This is promising. Finally competitive ARM design on server that can be purchased rather than renting on AWS. Although Amazon is also selling their Graviton but don't believe that would be a quick win simply because Nvidia has much better distribution channel.
What remains is Watt during benchmarks and price. Nvidia already put $20B on CPU sales on their book. If they could continue this momentum the era of ARM on server will finally arrive.
I mean I am happy if they kept FreeBSD to be Server focused. I have been using a Mac / Windows and deploying on Linux and FreeBSD, i don't see why both the consumer and the server / enterprise has to have the same OS stack all the time.
I don't see why the operating systems have to be different.
One should be able to run a GUI on a "server," if they choose to do so. It's not arduous; here in 2026, servers are allowed to have GPUs. It's really OK. (My mom says we're even allowed to run LLMs on the FreeBSD server-box in her basement.)
One should be able to run a stodgy, reliable database on a "desktop," if they choose to do so. That may be best with a good filesystem, redundant storage, and some ECC RAM. But it's good for desktop systems to have these things. (And ZFS is a built-in, first-class filesystem on FreeBSD.)
I mean... It's not like we're talking about the difference between an IBM Multiprise 2000 and an SGI Octane here. Those days are over. We're mostly just using PCs for all roles these days.
These PCs run the same code in the same ways, whether packaged and sold as "server" or "desktop" units. The CPU parts are frequently even cut from the same literal cloth: A "desktop" Ryzen CCD and a "server" Epyc CCD are born on the same wafer before being packaged up differently.
The line betwixt server hardware and desktop hardware is presently murkier and less-defined than it ever has been before. Why should the operating system be different?
on a server a GUI is optional. on a desktop/laptop a GUI is mandatory. supporting a GUI is more work. so if you are server focused, then why bother? making that effort only makes sense when you actually do want to also support desktops. and of course you can use a desktop device as a server. all my home servers so far have been desktop devices. but they never had a GUI because i had no use for it.
Agree. Other than drivers, which Server OS will have less to deal with. I also wanted other point out GUI is the majority of the work and polish on consumer OS. Far more than what ever kernel, driver, filesystem sitting behind it. And that is ignoring the apps.
Consumer OS is basically totally different beast, it is not that I don't want FreeBSD to go down that path. But given the limited amount of resources I would much rather they focus on doing something great.
It is one of the thing with consumer when they remember they brought it at the absolutely lowest price point when DRAM maker were bleeding money.
Those are not normal pricing. Before the pricing collapse in early 2020, 96GB DDR5 would have cost about $450 to $500. And I will need to restate again the cost of DRAM hasn't really changed much in the past 20 years. Its price just goes up and down in cycles.
So in reality it is more like going from $500 to $1300. But consumer felt it was more like going from $200 to $1300.
Crucial are already selling DRAM made by CXMT. And China are already throwing money at it. I doubt the memory bubble will burst in next 12-24 months. As in going back to money losing DRAM pricing. As they will all pivot to HBM or other money making products. But the bulk of lower end consumer DDR5 or LPDDR5 will goes to Chinese Foundry. Assuming they have figure out how to do them well. Which they have improved but are still so far away from industry leaders.
Normally memory maker will push the next DDR standard to market just to push out Chinese competitors, I am not sure it will work the same this time around. DDR5 have plenty of other usage / demands.
Historically the price has always trended downward. When I first got into computing $200 could buy you 128 MB (yes M) of ram. Really nice systems had 512 MB.
That's obviously changed over the decades as process shrinks have lead to higher memory density. We should generally expect that ram will cheaper up and until the point where process shrinks stop happening. They've definitely slowed, but they haven't stopped.
Exactly. My first computer had 48k, yes K of ram :-). My first PC has 2MB and made all my friends jealous as they had 1MB. Amiga 500 at the time had half.
I am keeping a piece of paper that came with my Tex Murphy game which stated that one could get 32MB of RAM for as little as $700 (1990s dollars) which would drastically improve the game!
>They've definitely slowed, but they haven't stopped.
Yes if you span into 40 years. But the spot price for DRAM floor was ~$2/GB in 2008 and touched that 2-3 times over the next 15 year. It wasn't until early 2020s it broke that into $1.
Process shrinks happen but majority of DRAM part can't be shrinked by process any more.
>Did dhh provide a recipe to install hyprland properly without having to install a full "distribution"? (I don't know, it's a real question)
I would guess in typical DHH fashion he would say it is Open Source. And I don't understand where this just Arch + Hyprland installation is coming from?
They have also customised the OS / distro so it install in less than 2 min on a super fast USB. Getting Laptops, both Framework and Dell are now on board, tested on Omarchy so they work out of the box. And so many other tiny things that just make the experience better. I say better but to most consumer, those are expected in the first place.
And this "expectation" people have been waiting for more than a decade.
> Getting Laptops, both Framework and Dell are now on board, tested on Omarchy so they work out of the box. And so many other tiny things that just make the experience better. I say better but to most consumer, those are expected in the first place. And this "expectation" people have been waiting for more than a decade.
As a fan of boring Dell laptops/desktops and owner of many, I can tell you they have been well supported in every distro I have tried (Debian, Fedora, Arch, SUSE)
According to Chrome Stats from Google 2019 [1], ~80% to 85% of images served are above bpp 1.0 ( Bit per Pixel ). Around ~95% are above bpp 0.5. I doubt this have changed if not gotten worse as images gets larger over the years.
There are images such as logo or specific patterns that works a lot better and compressed at low bpp ( below 0.5 ). But those are rare on web, and in the case here, most of the images are photography.
Which means at sub 0.3 bpp it is a ridiculously low bitrate even for Web photo.
HEIC on iPhone 17 is based on HEVC, H.265 hardware encoder on iPhone 17.
VVC / VTM is H.266 Codec, the successor of HEVC / H.265 most people may have heard of one of its encoder called x265.
ETM is H.267 Codec, currently the best in class video compression codec that is still in development.
I assume everyone knows about AV1 and AV2 already since we are on HN.
CPU Encoder, generally speaking produce better image quality while hardware encoder tends to be fast but lower quality. Both HEIC and AV1 are based on hardware encoder on the iPhone 17. ( At least that is my read of it )
In case anyone wondering. JPEG XL is not designed or yet to be optimised for low bpp. It excels at 0.8 bpp onwards depending of type of image. So the result of XL would likely be very bad. Similarly to normal JPEG and H.264 encoder.
I am wondering if this image codec is deterministic.
I would imagine if we use this on the web the actual image size would be a lot smaller. Meaning a lot of the artifacts we see shouldn't matter. And clicking on it would bring up a different image file.
There is finally a possibility in the future some half decent image could be included within 14K frame of the webpage.
Encoding and Decoding speed is actually useable on today's hardware. Decoding in sub 100ms on iPhone 17.
And on another notes, VVC is doing extremely well in terms of compression rate and encoding / decoding complexity.
Bjontegaard Delta-Rate (BD Rate) metric, proposed in 2001 by Gisle Bjontegaard, is a method for calculating the average difference between two rate-distortion (RD) curves.
It is extremely common in codec comparison, along with terms like PSNR, SSIM and VMAF ( which is newer and developed by Netflix so it tends to get explained a bit more )
>’m certainly curious if this is aimed at becoming the new default format for Apple devices.
I certainly hope not. Not unless it is deterministic and much much higher quality.
> I certainly hope not. Not unless it is deterministic and much much higher quality.
You're not comparing fairly. The author is intentionally using low-res images to illustrate how the compression works. You should compare these to, say, a JPEG compression at the same resolution and same bitrate. I think you'll find that this technique is quite an improvement to the compressions you already know and love.
JPEG has the great advantage that all JPEG artifacts look like JPEG artifacts. Newer codecs create artifacts that can be mistaken for part of the original image. That's a heavy price to pay for improved compression efficiency.
China might not win on GPU or Software Stack, but they will definitely win on the lower end of the stack which is Datacenters and Electricity Grid infrastructure building out. And I wouldn't be surprised they are 10x the rate of US.
Putting the human freedom issues aside, I don't see how Datacenter and Electricity production have much environmental or social concerns especially when they are built in mostly remote places. This isn't some manufacturing facility that leaks toxic substances into river. And China's annual solar installation rate is more than the rest of the world combined.
>Economically it might be more successful
Economics is what gives you power. US needs to wake up and start taking it seriously rather than feeling good about themselves being the best country in the world.
> US needs to wake up and start taking it seriously rather than feeling good about themselves being the best country in the world.
If they consider themselves the best country in the world they need to wake up anyway :) :) :)
Honestly, my boss approached me last year reminding me there was this company program about sponsorship for moving to the US (it is how he got there too) and I really had to make an effort not to laugh. Sure they will pay 3x as much as where I live but really, no. I'm very progressive, socialist, pro-LGBT etc. I would hate it there. I won't even fly there to visit at this point.
I would consider China for tourism much more at this point. They have this cyberpunk city I'd love to see. But thanks to Trump flying across the middle east is heavily disrupted so not this year.
Theres this weird false dichotomy people do in energy discussions where they imply that you can either be an authoritarian regime with no property rights and build power plants, or have a rule of law and not build any energy infrastructure at all.
Don’t get it twisted. Their superpower is that they actually just build solar panels, wind turbines, and nuke plants. The answer to should we build, is “yes.”
BTW, look at what is happening today, under this administration, at the US EPA sometime.
This is a two part question, the system and the idolise part aren't related.
You are ignoring a lot of quality Japanese product have been producing for more than 50 years, which answer partly why idolise Japan question. But the so called squeezing of subcontractors is true as of any other American companies if not worldwide and yet they failed ( or mostly failed ) to achieve similar standing to Japanese counterparts in terms of quality.
This is a noob question but wondering if anyone here could answer.
There are plenty of choices for Small and Medium size plane as well as private jet. Why are most commercial airline only buying Boeing and Airbus? And why others aren't making bigger planes to compete?
Industrially boeing / airbus are 100k large companies, i.e. need 300m+ base country size (or in EU case, block size) to support specialized workforce large enough for modern commercial aviation. Economically, fuel costs, i.e. engine maturity makes any entrant that's doesn't have parity engine core tech automatically none viable because simply higher costs due to lifetime fuel costs. (Geo)politically, there's lots of certification / safety drama that incumbents like US/EU will throw to undermine competitors so really also matter of geopolitical power - i.e. apart from engines, PRC COMAC using western components mostly for easier certification, they have large enough internal market to sustain development against economics. Almost no one else has those conditions, except India but they don't have industrial base.
Since topic is tankers, PRC/RU has their own tankers, i.e. for military aviation it's not strictly as difficult since fuel cost less issue. But for strategic aviation (transport/tanking) big efficiency working with commercial chassis / turbofan efficiency.
> Economically, fuel costs, i.e. engine maturity makes any entrant that's doesn't have parity engine core tech automatically none viable because simply higher costs due to lifetime fuel costs.
Neither Boeing nor Airbus make their own engines. They get them from CFM, GE, Rolls-Royce. Like everyone else. That's not the differentiator.
But it just costs insanely much to get an airliner certified. Even Boeing has been held back by its clients demanding a shared type certificate for the 737max which caused all those deaths due to mcas.
There were of course more players but they've been bought up. And some emerging brands that are excluded from our markets due to sanctions like the Chinese Comac and the Sukhoi Superjet. The superjet is particularly affected because some of its systems were designed by western companies like Honeywell and they've had to make last-minute replacements after the Ukraine war that didn't exactly work out well.
And there's some other players. Embraer is creeping closer to the 737/A320 market.
But anyway so it isn't just Airbus and Boeing really.
It's differentiator in sense engine manufactures are part of western aviation industrial complex and can limit access, i.e. COMAC sanctions. Unless you're western core, you do not have guaranteed access. I think Embraer is creeping on to bottom end of narrowbody, but they're far from wide body. And with respect to topic, modern strategic tanking are wide body size, i.e. Embraer E190 is not same heavy league as KC46, MRTT, Y20, Il-78, and Embraer has not demonstrated ability to go beyond regional narrowbody. Fielded widebody options is really just Airbus/Boeing duopoly, AVIC/Xi'an, UAC stumbling along from legacy USSR stack.
I travelled on an E195 recently and I have to say it was much more comfortable than any 737 or 320/321 I've been on. Unlike other regionals I've travelled on in the past which were noisy, cramped and bumpy.
Their squircle fuselage works out really well. Really next-gen which makes sense because the Boeing and Airbus designs are decades old. I think they certainly have the chops to scale it up that bit to reach regular narrow body. But I imagine they just don't want to because that might mean trade sanctions from the US.
But I think the restrictions are more artificial/protectionist than actual knowledge. I'm sure sooner rather than later we'll have comacs flying in Europe. The Russian ones depend really on whether they resolve the war situation.
There is a third, Embrarer. They have most of the market in small regional jets in some cases, but those are in reality very different than say a 777 or 787.
These two choices are conglomerates of what used to be a much larger set of manufacturers. In short Boeing, Airbus and it's suppliers are basically what is left of all the old big aerospace manufacturers.
Indeed. Embraer (Brazil) does jetliners carrying up to around 150 pax. So did Bombardier (Canada), though they sold their C-Series to Airbus (now the Airbus A220). Then there's COMAC (China) and UAC (Russia; also a conglomerate of Sukhoi, Tupolev, etc.).
These compete with the smaller versions of the Airbus A320 family (like the discontinued A318 "Baby Bus") or Boeing B737 family.
So, in that narrow-body and regional jet segment there are a few players.
But in the big wide-body (=2 aisles) long-range jets, there's only Airbus and Boeing.
Another manufacturer of regional aircraft with a significant market share is ATR, though it makes turboprop airliners, so not exactly the same category as Airbus A220 and Embraer E-Jet.
For private jets, in order of most to least deliveries per year (number per year in parenthesis):
Cessna (171)
Gulfstream (158)
Bombardier (157)
Embraer (155)
Cirrus (106)
Dassult (37)
Honda (12)
Which is nothing compared to:
737 (447)
767 (30)
777 (35)
787 (88)
These two choices are conglomerates of what used to be a much larger set of manufacturers
This. The entire market has been allowed to be monopolized through mergers and buy-outs. Russia used to have their own aerospace industry (and that fleet was reliable enough to be allowed to fly in Europe) but then Russia happened.
>Russia used to have their own aerospace industry (and that fleet was reliable enough to be allowed to fly in Europe) but then Russia happened.
It's absolutely irrelevant what Russia did or could have done here in this industry.
Same with Chinese planes. If they ever manage to make a competitive passenger plane, it will not be allowed certification by US and European authorities purely for political reasons, the same way how their EVs are not allowed for sale in the US or how they aren't allowed to have ASML EUV machines. This isn't a fair game, never was.
The decisions on purchase of aerospace units is 90% (inter-)national politics and only 10% meritocracy, since both Boeing and Airbus are massive defense players making advanced killing machines, and no country wants to directly or indirectly fund the defense industry of their geopolitical rivals.
When a third country needs to chooses between Airbus or Boeing for their flag carrier fleet, they don't objectively compare the operational history and tech specs of Airbus vs Boeing and make the decision based on that, they just ask themselves "do I want to be in bed with EU-France or with Uncle Sam as my main ally and provider for the next 30+ years". Hence why most oil-rich middle eastern states chose Boeing as the US is their main defense provider anyway and don't want to anger them, especially when Donald Orange makes a visit to your state.
That's just how politics works when you operate at that level. Handshakes, dinners and bribes. Always has.
Cost and complexity of certification? At least that was why Mitsubishi MRJ never went into production. They got the first prototype flying in 6 years, then sought to obtain certification for subsequent 9 years through various means before giving up and scrapping all ~10 examples. Kawasaki P-1 that flew in the overlapping timeframe is in production and in service, with IHI made indigenous engines albeit with teething issues, so it's not like planes and engines can't be made by anyone other than existing players. They just can't be sold, and therefore can't be* done.
What remains is Watt during benchmarks and price. Nvidia already put $20B on CPU sales on their book. If they could continue this momentum the era of ARM on server will finally arrive.
reply