News
If you're a seasoned anime fan, you'll understand the unique sense of excitement that comes with binging a brand-new show — and it's exactly how I feel about Crunchyroll's breakout series, BLACK TORCH.
As of writing, we're now six episodes into the epic adventure, following high-schooler Jiro, who just also happens to be a ninja descendant who can talk to animals. He's rescued an injured black cat named Rago, who is actually a powerful mononoke, or demon spirit.
I'll let Crunchyroll explain the next part: "After Jiro is mortally wounded protecting Rago from hostile spirits, Rago fuses with him, creating a human-spirit hybrid. Jiro is then recruited by Japan's covert Bureau of Espionage into a special unit called 'Black Torch' to fight rogue mononoke."
It's a classic anime storytelling format with a new twist, and so far, its six episodes have successfully blended high-intensity action with a strong emotional core. Even the intro song feels a cut above the rest, and I'm not surprised that anime fans are lapping it up.
With six episodes left to go, what can we expect from the second half of BLACK TORCH? According to English dub cast AJ Beckles (Jiro) and Ryan Colt Levy (Ryosuke Shiba), you're best off expecting the unexpected.
'It's more of what we've had... but bigger'"Expect the unexpected," Beckles tells me. "It's more of what you've got, but bigger. As we reach the end of the season and these episodes start to come out, there more high-octane, serious moments. There's a lot to look forward to, for sure."
It's a vague response, but this doesn't surprise me. Even the above trailer showed almost no signs of what was to come in the actual series, which was a refreshing decision. Of course, we can expect the drama to follow the original manga, but this can't be assured as a given.
But has there been any pressure for either actor knowing that these episodes are airing to a pre-existing fanbase?
"I don't know if it's necessarily like a pressure... I like to think of it as a responsibility," Levy says. "No matter what, there are people out there that you are making this for, and where we may never get to have that face-to-face feedback, we have access. People will put so much passion into that.
"We approach this no matter what with the intention to make sure that people who love it feel seen, and feel like they are spending time with the characters that they want."
"We've been cast and It's a ginormous responsibility, because it could have gone to anybody," Beckles agrees. "So the fact that it went to us means that we have a duty to do a good job. We're not necessarily thinking about the end result. The focus should be on, 'Am I thinking as Jiro is thinking? Am I serving this character?' Because if I do so to the best of my ability, someone will resonate with that."
I've got no idea how I'm going to feel about the upcoming second half of BLACK TORCH, but hearing the sheer passion that the voice cast has for its craft fills me with hope — and the joy that my initial burst of anime excitement seems set to continue.
A new Quordle puzzle appears at midnight each day for your time zone – which means that some people are always playing 'today's game' while others are playing 'yesterday's'. If you're looking for Friday's puzzle instead then click here: Quordle hints and answers for Friday, August 14 (game #1663).
Quordle was one of the original Wordle alternatives and is still going strong now more than 1,500 games later. It offers a genuine challenge, though, so read on if you need some Quordle hints today — or scroll down further for the answers.
Enjoy playing word games? You can also check out my NYT Connections today and NYT Strands today pages for hints and answers for those puzzles, while Marc's Wordle today column covers the original viral word game.
SPOILER WARNING: Information about Quordle today is below, so don't read on if you don't want to know the answers.
Quordle today (game #1664) — hint #1 — VowelsHow many different vowels are in Quordle today?• The number of different vowels in Quordle today is 4*.
* Note that by vowel we mean the five standard vowels (A, E, I, O, U), not Y (which is sometimes counted as a vowel too).
Quordle today (game #1664) — hint #2 — repeated lettersDo any of today's Quordle answers contain repeated letters?• The number of Quordle answers containing a repeated letter today is 1.
Quordle today (game #1664) — hint #3 — uncommon lettersDo the letters Q, Z, X or J appear in Quordle today?• No. None of Q, Z, X or J appear among today's Quordle answers.
Quordle today (game #1664) — hint #4 — starting letters (1)Do any of today's Quordle puzzles start with the same letter?• The number of today's Quordle answers starting with the same letter is 2.
If you just want to know the answers at this stage, simply scroll down. If you're not ready yet then here's one more clue to make things a lot easier:
Quordle today (game #1664) — hint #5 — starting letters (2)What letters do today's Quordle answers start with?• R
• M
• M
• L
Right, the answers are below, so DO NOT SCROLL ANY FURTHER IF YOU DON'T WANT TO SEE THEM.
Quordle today (game #1664) — the answers(Image credit: Merriam-Webster)The answers to today's Quordle, game #1664, are…
- RIDER
- MOLDY
- MEALY
- LYMPH
Another day featuring a word without vowels.
I already had the Y here, so it was simply a case of finding the P to make up LYMPH.
Y was the hero letter of the day featuring in the trickiest word but also helping out at the end of two others — MOLDY and MEALY — making for an easier game than usual.
Daily Sequence today (game #1664) — the answers(Image credit: Merriam-Webster)The answers to today's Quordle Daily Sequence, game #1664, are…
- FIBRE
- JUMPY
- LATHE
- LIVER
- Quordle #1663, Friday, 14 August: SEIZE, MYRRH, SANER, GLOAT
- Quordle #1662, Thursday, 13 August: TOPAZ, PURGE, WHINE, SIGHT
- Quordle #1661, Wednesday, 12 August: HOTLY, CROWD, HEART, SWASH
- Quordle #1660, Tuesday, 11 August: ONSET, BRAWL, TEPID, SLANG
- Quordle #1659, Monday, 10 August: GAVEL, CRUSH, SPEED, ETHOS
- Quordle #1658, Sunday, 9 August: KAPPA, WIDEN, FOLIO, STEEP
- Quordle #1657, Saturday, 8 August: FUROR, BUGLE, REUSE, SCOOP
- Quordle #1656, Friday, 7 August: AMISS, SHIRK, SALSA, RABBI
- Quordle #1655, Thursday, 6 August: AMUSE, BRAND, CEDAR, BLURB
- Quordle #1654, Wednesday, 5 August: MOUND, WRONG, STOOD, WEIRD
- Quordle #1653, Tuesday, 4 August: GOOEY, TARDY, HUMAN, CHALK
- Quordle #1652, Monday, 3 August: FUNGI, MUDDY, DEBUT, MANIA
- Quordle #1651, Sunday, 2 August: BOOTH, EASEL, REACH, MAKER
- Quordle #1650, Saturday, 1 August: OMEGA, BIRCH, SMOCK, EERIE
- Quordle #1649, Friday, 31 July: MANIA, SINGE, STOOL, LOFTY
- Quordle #1648, Thursday, 30 July: MODAL, IDLER, CHUMP, LUMPY
- Quordle #1647, Wednesday, 29 July: GAWKY, ALLOY, AUDIT, LATCH
- Quordle #1646, Tuesday, 28 July: SMITE, ROACH, RUDDY, GLOBE
- Quordle #1645, Monday, 27 July: TAFFY, FIGHT, GUILD, WAGER
- Quordle #1644, Sunday, 26 July: STOMP, LIMIT, FUNKY, STEAD
- Ukrainian drone forces have proved so successful against US troops in exercises that “destroyed” equipment had to be reintroduced into the simulated combat
- Over the course of the two-week-long exercise, US forces improved at camouflage, dispersing, and using electronic warfare against the aerial attacks
- The Combined Resolve military exercise took place in Germany earlier this year
US forces engaged in military exercises were stunned to be defeated quickly by Ukraine drone teams, so much so that the units were allowed to be re-added into the scenario.
Taking place in Germany earlier in 2026, the event was one of several exercises intended to simulate real drone warfare and develop responses and countermeasures to attacks. Ukraine's drone warfare has proved vital in its ongoing war with Russia, with its successes in the conflict defining the country as a pioneer of Unmanned Systems Forces and electronic warfare integration.
The US isn’t the only country to have engaged in wargames with Ukraine, with the UK, Estonia, and Sweden all recently outmatched
Recording a “kill”Despite the US incorporating long-range drones into its arsenal during the War on Terror, it has had mixed results in defending against drone attacks, both in combat and in exercises. The simulated conflict with Ukraine took place in April and May 2026, and while the Fort Hood-based US forces may have come away from the event with some pride, the speed and decisiveness of the Ukraine drone forces clearly took them by surprise.
Rather than using live ammunition in these exercises, an observer is employed to determine whether a “kill” has been made. This is usually achieved by the drone controller hovering the craft above a vehicle or other combat equipment long enough for a potential kill to have been made.
In this particular incident, the Ukraine drones were recording kills so quickly that vehicles had to be sent back as new units, like a video game respawn.
However, as the exercises progressed, the US forces were able to improve their performance. Their responses to drone attacks relied on faster camouflage and concealing, effective dispersal, and employing electronic countermeasures to block the drones.
Ukrainian drones dominateAs part of the training, Ukrainian drone controllers also switched sides to assist the US troops in making effective offensive use of drones.
The Wall Street Journal reports that Air Force General Alexus Grynkewich, NATO’s top military leader, stated that “You can’t get training like this in the United States,” but observed a noted improvement in NATO forces after training: “The change is remarkable.”
It isn’t just the Americans that have struggled with Ukrainian drone warfare in simulated scenarios. Other events in 2026 have revealed shortcomings in handling drone attacks in exercises involving the UK, Estonia, some US troops, and other NATO forces. In addition, a military exercise involving Swedish-led NATO troops and Ukraine's drone warfare resulted in the NATO side being outmatched.
- Dolby says that Dolby Vision HDR movies can be upgraded to Dolby Vision 2 using an automated process based on the existing mastering information
- This means support may be more widespread than expected, since apparently it will be relatively low-effort for streamers to achieve
- The Authentic Motion and Precision Black features will be excluded from this, because Dolby says they they require human judgment to get right
If you're wondering where all the Dolby Vision 2 media is going to come from, we've got good news: it's already here, at least partly. Dolby Laboratories told me that it's created a process that can upgrade existing Dolby Vision media to add most features of the new HDR standard using an automated process based on the existing mastering data — and there are only two key features that are excluded, because they're areas that depend on new input from creators.
With Dolby Vision 2 arriving on TVs from this month, I caught up with Dolby Laboratories' director of business strategy, Jonas Klittmark, to find out the details: what it could do for 4K Blu-rays, what benefits it'll bring to streaming and most importantly of all, what to watch.
One of the biggest concerns with Dolby Vision 2 is that creators just won't be inclined to master video for it when it doesn't have widespread support — but it's unlikely to get widespread support if there's no content. But it turns out that existing Dolby Vision masters can be adapted to use many of its features without being reworked.
"There is a bunch of new features in Dolby Vision 2 that don't rely on new content at all," Turkstra told me. "Because you already have all the Dolby Vision content in the world... the streaming providers will have more to say about this later, but the existing Dolby Vision library will start to get converted and the new metadata for Dolby Vision 2 will be added."
The Hisense UR8 is one of the first TVs to support Dolby Vision 2 — but there needs to content to make it worthwhile… (Image credit: Future)Klittmark estimates that there are around 33,000 movies and TV show episodes available in Dolby Vision. "There's no realistic chance that all that content is going to get remastered from scratch," he says.
"The cost of doing that is just astronomical. So what we're offering instead is a semi-automated way of doing it where we're using modern techniques such as machine learning to assist in the creation of the additional metadata that's needed to make the most of Dolby Vision 2."
He elaborated that "many of the features, including the image engine, including the new version of LightSense, the intensity control, the pro settings, and sports optimization, all of those features can work through a combination of existing metadata."
That metadata is based on existing mastering data, and as will all of Dolby Vision 2, the next-gen data is in addition to the existing data; it doesn't replace it, so the automation won't affect the visuals of, or compatibility with, TVs that have the first generation of Dolby Vision but not the second.
What Dolby Vision 2 features can arrive automatically – and what ones can'tThe two exceptions to the automated upgrade are Authentic Motion, which enables creators to control how strong your TV's motion processing is applied on a moment-to-moment basis, and Precision Black, which is designed to address dark scenes that looked great in the creators' mastering suite but appear too dark on TVs.
Klittmark said: "Authentic Motion is an example of a feature where it's just too sensitive. I don't know if you've been following the fairly polarized debate about motion that has been playing out over the years, with Tom Cruise making public service announcements, and things like that. That feature is so sensitive that we don't generate that metadata, it has to be manually authored."
I've seen Dolby Vision 2's Authentic Motion in action, and I think it's a great addition — it means that judder can be removed in scenes that need it, but motion smoothing can be deactivated in scenes that don't. I suspect that actually it probably could be automated to good effect, but as Klittmark says, I don't thing anyone would really trust that, so it's best that Dolby is leaving that to humans.
When it comes to Precision Black, Klittmark says this can't be automated because in order to get it right you need to know all about the creator's environment: "how dark was their ambient light, what kind of reference monitor were they using, et cetera... that cannot be speculated about or generated, so that needs to come manually from the creative."
When is Dolby Vision 2 not Dolby Vision 2?I asked if this meant that movies might be labelled as supporting different Dolby Vision 2 features, but Klittmark says the firm doesn't want to make things complicated in that way.
"We've taken the simplest possible approach for consumers' benefit, which is that all the content is going to be Dolby Vision. There is no asterisk. There's no 'Dolby Vision 2' content, for example. There's no 'Dolby Vision 2 Max' content or anything like that. It's just all Dolby Vision content," he says. "And that way, it's very simple for a consumer to know what to look for. If you see that the content is in Dolby Vision, then you are getting the best experience you can get with that particular movie or whatever it is you're looking for."
I think this makes sense — your TV will be labelled as Dolby Vision 2 or Dolby Vision 2 Max, and potentially could tell you which features of the format are active, but when it comes to browsing the library, it makes sense to simply use the Dolby Vision logo to indicate that you're getting as many advanced features as the service has added, regardless of what they are.
Klittmark pointed out that this would be new to Dolby Vision 2, either. "The original version of Dolby Vision had tons of features in it, and not all creatives chose to use all of them all of the time. And there was no way to know that. And that's essentially a choice up to the creative. Which of the Dolby tools do they need do they feel like they need to use for which content? We leave that decision to the creative," he said.
Thinking of buying a new TV?Try our TV size and model finder! You tell it how far you sit from your TV, we'll tell you what size to buy based on viewing angle advice from image quality experts, and we'll recommend our three top TVs at that size for different prices.
Networking never fails to surprise me when it comes to dependencies and its live nature – it’s a world that’s continually changing, and yes upgrades designed to improve performance or security are often the ones that cause outages.
Even carefully planned updates can have consequences that might initially be difficult to see during early pilots and test, due to a highly dependent mix of switches, firewalls, clouds and more – not to mention the challenges that come with interoperability in a world of mixed vendors.
Until now, it’s mostly been a firefighting exercise of responding to incidents and outages as they happen, admitting that they’re just an inevitable outcome of virtually any change.
Autonomous networking needs a safety-first designForward (formerly Forward Networking) wants to challenge this in its entirety, though, and the work it and companies like it are doing could fundamentally revolutionize network stability and uptime, significantly reducing costs.
Its solution is a mathematically accurate digital twin of the production environment, giving companies space to make changes and observe unintended impacts without actually affecting the real deal.
The principle isn’t anything new at all, and it’s been used in software development and other areas for decades, but where it could really shine is in an upcoming age of automated networking. Forward argues that providing a place for systems to autonomously test changes pre-deployment is a necessary stage to true automation.
I wanted to know whether technology like this could have prevented some major high-profile incidents lately, including those we’ve seen relating to Amazon, Cloudflare and more, and spoke to Forward’s co-founder and Chief AI Officer, Nikhil Handigol, on some of the technology’s biggest challenges, including how the digital twin could stay current and why pre-deployment proof could become essential in the world of autonomy.
- Forward proposes to prevent 100% of network outages caused by configuration errors. Are there any strings attached? And what does a “mathematically accurate digital twin of the entire production network” actually mean?
There are no strings, but let’s be precise on what that rests on. Forward Predict eliminates network risks through routing validation, end-to-end path analysis, firewall and ACL verification, segmentation and compliance verification, and automated regression testing.
The closest parallel is what software development did with continuous integration. No code goes to production without being tested first. Forward Predict brings that same discipline to networking.
Every proposed change is tested against the full production-equivalent model before it touches a live environment – routing validation, end-to-end path analysis, firewall and ACL verification, segmentation and compliance verification, and automated regression testing.
That model isn’t a sample or a simplification. It's built from the configuration and operational state of every device in the network — every vendor, every OS, every layer from L2 to L7, on-premises and in the cloud — using header space analysis to trace every possible path any packet can take.
The platform regularly collects state from the live network, and you can trigger a fresh snapshot on demand immediately before a push, so the "digital twin" you're testing against is always current. That continuous synchronization is what makes the math trustworthy: a proof is only as good as its inputs, and this is how we keep the inputs honest.
Run a change through that model and you get a deterministic outcome with evidence, not a guess. If it breaks something, if it opens a security hole, if it creates an unintended path, you see exactly why.
And because the answer comes from mathematical path analysis rather than a probabilistic risk score, it doesn't degrade as the network grows more complex.
Pair that with Forward AI and the platform does not just flag the problem, it iterates on its own until it lands on a change that is verified to deliver the intended behavior. Change windows that once took weeks now take minutes. There is no anxiety pushing a change live, because you already know how it ends. Weekends are for time off, not change windows and war-rooms.
- One point I wanted to raise is that you're essentially pegging it against a snapshot. Sysadmins will tell you that networks change. Firmware get updated, access points get refreshed, switches swapped and servers decommissioned. How do you make sure that Forward remains relevant throughout the lifecycle of the production network?
Forward Enterprise regularly collects configuration and state from every device in the estate, and you can trigger an on-demand snapshot immediately before pushing a change for absolute certainty.
When firmware gets patched, when an access point gets refreshed, when a switch is swapped, when a server is decommissioned, any change to the network is updated in the digital twin, it is a mathematically accurate model of the entire production network.
This is exactly why the underlying model holds up under scrutiny. Our engineering team validates the software against real behavior on live test hardware from every supported vendor, so the model's fidelity is proven before it ever reaches a customer network.
Then in production, that same discipline continues against the customer's own, constantly changing estate. Sysadmins are right that networks never sit still. Neither does the network digital twin. That is the only way proof stays proof instead of becoming a guess based on past behavior.
- Are there other companies that compete with Forward? How does your proposition differ from theirs?
The competitive threat isn't another vendor. It's a set of processes built over decades, in an era before a network could be modeled at all, that include change windows, war rooms, and the belief that the only place to truly test a network change is the production network itself.
Every other industry running mission critical infrastructure solved this problem years ago. Airlines train pilots on flight simulators instead of real aircraft. Pharmaceutical companies model outcomes before anything reaches a patient.
Manufacturing validates changes on a digital twin before touching the factory floor. Software development has run on continuous integration for two decades, where no code ships without being tested first.
Networking is the industry that never got its equivalent, not because nobody wanted one, but because building a mathematically accurate network digital twin is genuinely hard.
Against observability and monitoring vendors, the difference is which question gets answered. Those tools are essential for understanding what already happened on the network. Forward Predict answers a different question entirely, what will happen if this change is made, and it answers it with a deterministic proof rather than an educated guess.
Against other platforms using the words “digital twin,” the difference is a decade of foundational work building a model with header space analysis across all major vendors, every device, and every layer, rather than a partial view stitched together after the fact. That foundation cannot be a shortcut, and it is the part everyone else in the category is still missing.
- How do you see Forward evolving in terms of features and roadmap? Could this pre-emptive, predictive approach find a receptive audience in other sectors?
The roadmap moves in one direction, widening the aperture of what gets proven before it goes live. Today Forward Predict proves the outcome of changes to connectivity and security. Next is broader impact analysis.
We are not predicting what may happen, we are proving what a network will do, because a network's behavior can actually be modeled with mathematical certainty in a way human intent cannot.
That distinction is exactly why this belongs anywhere teams run complex, interconnected infrastructure with the same stakes, critical infrastructure operators running their own networks underneath everything else.
Wherever teams are changing complex systems without a way to prove the outcome first, the same problem exists, and the same fix applies.
Where this is headed fastest, though, is autonomous networking. As agents start proposing and executing changes at machine speed, a mistake stops being a single event and starts propagating at scale.
Proof is not an added feature there. It becomes the precondition for letting an agent operate at all.