The SR-71 mission set included trans-attack reconnaissance for SAC, where it would overfly a radiating Europe and USSR perform nuclear bomb damage assessment for follow-on strikes. Aside from nuclear bursts and leftover detritus from the ICBMs etc., it was anticipated to operate in an environment where nuclear-tipped SA-5 and other Soviet SAMs would be bursting here and there. It's EMP-hardened.

The Navy has been pushing for a large cruiser for literally decades now. The Trump-class is that under a different name. We're at the end of what we can cram in a hull that's nominally called a destroyer, especially if you want to host larger hypersonic strike weapons and ABM interceptors, plus the radars to manage the ABM battle.

Right but 'battleship'?

That's the same kind of silliness as calling the Russian Kirov class 'battlecruisers' in my opinion. It's not going to stand in the line of battle and exchange shells with other battleships is it?

There have been no battleship-versus-battleship combat since 1944. Airpower made them obsolete.

https://en.wikipedia.org/wiki/Battle_of_Leyte_Gulf

Kirovs are the size of WWI battleships and exchange the heaviest AShM missiles made with other battlegroups, so yes?

At the end of the day the Navy is seeing Orange Man offer a them a check to build the cruiser they've wanted for decades. They'll finally have the space and power to do all kinds of things they can't pull off on a destroyer hull (even if said "destroyer" is really a cruiser in other clothing). If they have to call it a BBN or whatever to make Trump happy they'll do it.

The US Navy needs fewer Kirovs, Yamatos and Ticonderogas. When (not if) the US enters the Pacific theater, they will be going up against Chinese land-based strategic AShMs and potentially 50+ Type 052 destroyers. The moment the US enters any capital ship into this theater, it will be like a 21st-century Battle of Midway with hypersonics, SAR satellites and ballistic missiles.

The Trump-class is pretty clearly one of those interim study projects that only exists to fill a perceived gap in force composition. Unlike the USAF NGAD, I don't think the Trump-class will ever see the light of day, let alone a battle versus a peer-power.

a single hit from enemy anti shipping missles would likely blow the whole thing up.

I agree with the manipulating the nostalgia angle, the real plan just isn’t anything like an actual battleship, and the approval is also because they’re using ‘battleship’ to manipulate the nostalgia.

If things continue they may be everyone's last console. The next gen might effectively be a video player and controller interface for content rendered outside of your home. No physical games (or saved game backups) at all under this scheme, of course.

Ask Google, Microsoft or Nvidia why it's not in fact the future of gaming. You need to be rich enough to have fast internet and pay an expensive subscription yet poor enough to fail to afford a GPU at home, all while accepting low bitrate and high input latency. It's not a large addressable market.

Stadia played Cyberpunk at max settings when none of the home systems could do it. I played it on a Retina iPad, and thought it looked fantastic.

I've heard people play Fortnite Festival (basically Rock Band, and very latency sensitive) over Microsoft/Amazon cloud streaming.

"low bitrate and high input latency" is what you'd expect rendering offsite, but in practice, it seems to work surprisingly well.

Also super fast loading times. Playing e.g. RDR2 on a physical console, by comparison, is painful.

Stadia worked absurdly well. One of the beta games was Thumper, a demanding rhythm game (I might be the reason why it was in the beta lineup actually). How well did that work out? Well, the latency made it hard to play... until I switched my TV to so-called "gaming mode". And then it was fine.

No kidding. The TV was introducing more latency than Stadia itself.

I also once tested it on hotel wifi. It worked great.

(Mind you, this was in Europe. YMMV.)

The business model however was terrible and killed the product IMO. Which, honestly, good: Stadia would have killed video game archiving efforts. If a publisher folds and Stadia and similar platforms stop streaming its streaming-exclusives, then those games are gone. Not worth it, no matter how well it worked.

yeah i felt so dumb with how constantly bad in home streaming was for me with latency and i just could not pinpoint the issue until one day i swapped my tv into gamemode and boom..... felt like i was playing on my desktop

Even if it worked for some games and users - this still demonstrates the fatal flaw of cloud gaming. Total vendor lock in and no user control.

If they decide one day to fuck you over, there is nothing you can do about it.

This is only a fatal flaw to you and me.

If Sony and MSFT decide to go down this route and it's the only way to play blockbuster games, it'll happen whether we like it or not because the market cares far more about playing the latest game than it does maintaining control over its own destiny so to speak. Very few consumers will eschew the Latest Thing and just happily play older games on their own PC (or console) to make a point to the publishers.

> If Sony and MSFT decide to go down this route and it's the only way to play blockbuster games, it'll happen whether we like it or not because the market cares far more about playing the latest game than it does maintaining control over its own destiny so to speak.

Maybe, or maybe it'll live a giant gap waiting to be filled in by a new competitor in the console market.

I still thing that if you burn enough people there will be pushback.

Pushback like there was to making Office 365 or Adobe subscription-only? Or pushback like there was to Google mostly killing AOSP? Or pushback like there was to ads in Netflix & friends? Or pushback like there was to <pick a subscription service> price hikes?

You'll note that all those things are still there, and their competitors are either not there or not significantly bolstered by people disgruntled by those changes.

Well I am sure the subscription shenanigans helped Krita and Libreoffice. Graphene OS seems to be gaining traction & Sailfish OS exists as well.

And making media subscription more expensive will just push more people to piracy, so they will just loose money due to their greed.

How do you pirate GTA:7 when your console never even gets the whole thing in its own memory? That's the point: piracy as we know it becomes largely impossible. Jailbreaking a console won't help. In fact, the publishers could open source the consoles and allow rooted consoles to play, because who cares?

The other point is cheaters. Knowing someone's running an aimbot makes it a whole lot less fun to play online.

> (basically Rock Band, and very latency sensitive)

I contest that rhythm games are the least latency sensitive. You'll automatically adjust to (consistent) latency and not notice it. You also don't get much feedback unless you miss a note.

Festival tells you how far off you are in milliseconds, and gives you extra points for perfect timing.

But that's still not an across the wire latency.

The “perfect” timing window in Festival is +/-5ms from gem position (10ms total afaik). Harmonix has always been exceptional with their beatmatching code. There are physical limits to the black magic of latency compensation, however. If your input has to travel over the wire before the game can see it, then it’s unavoidable without heavy compensation

I tend to think that Stadia (like many things Google) was killed before its time.

It didn't seem any worse than cloud gaming on Xbox/PS Now/Luna/Nvidia/etc.

We have had very... different experiences with stadia. It was not at all useful gor low latency games..

It is not suitable for competition ranked players. The average player will stomp over high ranked players.

Most people are not competition ranked players, they’re people who like to play a few hours of a game now and again.

I agree, but i play a few hours of games, now and again.. competitively.

I played the entirety of Assassins Creed Odyssey on Stadia. It did works surprisingly well. Xbox cloud gaming also worked well for me. It's great for non competitive games.

GeForce Now is doing much better.

Stadia had other issues beyond fast Internet, like asking a community focused on Windows/DirectX to add support for the Stadia/Vulkan stack.

At the last Stadia developers conference, a few months before it was killed, one of the talks was about doing something like Proton for Stadia.

EDIT: Here is the talk,

"How to write a Windows emulator from scratch"

https://www.youtube.com/watch?v=8-N7wDCRohg

At least GeForce allows you to play games you already own, unlike Stadia where you have to rebuy everything. IIRC in GeForce Now its just a matter of signing into your Steam account.

I don’t know much about gaming GPU workloads and how they differ from LLM inference workloads, but I’m definitely rich enough to pay for an LLM subscription and definitely too poor to afford the GPUs it would take to get anywhere near that level of inference.

Gaming is less expensive and more latency-sensitive.

That makes sense. But is the price difference between doing it on home hardware versus cloud hardware less significant than it is for LLM inference? And if so, is it mostly because LLMs need much more VRAM (or more exotic memory architectures)?

I wonder if this will maybe change in the near future or possibly already has? With things like starlink and fiber internet and data centers everywhere absolutely packed full of GPUs it seems like one or two more big players will try at least and possibly succeed with this model.

>You need to be rich enough to have fast internet and pay an expensive subscription yet poor enough to fail to afford a GPU at home

that sounds like a broadening market to me.

Why do you need to have low bitrate?

There are lots of techniques to hide latency.

I don't get the thing about the GPU: even rich people don't necessarily want to bother with owning a GPU thing. And whether fast internet is expensive depends on where you live. Here in Singapore 10 Gbps fibre to the home starts at around 20 USD per month; and we are in a really high cost of living country (we usually topped charts of 'most expensive city to live in').

You're conflating internet prices with cost of living when it's almost entirely unrelated. You live in one of the more densely populated areas around, that is significantly more important. I pay more than $100usd for 100mb bonded DSL because there are probably fewer people in my county than a Singapore city block.

Nowadays the average person can just buy a Steam Machine or something similar and stream to anywhere seamlessly, no subscription required. Most ISP's offer good upload speeds now compared to 10 years ago making it very viable.

Stadia worked great for me on a typical 100Mbps cable connection in the US. Online multiplayer was okay with latency, not wonderful but not bad either.

This has little to do with whether you can afford a GPU and everything to do with whether or not the publisher even makes the properties you want to play available on disc (hint: they won't).

Global access to high speed home Internet is growing not shrinking, there are plenty of ways to creatively manage latency especially when you control the endpoints, and at the end of the day the publishers can effectively force the transition by only making the latest blockbuster games available on the cloud model.

You and I will bitch and moan but the vast majority of consumers (really their parents who pay for them) will be happy their new console is small, light, potentially cheaper, and silent. No more boxes of games to store makes Mom happier. Eventually much of the hardware could be built into televisions from affiliated manufacturers so all you have to do is buy controllers and sign up for the subscription(s).

I still remember the outrage from Half-Life 2 requiring an internet connection to play due to the DRM. Similar arguments were made about games being rented instead of owned, objections about the internet requirement.

I've been saying this for years. These companies don't want you to own the compute hardware. They just want to flash the colored pixels in front of you.

Its true that owning hardware is really important especially at an age where privacy is becoming more and more critical. But the truth is hardware on the cloud is more efficient, it would allow devices to be literally be a screen and a wifi card, you can do anything anywhere. All consumer software is just colored pixels in front of you. To 90% of the planet a truly fast cloud streamed UI would be some much more capable and convenient. With generative AI more and more of our interfaces will also be generative. As the way we input things to most of our devices is literally pressing a location on flashing pixels, best experiences will naturally be optimized around a good screen and connection and that's just the nature of evolution for software being data <-> presentation medium

The main problem with cloud streamed is latency, and the other is fuck the cloud.

I, and probably a lot of the people here, spend so much time on the PC that every additional msec annoys me. If there's keyboard shortcut, then I'll learn it to avoid using the mouse.

Windows being a bloated slow mess even when run locally means that using it remotely all the time would be like torture.

I'm sure dealing with the 4k60 videos from my cameras would be a world of pain on the cloud too.

I don't need or want to pay a subscription for the tool I use for hours a day, with no advantages beyond vendor lock-in.

You need a computer either way... remote is basically all negatives when compared to local.

You are probably correct about 90% however.

90% of people are pretty ignorant about how their computers work, or how to use them more efficiently. They line up like lemmings to lose autonomy over their devices and pay subscriptions in exchange for the barest crumb of convenience.

One has to be naive not to understand that the whole cloud computing model is nothing else than being back at the timesharing days.

The big difference is the hardware evolution, now the rooms are full of server racks with sliding pizza boxes, and the terminals are prettier.

It's never going to work. Why? Fucking physics. Unless games ACUALLY become glorified movies the latency makes multi-player games unplayable and many single player games.

With multi-player games you are running a local simulation and it gets resolved server side and then reconciled locally. This hides lag and works great.

You lose this with streaming. Latency is constant and total. It's just not realistic.

You lack imagination. There's plenty of techniques to improve perceived latency. Especially when you control the hardware at the edge.

You mention one great technique yourself already.

No I'm just actually technical not a larper so I know what I'm talking about. There is no technique for playing a YouTube video of a video game where your controls are streamed to the remote box.

> There is no technique for playing a YouTube video of a video game where your controls are streamed to the remote box.

What does this have to do with anything? Why do you want to restrict yourself to such an austere setting?

> You mention one great technique yourself already.

I believe their point was that this can't be done, since the "local simulation" is no longer running locally; it has to resolve & hide the latency _before_ it's transmitted to you, and you don't know what latency that frame will observe.

You can fake a lot locally with relatively little processing power.

For a slightly silly example, see what mosh is doing to hide ssh latency.

But... the frame is still being rendered behind a hop of latency? How does your local client fix a frame that was rendered on a cloud GPU once it arrives, to mitigate the latency of its arrival?

> But... the frame is still being rendered behind a hop of latency?

Not necessarily, why?

The generalised idea that I see is: 'what can you do with beefy processing in the cloud (but behind a latency hop) and a very small amount of processing power locally?'

A VR example: you do most of the rendering on the server, but you do all the correction for small head movements locally. The server has to send a bit more (and not just as a flat image), but you get over the nausea.

Something like this has been state of the art for ages. And I only say 'something like this' because I don't remember the details.

As a generalised solution if you have crazy amounts of bandwidth to burn: the server is a latency hop behind, so it could send all possible futures that correspond to the inputs that might have happened during that latency hop, and the client picks what to display.

As written, this is crazy. But if instead of all possible futures you have a probability model of human input, and you send the most likely futures, you can probably make it work. Especially if you don't insist on sending raw video. (And even with raw video, you can save a lot of bandwidth by using a special codec that can compress multiple video streams simultaneously by taking advantage of inter-stream redundancies.)

As an example: with this system in the original Super Mario Bros your probability model would most of the time say that the user is most likely to just keep doing what they were doing (running right, or holding the jump button etc). Crucially, in the long run Mario will behave as if he jumped exactly when you pressed the jump button, but you might see a short visual glitch where Mario seems to be doing something wrong or weird, but it'll correct itself very quickly.

Similar to how mosh's predictions work.

If you add a bit of extra smarts locally, you can even add some cheap interpolation between possible futures to try to cover some that the server hasn't sent.

Mostly, what the local player and what the other players are doing will have differ between 'futures', but eg you'll mostly likely to be able to re-use the complex lighting rendering for the fully realistic leafy trees and blades of grass you see in the background between nearly all your possible 'futures'.

None of those techniques look generic though; you could write pipelines in games that worked like this. But you can't (easily) drop it into an arbitrary game.

And while you're right that the computation on the client end could be made lower (depending on the constraints of the game), you're now going to need to care about the hardware of the client beyond "can render a video stream". My (regretfully) smart TV has trouble rendering its own menus, it's not lifting a finger to do a game.

All good points. If you want to this to work across devices, you'd need something like a relatively standard 'browser' on the frontend (and perhaps that could even be the normal webbrowser souped up a bit.)

And for best effects, of course, you'd also want to get support from the game itself. AI coding will make this part easier, because there's probably relatively common techniques you can apply to your games to get most of the benefit.

Just to be sure: I'm just outlining what could be done to make remote gaming viable, even with the latency constraints. I'm not saying that this will definitely happen, nor that it's even particularly likely.

Perhaps even low end local devices will just be fast enough to give people good enough games anyway. Nowadays even phones are capable of astonishing feats.

I bet that would be great (for a specific subset of games). MMORPGs come top of mind.

Yeah, like we don't have enough centralization, no control & force feeding already. No thanks, take your Minitel 2.0 and shove it.

xbox 360 is my last console. disconnected 100% from internet. good times only

Many cameras include microphones that the user has no intention of enabling. These are like that, since they're in a state where no-consent audio recording is illegal.

No audio recordings are made or handed over to anyone. Handing video over to a service provider to do things like weapons detection is not uncommon and not inherently a privacy problem. It's done in hospitals today.

how accurately can a model do lip reading these days?

A friend of mine was involved with AVLIS and Pu-AVLIS. There were/are significant counterproliferation concerns with them. I'm not sure if that has anything to do with the lack of major commercialization, but I wouldn't find it shocking if it were true. As I understand it, it's one of those technologies that the US doesn't strictly need, but we really don't want the Iranians (or similar weapons-pursuing state) to have it.

It's a U.S. national security policy to prevent countries which don't already have capabilities to enrich uranium or reprocess spend nuclear fuel from aquiring such capabilities.

Technologies which have low energy consumption and need only small space and therefor are easier to hide, such Laser isotope separation, US wants to restrict to any country, even allies.

> actual understanding of the pros and cons of union membership

Many of us have worked in $MEGACORPS with substantial union representation among the employees, have seen what it does and how it works in practice, and as a result want nothing to do with it. I wasn't represented due to my job classification, but now due to my firsthand exposure to unions I absolutely never will be by choice.

No way, no how, not ever.

Translated: Management ("wasn't represented due to job classification") doesn't like unions. Shocker.

I know plenty of line level engineers (mostly non-software) who's main interaction with unions is things like waiting for a week to move their desk to the other wall of their cubical because that's a union job and they're not allowed to touch the Ethernet cables.

Notice the story has cubicals, how nice! Most these stories are pretty old. I suspect because there are less jobs in some big F500 campus with colocated engineering, accounting, etc and a unionized manufacturing plant all in the same place.

> open competitive programs based on project quality

This will never, ever happen. There will always be bonus points available, even if they're awarded to "conservative"-leaning feel-good attributes like veteran-owned sponsor businesses.

These investments are likely to always fail at their declared purpose. Better to put the money towards free childcare and maybe trying to convince parents to read to their kids.

> There will always be bonus points available, even if they're awarded to "conservative"-leaning feel-good attributes like veteran-owned sponsor businesses.

Why are you singling out "conservative" causes?

The government should limit innovation and direct resources to proven ROI spending such as free daycare, nutrition, and public infrastructure

Your consent is given when you choose to pay for a ticket and attend an event there.

For publicly-funded venues, you may have a point. But if Bimbo's or the Warfield in San Francisco wants to ban someone and it's not for a reason prohibited by law, more power to them. Casinos have been doing this for many years, and generally speaking businesses have the right to refuse business to anyone. Make them post signs if that makes you feel better, but they should be able to ban, say, a person who attacked a patron or staff at the venue (and then was swiftly let out of jail by an activist judge).

Funny you mention San Francisco since 1015 Folsom and other places now use facial recognition to bar entry to (what I suspect are) bad actors. I suppose it was inevitable that as we allow more bad behavior through diminished societal enforcement that this would move into the private sphere. It's interesting to observe that places like San Francisco have moved into this kind of multiple-parallel-government structure.

Bad actors?

Oh no.

There was a fight one time in the place? Do you mean bataclan bad actors?

The claim of a few bad apples is a ridiculous claim to require facial recognition. Prior to facial recognition, situations were dealt with. It's not a new problem, nor did a hard ban policy solve the problem. After, facial recognition this seems to be a fear mongering technique that just allows for customer profiling even more than it did before.

> AI in its current form has limited usefulness for most people.

That's not what I'm seeing. My mom always wanted Google to just answer questions, and now ChatGPT can. She uses it enough in her daily life that she bought a subscription.

Yes, she knows it hallucinates and you have to double check everything, but so far she finds a ton of use for it even with those caveats.

Now, I agree that a personal servant robot would get a ton of business. Even at new-car prices, it's still cheaper than a human caretaker/maid/butler/etc. And the maid usually doesn't also mow the lawn on a hot day, while a robot would potentially do all kinds of different things without complaint.

> That's not what I'm seeing. My mom always wanted Google to just answer questions, and now ChatGPT can. She uses it enough in her daily life that she bought a subscription.

Reminds me a lot of AskJeeves :)

Google summaries from websites mainly did this before “AI” came around. Now I just don’t know the source unless I dig myself. Would rather it went back to before, or at least they implemented the veracity checks better.

I thought it was interested that the movie looper had argidrones, but what inspired that wasn’t part of the training set or discussion online. I get remarkably bad answers and search results which seem misleading at best.

> Now I just don’t know the source unless I dig myself.

Claude includes links to references.

What's a problem though is that while it includes a reference, its distillation from the source, or worse, from the combination of sources, is often plain wrong. I check them regularly, and it made me very distrustful of Claude's capacity to faithfully summarise or explain from a source.

When asked to give specific links, it's usually even worse.

I don’t disagree. Many times when I’ve checked the source links a language model provides, I get concerned. It’s definitely important to stay vigilant and exercise critical thinking, but the model’s ability to synthesize information and converse with it in natural language still blows my mind sometimes

So does ChatGPT. The hallucination problem is not completely solved, but much better than a few years ago, especially if you use reasoning mode, where it’s more likely to spontaneously do a web search

Those references often have no actual relation to the answer. A surprising amount of the time, if you check them they’ll prove it wrong.

I have a feeling that humanoid robots will have other, more intimate, tasks first. Our most primal drives seem to drive the advancement of technology.

Also, mowing the lawn on a very hot day is pretty bad for the grass.

I knew someone was gonna go there.

Yeah, I felt that someone had to speak up about the health of the lawn.

I think more non-tech workers are buying subscriptions than people in industry.

Neither I not almost all coworkers I know do not pay for AI subscriptions (or companies do).

But quite a few people’s sisters and parents and friends are paying $20-200/month for ai to help them with their “projects.” Whatever that means.

FWIW, Google does just answer questions now.

We already have robot grass trimmers and they work pretty well. Why would you want a crude, inefficient, facsimile of a human push or power an inefficient form of mowing the lawn?

Cameras worked pretty well, but most people take photos with their phones nowadays. Pro photographers still mostly use cameras though.

I can envision a future where people have a humanoid trimming a small backyard but at the same time the maintenance of e.g. a golf course would be done by dedicated robots.

EDIT: my point is that just like smartphones replaced a number of specialised devices for most people, a humanoid robot could do the same, by virtue of being a general purpose machine

I don’t want a crude facsimile.

I want a pretty good facsimile that can cook, clean, mow, carry stuff around the yard.

I want this much more than a robot lawnmower + robot vaccuum + plus other single purpose automatons.

For when they don't work pretty well and for a thousand other things besides trimming grasss.

Because houses and tools are made for humans.

You're honestly comparing a computer that can poorly answer questions to a fully functioning robot that does chores?

Well, one of those is already a reality in each persons pocket, and the other is a vision needing a lot of money and effort to implement at all

That's laughably false propaganda. Maybe one person one time stole something for jail health care, but the hundreds of thousands of other shoplifters do it for other reasons, ranging from being unable or unwilling to control their childish I-want-that impulses to being part of organized crime rings.

To pretend shoplifting at its core stems from not having socialized health care is the height of chicanery.

I mean, yours is just as false by putting everything in a different container.

The issue is all this stuff is complex and multiple causative.

For example metal thefts are by far going to be executed by two groups. 1. Criminal gangs stealing large amounts at once. 2. Methheads looking for anything they can sell for their next fix.

Group 2 can certainly be helped with clinics and healthcare.

I find it bizarre you imagine a world where immoral or amoral persons are somehow not at fault for their own behavior. When some turd goes into a Walgreen's on Market St. in SF and fills a bag with goods to resell, they aren't doing it because society made them do it. They aren't stealing to feed themselves or their kids. They're doing it simply because they can and they don't find anything wrong with it.

The drug users stealing copper can only be helped if they truly want the help. A great many are not ready for that help yet. As San Francisco has amply shown, no amount of money poured into the existing homeless industrial complex will change that fact.

See, I imagine you live in a black and white world where people are good and bad. Bad people do things like steal from walgreens and should be punished with death. Good people simply make mistakes, like steal hundreds of thousands via wage theft from their employees, and when they are caught they should be forced to pay 1% of that as a fine.