The cat sleeps more. It jumps up onto things less. It takes longer to get comfortable and doesn’t race around the way it used to. This reads as an obvious story: the animal is getting old. It’s natural. It’s expected. Most owners I know accept this narrative without a second thought because it’s tidy and it requires nothing of them.

What veterinary research keeps saying, with increasing insistence, is that the tidy narrative is almost certainly wrong.

The numbers are difficult to ignore

Studies consistently find that between 64% and 92% of cats over the age of ten have arthritis. By the time a cat reaches twelve, the figure for osteoarthritis alone is around 90%. In cats older than fourteen, 82% show joint involvement in imaging studies. These are not marginal findings from a single outlier study. They replicate across research institutions and countries.

The gap between those numbers and what most cat owners believe about their pets is enormous. And it exists almost entirely because cats are extraordinarily good at not telling you they are hurting.

Why cats conceal pain

This is not stubbornness or stoicism in any meaningful sense. It is evolutionary architecture. Cats occupy an unusual ecological position — they are predators, but they are also prey. A wild cat that signals injury or weakness invites attack. Over millions of years, the ones that survived were the ones that showed nothing.

That instinct does not disappear in a domestic animal sleeping on a heated blanket in a flat in Amsterdam. The cat that is quietly arthritic does not limp dramatically, does not vocalise its pain, does not do anything that reads obviously as suffering. It just… adjusts. It stops jumping to the high shelf. It grooms itself less around the hips. It finds a position and stays in it. To an owner watching casually, this looks indistinguishable from the ordinary changes of age.

Researchers from the University of California Davis and others have noted that owners frequently report these behavioural changes during routine vet visits, describe them as age-related, and don’t flag them as pain indicators — because they don’t know that’s what they are. The cat isn’t complaining. The owner isn’t alarmed. The pain goes unmanaged.

What to actually look for

The clinical picture of a cat in chronic pain is subtle, almost by design. The American Animal Hospital Association lists the signs: reduced activity, reluctance to jump, changes in grooming patterns, altered sleep positions, irritability when touched in certain places, and withdrawal from interaction. None of these is dramatic. All of them can be read as “just getting older.”

The distinguishing question a vet will ask is not whether any of these things are happening but whether they represent a change from the cat’s baseline. A cat that has always been calm and low-energy is probably fine. A cat that used to leap to the windowsill every morning and now doesn’t is a different conversation.

The treatment gap

There is now a real and expanding range of interventions for feline chronic pain — including newer monoclonal antibody treatments specifically approved for cats — but they require a diagnosis, which requires a vet visit, which requires an owner who suspects something is wrong.

That chain breaks at the first link when owners are operating on the assumption that the slowdown is natural. I find myself thinking about how many cats are spending years in unmanaged pain while their owners interpret the stoicism as contentment. The cat doesn’t complain. The owner assumes it’s fine. Nobody’s lying. But the outcome is the same as if both of them were.

If your cat is over ten and has changed its movement patterns in the last year or two, the veterinary consensus is clear: don’t wait for obvious signals. The species you’re dealing with evolved specifically not to give you any.

The post Most cat owners I know assume their older cat is just slowing down with age — veterinary research suggests up to 92% of cats over ten have arthritis, and most are hiding it appeared first on Space Daily.

In today's video I speculate about the future of AI

Ancient Slashdot reader Mark Round writes: Longtime reader here (since mid-1999 -- Hot Grits! Oog the Caveman! Beowulf clusters!), and I can still remember posting back on Slashdot's own 5th anniversary. Time's rolled on: my own blog just turned 25, and it's now roughly 40 years since I first sat down at a computer. So I went digging through archive.org, old backups, and a box of ZIP disks, and wrote up a long look back at four decades of computing through the one website that's been my online home along the way. It runs from my first 8-bit micro and a 1,200-baud modem through discovering the actual Internet at university (and burning far too many hours on Slashdot and sister sites like freshmeat.net), past gloriously pimped-out Enlightenment Linux desktops, all the way to the modern cloud-native world. Plenty of dodgy screenshots, terrible code, and fond memories of long-gone haunts like kuro5hin.org and Linux Coffee Talk along the way.

Read more of this story at Slashdot.

They’re evolving faster than we can stop them

The post How City Rats Are Becoming Resistant to Poison appeared first on Nautilus.

One of the world’s leading technology companies, Nvidia, has just unveiled a new data center cooling system that uses a liquid coolant hotter than a hot tub to cut energy consumption.

The Californian company’s latest Rubin platform can operate using liquid coolant at temperatures up to 113 degrees Fahrenheit (45 degrees Celsius). The approach greatly reduces the need for energy-intensive chillers and cooling infrastructure.

According to the company, the feat is part of the DSX AI factory reference design, which introduces fully liquid-cooled infrastructure throughout the entire system. Every processor, networking part, and high-performance computing element is cooled by liquid in a closed-loop system.

Ali Heydari, Nvidia data center cooling and infrastructure director, noted that the DSX AI factory design reduces nearly all water use and cuts a substantial amount of the power necessary for cooling. “With dry-cooler-based designs, it’s a closed-loop system with no evaporative water cooling – outside of maybe one percent of the year when we might need chillers in some climates,” he added.

A new cooling method

Cooling has accounted for up to 40 percent of total power demand. Traditional data centers utilize large volumes of cooled air to remove the heat generated by servers. This requires extensive cooling infrastructure, like fans, chillers, cooling towers, and carefully managed hot and cold aisles.

Meanwhile, Nvidia’s Rubin platform captures heat directly at the chip level. It can reportedly reduce cooling water consumption from about 2.6 million gallons per megawatt per year to almost zero in suitable climates.

The system relies on a liquid coolant made up of 75 percent water and 25 percent propylene glycol, which circulates through cold plates attached to processors. The liquid enters at temperatures of up to 113 degrees Fahrenheit and exits at about 131 degrees Fahrenheit (55 degrees Celsius), once it absorbs the heat generated by the chips.

Since the heat is removed directly from the source, the processors can continue operating at full performance even with much warmer coolant temperatures than traditional cooling systems would allow.

Lower energy and water use

As per Nvidia, the higher operating temperatures unlock considerable efficiency gains. Industry estimates suggest that increasing chiller temperatures by just 1.8 degrees Fahrenheit (one degree Celsius) can reduce cooling energy costs by four percent. In practice, this means that a 50-megawatt hyperscale facility can save more than USD four million per year in cooling-related energy and water costs by adopting liquid-cooled infrastructure.

“In the right geographic location, with the right system design, you don’t need any refrigeration equipment,” Richard Whitmore, president and CEO of Motivair, Schneider Electric’s advanced cooling division, pointed out. “You can just put big radiator coils outside and use the air temperature for all your cooling.”

The move to complete liquid cooling also required a major server redesign. Earlier liquid-cooled servers used a hybrid approach, cooling CPUs and GPUs with cold plates while relying on air cooling for other components.

Nvidia engineers reworked cooling pathways across the entire server and created a system that uses a single liquid loop to cool multiple high-power components. The redesign has also improved space efficiency. The company revealed that fully liquid-cooled servers can reach higher rack densities.

This allows more computing power to fit into a smaller footprint. “Once the watts per chip crossed a certain level, liquid cooling became mandatory,” Whitmore concluded in a press release.

Named Jalapeño, the new processor was designed specifically for the unique needs of OpenAI's inference systems.

💾

Voyager aux États-Unis, au Canada ou au Mexique pendant l’été implique souvent le même besoin : rester connecté sans se compliquer la vie. Entre la recherche d’itinéraires, les réservations de dernière minute, les messages aux proches et l’usage constant des applications de navigation, la...

En levant les yeux vers le ciel nocturne dans la nuit de lundi à mardi, peut-être avez-vous eu la chance d’apercevoir une brève mais intense traînée de lumière. Mais de quoi s’agissait-il ?

💾

In a laboratory in Broomfield, Colorado, 98 atoms are suspended in mid-air, held in place by electric fields and cooled to temperatures close to absolute zero.

Each atom is far smaller than anything the naked eye could ever see, yet each carries information in a form that has no counterpart in classical physics.

Together, they form Helios, a new quantum computer built by the British-American company Quantinuum. Quantum computers use the power of quantum mechanics, the rules that govern how physics operates at atomic and sub-atomic scales. Those that use Helios’ model of suspended atoms are known as trapped-ion.

A paper published in Nature describes it as a 98-qubit processor with very high accuracy and performance that pushes beyond what can easily be simulated on classical machines. That sounds impressive, but the important question is not simply whether this is a bigger quantum computer (the previous biggest, System Model H2, had 56 qubits). It is whether it is a better one.

Quantum computers are not just faster versions of ordinary computers. The qubits (quantum bits) that they use to process information can exist in quantum states that do not behave like the ones and zeroes of conventional digital technology.

This allows some calculations to be arranged in ways that may eventually outperform even the largest supercomputers. The possible applications are fascinating: new materials, better optimisation methods, improved chemistry simulations and new approaches to cryptography.

The difficulty is that qubits are extremely fragile. They are disturbed by temperature variations, imperfect control, unwanted interactions with the environment and, in some systems, even the act of moving information around the device.

For this reason, the race in quantum computing is not only about having more qubits. It is about having more good qubits, controlled accurately enough to perform long and meaningful calculations.

Why it matters

This is why Helios’ result matters. Quantum computing has been promising to change the world for decades, but many announcements still tend to focus on the number of qubits.

This is like judging a race by the number of runners at the starting line. What matters is how many reach the finish, and in what condition. Helios takes both sides of that challenge seriously. Not only is the 98 qubits relatively large; it also reports very low error rates at this scale.

Errors are more common with quantum computers than with classical ones, so error correction is a big challenge in this area.

The Nature paper gives an average error rate for single-qubit gates of about 2.5 in 100,000 for Helios. A quantum gate is the building block of a circuit in quantum computers. For two-qubit gates in Helios, which are harder and more important for useful computation, the average error rate is about 7.9 in 10,000. This is similar to the best demonstrations of around 5 in 10,000 errors.

Quantum operations are cumulative. A small error in one step may not matter much, but a useful quantum algorithm may require thousands, millions or more operations. Lower error rates mean that more complex calculations become possible before the quantum information falls apart.

Helios’ other notable feature is all-to-all connectivity. In many quantum computers, qubits can interact only with their nearest neighbours, rather like people who can speak only to those sitting next to them. If two distant qubits need to interact, the information must be moved through a chain of intermediate steps. Each extra step adds time and error.

In Helios, any qubit can in principle interact with any other. This is especially valuable for algorithms where the required pattern of interactions does not fit neatly onto a fixed grid.

Quantum railway

The hardware behind this is also interesting. Trapped-ion quantum computers such as Helios use charged atoms as qubits. These ions are held using electric fields and manipulated with laser pulses.

The approach is known for high accuracy, but scaling it up while preserving that accuracy is technically difficult. Helios uses barium ions in what is called a quantum charge-coupled device, or QCCD, architecture. A useful way to picture it is as a tiny quantum railway.

Ions can be stored in memory regions and physically moved into operation zones when the computer programme needs a calculation to be performed using particular qubits. In those operation zones, carefully controlled laser pulses perform the basic steps of a quantum algorithm, known as quantum gates. These gates change the quantum state of one ion, or link the states of two ions together, allowing the computer to process information. In Helios, a ring-shaped storage area and a junction help route the ions around the device.

This separation of storage, movement and computation is not just smart engineering. It is a sign that quantum computing is becoming more like a full computing system, rather than a collection of impressive laboratory components.

The machine also uses software that can make routing and control decisions while a program is running. In practice, this means deciding which physical ion should represent each qubit, which ions need to be moved into the operation zones, and in what order the quantum gates should be carried out. This is important for more advanced quantum programmes, especially those where later steps may depend on measurements made during the computation.

And the paper reports that Helios can run random quantum circuits that would be extremely difficult to simulate on classical machines. That is an important benchmark, but not the same as having a generally useful quantum computer. Random circuit sampling tests the power and complexity of the machine; it does not, by itself, solve problems in medicine, climate science or engineering.

So how big an advance is Helios? It is a serious one, because even if it is not the arrival point of a quantum revolution, it brings together scale, accuracy, connectivity and programmability in one machine.

It is a reminder that transformative technologies rarely arrive in a single leap; they are built step by step, atom by atom, until the impossible starts to look engineered.

Domenico Vicinanza does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.

Asylum accommodation costs have more than trebled to over £15 billion and made a billionaire out of Clearsprings owner Graham King. This scandalous racket needs to end, says Madeleine Gillies.

The post Asylum Accommodation is Ripping off the Taxpayer  appeared first on The Daily Sceptic.

💾

Déshydratation, baisse de tension, insuffisance rénale, mais aussi sensibilité accrue au soleil : certains effets indésirables des médicaments peuvent être amplifiés lors d'une canicule. Un risque encore méconnu qui pousse les autorités sanitaires à renforcer leurs recommandations chaque été.

After years of struggling to curb my screen time, apparently all I needed was a $59 hunk of plastic.