is making the switch to OLED.

Image: Samsung

 

That does it for the gaming-focused monitors, but Samsung is also announcing two others that are more focused on everyday productivity. The SmartMonitor series is finally going OLED with the new Smart Monitor M9. And this year, Samsung is throwing in a ton of AI capabilities including AI Picture Optimizer. Like on its TVs, this feature “analyzes input signals to determine the type of content being viewed — such as gaming, video, or productivity applications — and automatically adjust the display settings for the best visual experience.” This also works for gaming and can detect the genre of what you’re playing.

 

AI Upscaling Pro can make lower-res content look crisper at the M9’s 4K resolution, and the monitor has a maximum refresh rate of 165Hz. That’s a significant upgrade from the M8, which topped out at 60Hz. Samsung says moving to OLED also allows for a new “ultra-slim” design for the Smart Monitor M9 that should take up considerably less desk space.

 

The last monitor being announced (at least for today) is the new 37-inch ViewFinity S8. It’s got a 16:9 aspect ratio, built-in KVM switch, 90W USB-C passthrough charging, and covers 99 percent of the sRGB color gamut. Peak brightness for this 4K LCD display tops out at 350 nits, which should be perfectly suitable for anything you’ll be doing at a desk.

 

Pricing and release dates are still to come, but The Verge will be in Las Vegas for CES 2025 in a matter of days. So you won’t have to wait much longer for some first-hand impressions of Samsung’s new monitors.

 

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every day for many years.

News posts... 2023: 5,800+ | 2024: 5,700+

RIP Matrix | Farewell my friend  

The search giant released many new products, features, and changes in the generative AI space. Let's recap major Google Gemini features and changes introduced throughout the year. In your free time, check out various products and services Google discontinued in 2024 and the Instagram features wishlist.

Note: The list is not exhaustive and may not include all the Gemini features released in 2024.

Renaming Bard chatbot to Gemini

One of the biggest changes we saw earlier this year was Google rebranding Bard to Gemini, bringing the naming scheme in sync with its pre-existing Gemini models. The company also tagged the freshly-baked Gemini 1.0 Pro model along with it and introduced Gemini in over 40+ languages and 230 countries globally.

A Google engineer later revealed "Why Gemini is called Gemini" and said that the AI model has a two-fold meaning behind its name. Its ability to handle different data types aligns with the dual-natured personality of the Gemini zodiac sign. The name also draws inspiration from NASA's early moonshot program called Project Gemini.

Mobile apps, paid version of Gemini, support for Chromebook Plus

Google also introduced the Gemini app for Android in February, eventually replacing Google Assistant as the default voice assistant for Android. At the time, iOS users could access the Gemini chatbot via the Google app.

In the same month, the search giant launched a paid subscription for its AI chatbot called Gemini Advanced, which gives users access to the most capable Gemini models, such as Gemini Ultra 1.0, 1.5 Pro, and experimental stuff such as Gemini-Exp-1206.

Furthermore, it brought various Gemini features, such as "Help Me Write," to the Chromebook Plus laptops, accessible via a dedicated Gemini button on the home screen app shelf.

Support for Google Maps

The capabilities of the Gemini AI chatbot can be enhanced with the help of Gemini extensions. In March, Google updated Gemini with support for Google Maps, allowing users to issue commands to trigger navigation from within the chatbot.

For instance, you can issue commands like "Navigate me to [X]," Gemini will display data such as distance, estimated time, and a link to Google Maps. The navigation will automatically start after a few seconds.

Vids

As the name suggests, Vids is a new Gemini-powered video creation tool Google launched in April this year to easily create training videos, presentations, updates, or marketing content. It offers a timeline-style user interface to assemble video assets from Google Drive or elsewhere, record voiceovers, or film yourself directly from the app.

You can collaborate with others and control who can edit, comment, or view your videos. Note that Google Vids is part of the Workspace suite and is a paid offering.

YouTube Music extension

Another add-on for the chatbot was a new YouTube Music extension in May. It allows Gemini users to connect YouTube Music with the chatbot to search their favourite songs, play radio, discover new artists and playlists, and more.

Gemini version updates and new models

The year 2024 saw several updates to Gemini models as well. Gemini 1.5 Flash was introduced in May as a lightweight LLM, which excelled "at summarization, chat applications, image and video captioning, data extraction from long documents and tables, and more."

This was followed by a smaller variant of 1.5 Flash (known as Gemini 1.5 Flash-8B); a new Gemini 1.5 Pro model with improved coding and complex prompt performance; and a new Gemini 1.5 Flash model with significant improvements.

Last month marked the debut of Gemini 2.0. Google announced the experimental Gemini 2.0 Flash model with support for natively generated images mixed with text and steerable text-to-speech (TTS) multilingual audio.

Ask Photos assistant

Google I/O 2024's limelight might have shadowed the Ask Photos assistant for Google Photos announced at the conference. The digital assistant powered by Gemini AI models can be used to pull photos/videos from your gallery, generate personalized captions, explain what's going on in them, or create trip highlights after a vacation.

Expanding Gemini to the education sector

In May, Google also expanded Gemini for Workspace into the education sector by introducing two new add-ons: Gemini Education and Gemini Education Premium. The latter allowed access to features like AI-powered note-taking and summaries in Meet and AI-enhanced data loss prevention.

Gemini side panels for Workspace apps

Google's pursuit of putting its AI chatbot everywhere possible paved the road for Gemini side panels for Workspace apps. In June, the search giant launched Gemini side panels for its web apps, including Google Docs, Slides, Drive, and Gmail.

The side panels are customized according to the app. For instance, Gemini in Gmail can summarize emails, suggest responses, and summarize email threads. Meanwhile, it can generate new presentation slides in Google Slides and create custom images.

In November, the Gemini side panel was also added to Google Chat. Like other apps, it can summarize conversations and save users from the pain of reading the entire thing.

Gemini Live

In August, Gemini Live was launched at the Pixel hardware event to give smartphone users a natural conversation experience with the AI chatbot. You can talk to Gemini like a normal back-and-forth chat, leave the conversation at any time, and continue it later.

You can also continue the discussion with Gemini while the app is in the background or your phone is locked. Gemini Live was initially included with Gemini Advanced, but now it's available through the Gemini app on Android and iOS. The feature was later expanded to add support for over 40 languages.

Custom Gems for Gemini

Custom Gems lets you create versions of the Gemini chatbot tailored for specific use cases. For instance, you can brainstorm ideas for an upcoming event, remember detailed instructions, or create a learning coach.

The premium feature was rolled out to Gemini Advanced, Business, and Enterprise users in over 150 countries. You can go to the Gem manager to access pre-made gems or create new ones by clicking the "+ Add Gem" button and typing its purpose and goal.

Imagen 3 and Whisk image generator

Google's highest-quality text-to-image model, Imagen 3, was launched in October following its announcement earlier this year. The latest image generation model works with Gemini apps for users in all supported languages.

Imagen 3 better understands natural language and the intent behind the prompt. Users can write prompts to generate photorealistic landscapes, textured oil paintings or whimsical claymation scenes, and issue subsequent commands to refine it further.

Earlier this month, Google used Imagen 3 to power a new generative AI tool called Whisk, which takes images as input to generate new images.

Gemini in Opera browser, Snapchat integration

Opera joined hands with Google Cloud to leverage Gemini's capabilities for its in-browser AI known as Aria, which already utilizes the power of multiple large language models (LLMs). The partnership brought Google's text-to-voice and image generation capabilities to the Opera browser.

Meanwhile, Snapchat was another third-party platform that gained access to Gemini's features in 2024. The social media giant partnered with Google to make the My AI chatbot smarter than before and offer a multimodal experience. The search giant claimed that Gemini's integration with My AI improved user engagement within the US by 2.5x.

Deep Research

If you use Google's generative AI chatbot for extensive research sessions, a new AI-powered assistant called Deep Research can help with your exploration. It is designed to cut down hours of research by doing in-depth document analysis, document summarization, and pulling key insights from large amounts of information.

Note that Deep Research is a part of Gemini Advanced (a paid offering) and is available in over 45+ languages in 150+ countries globally.

Natural language search in Google Maps

In a new update to its navigation app, Google added a Gemini-powered search feature to Google Maps. In other words, you can type natural language queries in the Google Maps search bar. For instance, you can find new places by typing "things to do with friends at night." The chatbot creates summarized reviews about places you read to get an idea about them.

Play songs from Spotify

Not just YouTube Music, Gemini also added support for the rival music streaming app Spotify. The AI chatbot can play Spotify music on your Android device without the need for any app-switching.

You can use the Gemini chat interface to request songs, browser playlists, and search for music through lyrics, among various features. However, you need a Spotify Premium account for the integration to work.

Gemini controversies

It's not just new features, Google's generative AI chatbot has landed in several controversies as well. In February, Gemini's image generation feature was in hot waters after showing bias in generating images of people. As a result, the image generator was temporarily halted, and Google had to explain the unexpected behavior.

An X user claimed Gemini summarized a PDF file automatically and without explicitly being asked to, noting that the required setting was already disabled.

Another report claimed that Google hires contractors to rate Gemini's responses based on qualities like accuracy, clarity, and safety and compare the responses with those from Anthropic's LLM Claude. The contractors noticed similarities between the responses from the two platforms. While comparing models isn't uncommon, the report didn't receive a satisfactory response from Google.

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every day for many years.

News posts... 2023: 5,800+ | 2024: 5,700+

RIP Matrix | Farewell my friend