Feed aggregator

Some important mortgage rates ticked up. But rate cuts from the Fed should help mortgage rates fall in the long term.

• Google Vids rolls out to Workspace users worldwide
• Vids will allow users to quickly and easily create work-focused video content
• Gemini AI helps with theme, design, text - and even voiceovers

Creating video content at work is about to get an AI boost thanks to a new tool launching on Google Workspace.

The company has today announced the launch of Google Vids, which utilizes its Gemini AI platform to let users create business-focused content from within their browser.

First revealed at Google Cloud Next 24 in April 2024, Vids is described as "an AI-powered video creation app for work", and will now be available to more Workspace plans.

Google Vids rollout

"We know that creating and editing video has felt challenging because in the past it has required special skills or licensing, but with Google Vids the power of storytelling is in your hands," a Google Workspace update blog post announcing the launch said.

"Now teams can build and tell compelling stories at work, scaling and landing their message to engage coworkers in an effective way."

Going forward, Vids will form part of the wider Google Workspace portfolio, sitting alongside existing apps such as Docs, Slides and Sheets, with full interoperability between all.

Users can pick from a number of existing templates, or use Gemini's "Help Me Create" tool to generate a "fully editable initial storyboard" using a Google Doc.

Once users have picked the visual style they want, Vids will create a first video draft using suggested scenes from millions of high-quality stock videos and images, alongside suggestions for text, scripts per scene, and even background music.

Users can then pick from animations, transitions, and photo effects, and even add in their own media from Google Drive or Photos, allowing companies to easily add in branding.

Vids will even add an AI voiceover, with a number of preset Gemini options available, or offer users the chance to record their own by reading along to a teleprompter, add a video recording, or screen recording of slides or other assets.

Google Workspace highlighted a wide range of possible use cases for Vids, from customer service teams creating better support by adding videos to help center articles, to training terms delivering company-wide guidance, marketers using video to share recaps of key campaigns and events, and project management teams sharing meeting recaps, business updates and report share outs.

Google Vids is now rolling out to Workspace users on tiers including Business Standard, Business Plus, Enterprise Standard, Enterprise Plus, Essentials, Enterprise Essentials and Enterprise Essentials Plus members.

You might also like

• We've also rounded up the best video editing software for beginners right now
• These are the best free office software choices we've found
• Lots more AI could be coming to Google Workspace soon

• Apple’s High Power Mode ramps up cooling to enable better performance
• It’s now available in more Macs than before
• Limited testing suggests its benefits might not be enormous

We already know that the M4 Pro chip inside Apple’s latest Macs is impressively powerful – it blows holes in the Mac Pro’s M2 Ultra, after all. But you can eke even more juice out of this little chip with a quick software tweak, and it’s now available in a larger range of Macs than ever before.

Dubbed High Power Mode, the setting in macOS Sequoia ramps up your Mac’s fans, providing better cooling to the machine. That in turn allows it to max out its performance, which is particularly useful in sustained workloads like video rendering and color grading, where increased cooling and performance can noticeably shorten working times.

High Power Mode is available in the MacBook Pro and the Mac mini, provided they’re outfitted with the M4 Pro chip. This is an improvement on the previous situation, where only the MacBook Pro with M3 Max, M2 Max or M1 Max chips could use this setting. It means that not only is the barrier to entry lowered in terms of the chip you need, but it’s also now available in the Mac mini for the first time.

(Image credit: Apple)

However, you might not find that High Power Mode provides a huge boost in performance. Ars Technica tested the setting in the Mac mini, where reviewer Andrew Cunningham said that the gains were “essentially negligible” yet came with “considerably increased fan noise.”

During our time testing the new M4 Pro MacBooks, we also found that turning on High Power Mode made only a little difference in benchmark results, and certainly nothing too mind-blowing.

But before you write the feature off entirely, Cunningham did acknowledge that his tests were brief and that you might see a larger effect over a longer period of time. That chimes with Apple’s note that High Power Mode is designed for “sustained” or “intensive” workloads.

We would be very surprised to see Apple include a feature like High Power Mode if it provides no significant benefits while also bringing a noisy drawback, and it’s likely you’ll see some improvement while the mode is enabled for more sustained workloads. But without testing it ourselves, we can’t be absolutely sure.

Still, it’s good to see this feature extended to more affordable Macs. It means that you can get High Power Mode for $1,399 (in the M4 Pro Mac mini) as opposed to the previous starting price of $3,199 for the M3 Max MacBook Pro. Regardless of how much of a performance boost it provides, that’s nothing to sniff at.

You might also like

• Apple M4 Pro: everything we know about Apple's midrange SoC
• Apple's M4 Pro chip is so powerful it puts the Mac Pro to shame
• Apple MacBook Pro 14-inch M4 (2024) review: one of the best Pro laptops around just got better

• New report shows ongoing need for human intelligence in cybersecurity
• AI continues to dominate the threat landscape
• Higher pay is the biggest motivator for security researchers

A new report from HackerOne has claimed almost half (48%) of security leaders believe AI is one of the most significant threats to their organization, and nearly 10% of researchers now specialize in AI in order to meet growing demand for AI expertise.

The top vulnerability reported to a bug bounty program was Cross-site Scripting (XSS), and for pentest this was misconfiguration emails. Bug bounty programs tend to focus on real-world attack vectors and user-level issues, whilst pentests focus on discovering architectural and systemic weaknesses.

AI is continuing to dominate the conversation and the landscape, with 55% of all reported vulnerabilities being AI safety issues, and a staggering 171% rise in AI assets in scope on the HackerOne platform.

Tech industries do better

The organizations most successful in actively reducing common vulnerabilities were security mature and tech-focused industries like e-commerce and online services. In fact, Web3 companies had 65% fewer reports for XSS than the average.

Motivation for security researchers is primarily financial, with 77% stating potential income as a driving factor, followed closely by the opportunity to learn new skills (64%).

Most respondents (67%) believe unbiased reviews by humans are the most effective tactic in uncovering AI security and safety issues. AI's tendency for bias means that human intelligence will always be needed in cybersecurity.

“Even the most sophisticated automation can’t match the ingenuity of human intelligence. The 2024 Hacker-Powered Security Report proves how essential human expertise is in addressing the unique challenges posed by AI and other emerging technologies,” said Chris Evans, HackerOne CISO and Chief Hacking Officer.

Humans excel in manual exploitation and reconnaissance, which are two hacking skills which require human creativity, such as spotting unique weaknesses from an outsiders point of view, or uncovering overlooked domains.

You might also like

• Take a look at our pick of the best AI tools around
• Microsoft reveals major Chinese botnet is attacking users across the world
• Check out our choices for best malware removal

These are the cortisol imbalance symptoms to watch out for, along with doctors' treatment advice.

• Three in 1,000 bankers click on a phishing link every month, report claims
• Russia is believed to be the leading force behind banking attacks
• Banks are wary about new tech like GenAI, which does go some way to helping

New research has claimed a concerning amount of banking employees click on a phishing link each month, making it one of the most common threats in the sector.

A report from Netskope found three in every 1,000 employees would fall victim to such scams, so of the estimated 362,000 banking employees in the UK in 2023, this equates to more than 1,000 workers clicking a dodgy link sent to them.

According to the report, hackers are getting a lot of their success from designing phishing pages to mimic the target bank’s website, stealing bank account information and login credentials to commit fraud.

Phishing is rife in banking

Netskope identified Downloader.SLoad (Starslord), Infostealer.AgentTesla, Trojan.FakeUpdater, Trojan.Parrottds and Trojan.Valyria as the most recently used malware families to be aware of, highlighting that Russian criminal groups are the most likely to target this industry.

Despite the serious threat of phishing attacks, banks were found to be more hesitant about adopting new technologies than other industries, with 87% of banks using generative AI compared with the cross-industry average of 97%. More than half of banks also use Data Loss Prevention measures to manage data going into GenAI apps.

"[Banks] are more aggressive at blocking apps without a legitimate business purpose and using DLP to control what can be sent to allowed apps," commented Netskope Threat Labs Director, Ray Canzanese.

The company’s advice to banks, as well as all other industries, is to inspect all HTTP and HTTPS downloads to prevent malware from infiltrating a network. Companies should also ensure that high-risk file types are thoroughly inspected with status and dynamic analysis.

Other basic internet hygiene practices can be adopted by workers, such as questioning the authenticity of any emails they receive and taking part in training campaigns. Boosted protection, like the use of multi-factor authentication and passkeys, can also help to prevent unwarranted access to accounts.

You might also like

• Check out the best password managers
• Keep your online footprint safe by using the best VPNs
• Google Cloud is making multi-factor authentication mandatory for all users

This holiday season, get something meaningful for your in-laws. They'll love and cherish these thoughtful gifts forever.

Shopping for a fitness buff this holiday season? Whether you're shopping for a workout warrior or a fitness newbie, we've got you covered. Here are our editor-approved best fitness gifts in 2024.

Be the talk of your next White Elephant gift exchange by showing up with an excellent gift. Here you’ll find our editor-approved picks for the best White Elephant gifts that people will want to steal.

The SpaceX rocket will launch during the late afternoon so its descent into the Indian Ocean is visible.

This review first appeared in issue 360 of PC Pro.

WatchGuard’s Firebox M390 shows that high-end security appliances don’t always command a high price. Targeting busy SMBs with up to 250 users, this affordable 1U rack model boasts a top raw firewall throughput of 18Gbits/sec, dropping to a respectable 2.4Gbits/sec with all UTM security services enabled.

A simplified licensing scheme makes it easy to manage costs. We’ve shown the price of a Total Security Suite (TSS) subscription, which enables everything WatchGuard has to offer. This includes gateway antivirus, anti-spam, web content filtering, application controls, intrusion prevention services (IPS) and an advanced persistent threat (APT) blocker with cloud sandboxing.

You also get WatchGuard’s reputation enabled defense (RED) cloud-based URL filtering, ThreatSync XDR for collection, correlation and automated responses to threat events, and DNSWatch to monitor client DNS requests and block access to known malicious domains. It has enough power to run the Cylance AI-based IntelligentAV malware scanning engine; the elderly dual-core Celeron in the M370 has been upgraded to an 8-core NXP LS2084A CPU.

There’s room to grow, as the front expansion bay accepts modules with gigabit, 10GbE or 2.5GbE multi-gigabit ports. Note that the PoE+ services on the multi-gig module will be disabled as the M390 doesn’t support the required 54V power supply, which is only available on the M590 and M690 models.

The appliance is easy to deploy. Its web console wizard sorts out the LAN and WAN network interfaces and applies a base set of firewall policies that include blocking common undesirable web categories. We had already registered the appliance’s serial number with our cloud support account so the wizard grabbed our TSS feature key and applied it.

The Firebox M390 offers plenty of features for the price (Image credit: Future)

The M390 can be managed in standalone mode, but businesses with multiple appliances will prefer WatchGuard’s Cloud portal, which is included in both the Basic Security Suite (BSS) and TSS subscriptions. Two choices are available: you can keep local management and set the appliance to send its logs to the cloud for remote monitoring and reporting or move it all into the cloud.

Full cloud management is activated by allocating the appliance from your account inventory and enabling it from the local console. The portal steps through WAN port setup and applying new admin passwords, after which the appliance disables local management.

Cloud configuration is simple as the portal groups all security settings for the appliance in one page. From the content scanning section you activate antivirus scanning, APT blocking, IntelligentAV and anti-spam policies for incoming SMTP, IMAP or POP3 traffic.

The network blocking section covers botnet detection, IPS, port and site blocks and detection of Tor exit points. WatchGuard’s WebBlocker service is accessed in the content filtering section and offers 166 URL categories that can be allowed, blocked or set to display a warning page to users.

The Firebox M390 can be managed locally or remotely (Image credit: Future)

Each content filter action policy manages both web access and application controls. The latter presents nearly 1,300 predefined app and protocol signatures, with Facebook and X activities getting the lion’s share of the social networking category. They’re easy to apply; when creating firewalls rules, you simply choose the action policy you want to assign.

The portal’s monitoring page provides a wealth of information on all activities. Graphs and charts are provided for live activity, traffic, detected malware and botnets, application usage, blocked websites, top clients and much more.

The Firebox M390 will appeal to SMBs with a large user base. It combines strong performance with an incredible range of security measures all at a competitive price. It’s easy to deploy and WatchGuard’s cloud portal delivers excellent remote management and monitoring features.

We also rated the best anti-virus software for PC.

Anthropic's newest AI model is now available, but Claude 3.5 Haiku costs 4 times more than its predecessor.

The new Claude 3.5 Haiku costs $1 per million input tokens and $5 per million output, that's a huge increase compared to the 25-cent input per million and $1.25 output per million of the previous model.

Anthropic had initially claimed the new AI model would remain at the same price point as the older version, but the company took to X to say that the price would increase due to higher results than expected in benchmarks.

The tweet read, "During final testing, Haiku surpassed Claude 3 Opus, our previous flagship model, on many benchmarks — at a fraction of the cost. As a result, we’ve increased pricing for Claude 3.5 Haiku to reflect its increase in intelligence,"

It's worth noting that this new 3.5 Haiku is far more powerful than the previous 3 Haiku AI model with a more up-to-date knowledge base and the ability to write longer text answers. It may be more powerful, but it's still missing some features found in the older model such as image analysis. Anthropic has said Claude 3 Haiku will remain available for users who want to use the image processing that the new 3.5 Haiku is missing. Anthropic says image input will arrive at a later date.

During final testing, Haiku surpassed Claude 3 Opus, our previous flagship model, on many benchmarks—at a fraction of the cost.As a result, we've increased pricing for Claude 3.5 Haiku to reflect its increase in intelligence: https://t.co/UWTd1RDRVE.November 4, 2024

Claude 3.5 Haiku

Claude 3.5 Haiku is a more advanced version of the cheaper 3 Haiku and Anthropic says it outperforms the company's previous flagship AI model, Claude 3 Opus.

In that sense 3.5 Haiku is worth the increased cost, allowing users to access a far more powerful AI at a lower cost than 3 Opus which is priced at $15 per million input tokens and $75 per million output tokens. If Anthropic's claim that 3.5 Haiku outperformed 3 Opus on benchmarks is true then the 4x price hike compared to 3 Haiku is far easier to digest.

Claude 3.5 Haiku is available now and you can request API access directly from Anthropic's website.

You Might Also Like...

• A new AI feature can control your computer to follow your orders
• You and your friends can now share and remix your favorite conversations with the Claude AI chatbot
• Anthropic wants its AI assistant Claude to be your favorite coworker

This review first appeared in issue 360 of PC Pro.

TP-Link’s first Wi-Fi 7 business access point (AP) sets a high standard for the rest to follow, as this slimline discus has a mighty BE19000 rating. The EAP783 achieves this with a claimed 1,376Mbits/sec on the 2.4GHz radio, 5,760Mbits/sec on 5GHz and a speedy 11,520Mbits/sec on 6GHz, with the latter enabling the ultrawide Wi-Fi 7 320MHz channels.

It’s also one of the first APs to support the multi-link operation (MLO) feature, which allows devices to connect to the 2.4GHz, 5GHz and 6GHz bands simultaneously for increased throughput and reduced latency. The only downside is that very few end-user devices currently support MLO, and Windows users will have to wait until Windows 11 24H2 is fully released.

The EAP783 has other appealing credentials. It provides two 10GbE ports, with the first requiring an 802.3bt PoE++ power source; there’s also an input for TP-Link’s optional 12V adapter. The second 10GbE port is designed to be aggregated with the primary network port for a higher network connection speed.

Despite the AP’s 12 internal aerials, TP-Link has kept the size down to a manageable 280mm diameter and a thickness of 47mm. It’s heavy, though, tipping the scales at 1.4kg – around 15g more than Zyxel’s WBE660S.

The AP comes with two 10GbE ports (Image credit: Future)

The EAP783 supports standalone mode, where the web console’s quick start wizard requests a new username and password, after which it removes the default admin account for extra security. Eight SSIDs per radio can be created, and the wizard asks you to configure the first three and provide new encryption keys.

For our real-world Wi-Fi 7 performance tests we connected the AP to the lab’s Zyxel XS1930-12HP 10GbE PoE++ switch and used a Lenovo desktop client running Windows 11 Pro 24H2 Insider Preview and fitted with a TP-Link Archer TBE550S Wi-Fi 7 PCI-E adapter. With Windows reporting a Wi-Fi 7 connection speed of 5,764Mbits/sec, we saw large file copies between the client and a Windows server on our 10GbE LAN averaging close-range speeds of 323MB/sec, dropping to an equally impressive 280MB/sec with the AP placed ten meters away in an adjoining room.

The web console provides a separate section for MLO where you create a new SSID and decide which of the three radios to assign to it. Performance-wise, we saw no benefits from a triple radio MLO SSID as the Windows 11 client reported the same speed as a standard Wi-Fi 7 connection and our copies were no faster.

TP-Link offers plenty of cloud management choices. You can deploy its hardware or free software controller on-site or use its cloud-hosted controller, which has a yearly fee of £165 for 250 devices. We use TP-Link’s OC300 dual-port controller appliance in the lab as its £130 price enables lifetime cloud management of up to 500 Omada APs, switches and routers.

The EAP783 can be cloud-managed from TP-Link’s Omada platform (Image credit: Future)

All controllers are viewed from the Omada portal home page, and selecting the OC300 transported us to its own console. This presents a ribbon across the top showing all cloud-managed devices, and selecting a site below provides options to create multiple dashboards with a wide range of widgets.

The EAP783 initially appears as “pending” in the portal, and all you do is adopt it and assign it to a site. It only takes a minute, after which the AP disables its local web interface for security reasons, takes all its settings from the controller and broadcasts your preconfigured SSIDs.

It’s easy to add more SSIDs, apply a desired personal or enterprise encryption scheme, set client and SSID rate limits and define schedules to determine when they’re active. For the EAP783, we could enable any of all of its three radios, and we noted the portal has an option to create MLO Wi-Fi 7 SSIDs.

The EAP783 will appeal to businesses looking to make an early transition to Wi-Fi 7. It delivers excellent performance and is MLO-ready, and TP-Link’s Omada cloud platform offers great remote management services.

We also rated the best routers for small businesses.

• Amazon is fixing the Kindle Colorsoft screen issue
• A switch of materials may be to blame
• We noticed minimal problems on our review unit

Amazon has previously acknowledged a yellowing screen issue with a "small number" of Kindle Colorsoft devices, and told us a fix is in the works – and now we have a bit more information about what might have gone wrong.

Well-known industry analyst Ming-Chi Kuo has weighed in on the issue to say the problem has been traced to Amazon switching to a different OCA (optically clear adhesive) from the one used in the best Kindles with a monochrome display.

Presumably, that adhesive is what's causing the yellowing issue. A significant number of users have reported a yellow band across the screen of their Colorsoft unit, though in our very positive Amazon Kindle Colorsoft review we found the issue was barely visible unless you're specifically looking for it.

According to Kuo, the switch of materials was partly down to the lower contrast of the color screen. The analyst says "component suppliers have developed several hardware solutions" but "Amazon seems to be leaning toward a software-based fix".

The best Kindle yet?

According to recent reports, Kindle Colorsoft displays are showing yellow discoloration bands along their bottom and sides - an issue traced to Amazon's use of a different OCA than that used in grayscale e-paper displays. This change was implemented to enhance the limited…November 7, 2024

Based on what Kuo has said, it would seem that Amazon has solved the problem by tweaking the software on the Kindle Colorsoft, rather than changing anything to do with the hardware – so any models sold from now on should be fine.

Amazon has told us that "appropriate adjustments" have been made to solve the issue, and that anyone who finds the yellowing a problem on their Colorsoft device can get in touch with Amazon support for a replacement or a refund.

The new e-reader, the first Kindle with a color screen, will set you back $279.99 / £269.99 (with pricing in Australia still to be announced). In our review, we said it "might be the best Kindle yet" – and suggested that once you make the switch to color, you're likely going to find it difficult to go back.

While this screen issue doesn't seem to be too widespread, and has apparently been quickly fixed, it's obviously something Amazon could have done without – especially as it marks a new color era for the Kindle.

You might also like

• There's just one problem with my beloved Kindle
• Kindle vs Kobo: how to choose the best ebook reader
• Amazon launches new Kindles for 2024

A hospital in Pittsburgh has become the latest medical center to join a ground-breaking Apple Watch trial in the US that's helping patients with atrial fibrillation ditch blood-thinning medication.

The REACT-AF trial began in July 2023 and is led by researchers at Johns Hopkins Medicine. Randomized patients (half of the overall participants) wear an Apple Watch for 14 hours a day, paired with an iPhone pre-programmed to monitor irregular heart rhythms.

If patients can go more than 30 days without an episode of erratic heartbeats lasting longer than an hour, they can stop taking their anticoagulant medication.

It's a trial that could transform cardiovascular healthcare, reduce the risks of complications associated with the medication, and save a whole lot of money in the process.

Apple breaks more ground in health

(Image credit: Future)

The Allegheny General Hospital recently announced that it was among the nation's top enrollment sites for the trial, and its principal lead on the trial is full of praise for the initiative.

"The purpose of this trial is to better understand how to personalize anticoagulation safely and effectively, especially for patients with isolated atrial fibrillation episodes," Amit Thosani, MD, remarked, noting that blood thinners prescribed to reduce stroke risk in Afib patients is also associated with increased risk of internal bleeding.

As noted, patients monitored can come off blood thinners if they go 30 days without an episode, a benefit keenly felt by Pittsburgh local Arthur Schiebel. “I’m not used to taking all this stuff," he told the Pittsburgh Post-Gazette, "I just want to get off all this medication."

After two years of treatment to get Afib under control, Thosani enrolled in the REACT-AF study, and says he has been off his blood-thinning medication since February "with no issues." “I never wore a watch in my life but if this thing will keep me off the medication, I’m wearing it," he said.

Afib tracking is just one of the Apple Watch's suite of health features, which were recently bolstered by the introduction of sleep apnea detection to the best Apple Watch models, the Apple Watch Series 10 and Series 9, and the Apple Watch Ultra 2. Along with features like Fall and Crash Detection, the Apple Watch is fast becoming an indispensable health-monitoring tool that can help to prevent disease and injury for wearers.

You may also like

• Here's why the Apple Watch Series 10 is now my favorite big-screen smartwatch
• Future Apple Watches could get a genius way to track your health, thanks to smart straps
• A key Apple Watch health feature just got a landmark stamp of approval

• Nvidia RTX 5000 and 6000 GPUs have been filed at the EEC
• These filings from a Chinese GPU brand are speculative by nature
• They do, however, hint that the RTX 5000 series is close to debuting

Nvidia’s RTX 5000 GPUs have been spotted in trademark filings, along with RTX 6000 models – which really is getting ahead of the game.

VideoCardz noticed that leaker Harukaze5719 on X posted the details of trademark registrations for Nvidia graphics cards, which have been filed by Sinotex – owners of the Ninja GPU brand in China – with the EEC (Eurasian Economic Commission).

RTX 60, RTX 50, RTX 40...It's just a trademark registrationhttps://t.co/stNKemJcyK pic.twitter.com/w2VUJiix36November 7, 2024

These filings are typically made just ahead of the release of new hardware products, though they are speculative in nature – as the RTX 6000 models show quite clearly. Those GPUs being mentioned at this stage is a total shot in the dark, of course.

Also mentioned here is an RTX 5090 Ti and RTX 5090 Super, and in fact the full raft of Super and Ti options down the Blackwell range – the same is true for the RTX 6000 models in the filings, too. This is Sinotex simply covering its bases as to the up-and-coming graphics cards it may need to manufacture.

(Image credit: Future) Analysis: Heavy on the seasoning here – but with a hint of hope

So, why is this even worth talking about? Well, we must be heavy with the seasoning – and ignore the mention of RTX 6000 GPUs for starters. We don’t even know yet if Nvidia’s following generation after next-gen Blackwell GPUs will be RTX 6000, or something different – or maybe Nvidia won’t even be making GeForce graphics cards in 2027. Maybe AI will have shoved gaming GPUs entirely to the sideline at that point for Team Green.

The fact that a Chinese GPU maker is now moving to register RTX 5000 models is, however, a weighty hint that we’re on track for the purported reveal of Blackwell graphics cards at CES 2025. (And the company has thrown in a bunch of RTX 6000 filings for good measure, at the same time).

As to any hint of this meaning Nvidia will have an RTX 5090 Super or Ti version, we wouldn’t read anything into the mention of those GPUs here – as noted, this appears to be a broad blanket move by Sinotex to cover all its bases.

The expectation is that Nvidia will reveal the RTX 5090, 5080 and possibly 5070 desktop graphics cards at CES 2025, with a potential launch of Blackwell mobile GPUs for gaming laptops alongside those GPUs. Although perhaps this fresh leak could be another hint that we might get a revelation from Team Green earlier, towards the end of 2024, as a recent rumor suggested. This might only be a teaser, though – if anything at all.

You might also like

• Nvidia’s RTX 5090 now rumored to have superfast clock speeds – as well as being super-slim – could this GPU be too good to be true?
• Nvidia rumored to launch whole range of next-gen GPUs by March 2025, including RTX 5060 and 5060 Ti – but we’re not convinced
• We could be witnessing the death of the graphics card in real time right now, and I couldn’t be happier about it

• Voting for the Golden Joystick Awards closes soon
• You can vote for your Ultimate Game of the Year and Best Game Adaptation
• Nominees include Balatro, Dragon Age: The Veilguard, and Silent Hill 2

If you've yet to vote for your Ultimate Game of the Year choice at this year's Golden Joystick Awards, you've got until later today to cast your vote!

This year's Golden Joystick Awards takes place on November 21 and will be presented by Ben Starr and a host of well-loved voice actors and developers. Right now, only the final two categories are open for voting: Ultimate Game of the Year and Best Game Adaptation. You'll have until 3pm PST / 6pm EST / 11pm GMT today to get your votes in.

You can vote over at our sister site, GamesRadar+, on their dedicated Golden Joystick Awards page. There's no need to sign up for anything, but as a nice little bonus, those who do vote can claim a free gaming eBook usually worth $18.99 / £12.99. To claim it, simply enter your email while casting your votes to be contacted about your free eBook selection.

Then, you can head on over to YouTube, Twitch, Facebook, X / Twitter, or Steam to watch the Golden Joystick Awards live on November 21. GamesRadar, PC Gamer, and ourselves at TechRadar Gaming will also be streaming the event live as it happens.

Without further ado, the nominees for each of those final two categories you can vote for are as follows:

Ultimate Game of the Year

• Animal Well (Billy Basso, Bigmode)
• Astro Bot (Team Asobi, PlayStation Studios)
• Balatro (LocalThunk, Playstack)
• Black Myth: Wukong (Game Science)
• Call of Duty: Black Ops 6 (Activision)
• Dragon Age: The Veilguard (BioWare)
• Final Fantasy 7 Rebirth (Square Enix)
• Helldivers 2 (Arrowhead Game Studios, PlayStation Studios)
• Metaphor: ReFantazio (Atlus)
• Satisfactory (Coffee Stain)
• Silent Hill 2 (Bloober Team, Konami)
• Tekken 8 (Bandai Namco)

Best Game Adaptation

• Borderlands
• Fallout
• Halo Season 2
• Knuckles
• Like a Dragon: Yakuza
• Tomb Raider: The Legend of Lara Croft

You might also like...

• PS5 Pro review: a high-performing enthusiast console with a pricey sting in the tail
• I reviewed the PS5 Pro and recommend these 7 enhanced games to try first
• Evercade Alpha review: authentic countertop arcade fun

Have you recently dumped a partner over their politics? Or chosen to stay friends with someone who voted differently this year? NPR wants to hear how politics have impacted relationships.

(Image credit: Paul Hennessy/Anadolu)

APYs are likely to keep falling after yesterday's Fed rate cut.

Apple, like most phone makers, puts a lot of focus on improving the cameras from one model to the next, and we now have an idea of what the next two generations of camera updates might look like.

Numerous leaks have already pointed to 48MP telephoto cameras for the iPhone 17 Pro and iPhone 17 Pro Max, with the whole iPhone 17 line possibly getting 24MP front-facing cameras. But with the iPhone 18 line, Apple might focus on the aperture rather than megapixels.

This is according to reputable analyst Ming-Chi Kuo, who in his latest post on Medium (via @Jukanlosreve) claimed that Apple will be equipping the main camera on the “high-end” iPhone 18 models with a variable aperture.

By high-end he presumably means the iPhone 18 Pro Max and likely also the iPhone 18 Pro, but he doesn’t mention the phones by name.

What Kuo does say is that this variable aperture would "significantly" enhance the photography experience. The aperture – the size of the lens opening – on most cameras is fixed, and having a variable aperture means you can control how much light reaches the sensor. This in turn gives you more control over exposure – you can open the aperture to its widest setting for better low-light shots, for example – and it also enables you to create optical depth-of-field effects to blur backgrounds, rather than relying on software processing.

So this sounds like it would be a worthwhile upgrade, and while we’d take Kuo's reporting with a pinch of salt for now – especially as this is a very early leak – he does have a great track record for Apple information.

Not the first variable-aperture phone

The Xiaomi 14 Ultra has a variable aperture (Image credit: Future / Roland Moore-Colyer)

If this claim does prove to be accurate, the iPhone 18 Pro and Pro Max wouldn't be the first smartphones with a variable-aperture lens. Samsung equipped the Galaxy S9 series with variable-aperture cameras way back in 2018, and also used variable aperture-cameras in the Samsung Galaxy S10 line.

We were quite taken with the tech in our Samsung Galaxy S9 review, noting that it particularly aided low-light shots, but it wasn’t something Samsung stuck with beyond the S10 series, so it presumably didn’t lead to increased sales.

More recently, the Xiaomi 14 Ultra also has a variable aperture camera, which coupled with a 1-inch sensor resulted in “fantastic low-light performance and unrivaled image quality” according to our Xiaomi 14 Ultra review.

You might also like

• iPhone 16 review: elevating the base to new heights
• Best iPhone: which Apple smartphone reigns supreme?
• iPhone 17: latest news and rumors for every expected model