HNHacker News
TopNewBestAskShowJobs

arghwhat

9,034 karma · joined January 12, 2017

no.
submissionscomments
arghwhat··on Liquid Glass? That's what your M4 CPU is for
> I was under the impression that for GPU accelerated GUIs, all windows are rendered to a render target.

Each window renders to one or more buffers that they submit to the display server, which will then be either software or hardware composited ("software" here referring to using the GPU to render a single output buffers vs. having the GPU scanout hardware stitch the final image together from all the source buffers directly).

Note that in the iPhone cases, the glass blur is mostly an internal widget rendered by the app, what is emitted to the display server/hardware is opaque.

> It might be that windows underneath have gone to sleep and aren't updating,

The problem with blur is when content underneath does update, it requires the blur to also update, and rendering of it cannot start until the content underneath completed rendering.

> Software renderers typically do the optimisation you're suggesting to reduce on memory and CPU consumption,

I am solely speaking about GPU-accelerated rendering, where this optimization is critical for power efficiency. It's also required to propagate all the way down to the actual scanout hardware.

It also applies to CPU rendering (and gpu-accelerated rendering still CPU renders many assets), but that's not what we're talking about here.

> I think some VNC-like protocols benefit from this kind of lazy rendering, but the actual VNC protocol just diffs the entire frame.

Most modern, non-VNC remote desktop protocols use h264 video encoding. Damage is still propagated all the way through so that the client knows which areas changed.

The frames are not "diffed" except by the h264 encoder on the server side, which may or may not be using damage as input. The client has priority for optimization here.

> Their very nature makes the kinds of graphics demands of a 2D GUI very negligible.

An iPhone 16 Pro Max at 120 fps is sending 13.5 Gb/s to the display, and the internal memory requirements are much higher. This is expensive.

Not rendering a texture and being able to pass it off to scanout hardware so that the render units can stay off is the difference between a laptop giving you a ~5 hour battery life and a 15-20+ hour battery life.

The GPU could texture your socks off, but you're paying a tax every microsecond your GPU's render units are active, which matter when you're battery powered or thermally constrained. This is why display servers and GUI toolkits go through lengths to not render anything.

arghwhat··on io_uring is faster than mmap
mmap isn't "serial", the code that was using the mapping was "serial". The kernel will happily fill different portions of the mapping in parallel if you have multiple threads fault on different pages.

(That doesn't undermine that io_uring and disk access can be fast, but it's comparing a lazy implementation using approach A with a quite optimized one using approach B, which does not make sense.)

arghwhat··on Liquid Glass? That's what your M4 CPU is for
As long as the lower blur layers are not fully occluded by opaque content, then yes - they all need to be evaluated, and sequentially due to their dependency. This is also true if there is transparency without blur for that matter, but then you're "just" blending.

Note that there are some differences when it's the display server that has to blur general output content on behalf of an app not allowed to see the result, vs. an app that is just blurring its own otherwise opaque content, but it's costly regardless.

(There isn't really anything like on-screen vs. off-screen, just buffers you render to and consume. Updating window content is a matter of submitting a new buffer to show, updating screen content is a matter of giving the display device a new buffer to show. For mostly hysterical raisins, these APIs tend to still have platform abstractions for window/surface management, but underneath these are just mini-toolkits that manage the buffers and hand them off for you.)

arghwhat··on Liquid Glass? That's what your M4 CPU is for
The usual mechanism is to mark widgets that changed dirty, accumulate the bounding boxes of such dirty areas, take the next swapchain buffer and get its invalid regions, iterate through the widget tree and render anything that intersects with the bounding box or invalid regions, and submit the buffer + the dirty areas to the display server/driver.

And yeah, having a render step depend on the output of a previous non-trivial render step is Bad™.

arghwhat··on Liquid Glass? That's what your M4 CPU is for
You misunderstood, this is not about computing individual pixels but only selective rerendering graphical elements which have been changed, and in turn figuring out the total area of change. This propagates through the entire stack to let the GPU scanout hardware know which tiles have changed, and allow partial panel self refresh updates (depending on hardware).

Rendering is still done in bulk for the changed areas, avoiding rendering expensive elements (e.g., transformed video buffers, deeply layered effects, expensive shaders). It's a fundamental part of most UI frameworks.

arghwhat··on Liquid Glass? That's what your M4 CPU is for
The problem with these kinds of blur effects is not the cost of a gaussian blur (this isn't gaussian blur anyway as it has a lens effect near the edges). It's damage propagation and pipeline stalls.

When you have a frosted glass overlay, any pixel change anywhere near the overlay (not just directly underneath) requires the whole overlay to be redrawn, and this is stalled waiting for the entire previous render pass to complete first for the pixels to be valid to read.

The GPU isn't busy in any of this. But it has to stay awake notably longer, which is the worst possible sin when it comes to power efficiency and heat management.

arghwhat··on Liquid Glass? That's what your M4 CPU is for
Blur is expensive because it propagates damage and has to wait for all previous rendering to be complete.

When one pixel changes underneath a blur the entire blurred area needs to be redrawn, meaning that all elements on top needs to be redrawn. As the blur cannot render before the underlay is finished, the graphics pipeline is stalled. Fancy blur look past the area immediately underneath to more accurately lens effects, meaning each output pixel reads a lot of input pixels.

When it comes to power and heat management, the goal is to be able to power the GPU down again as fast as possible, and this kind fo thing prolongs it quite a bit. There may be a point where efficiency makes the result acceptable, but it's always going to be much worse than not doing that.

arghwhat··on Nintendo Switch 2 Dock USB-C Compatibility
This is not true, DisplayPort and HDMI is not compatible, but the dedicated converter chips are quite efficient at their job. One is present inside every USB-C to HDMI cable, dongle or dock, as there is no such thing as USB-C HDMI alt-mode. Heck, even laptop HDMI ports are usually implemented through a DP converter.

You might be thinking of DisplayPort++, which is a special port supporting both DisplayPort and HDMI, allowing passive adapters. This is not available in the USB-C DisplayPort AltMode.

arghwhat··on Nintendo Switch 2 Dock USB-C Compatibility
It's not. Every USB-C to HDMI cable in existence is a DisplayPort alt-mode cable with a DP to HDMI converter at the end. They don't require active cooling, even if they get a little warm.
arghwhat··on Nintendo Switch 2 Dock USB-C Compatibility
Switch currently runs in one of two modes: Docked and handheld. Games are written to expect those two modes, with rendering adjusted to those specifically and nothing else. Think of all the render settings in a PC game (in reality there are way more things the developer can adjust, but just for the sake of illustration), and assume there are instead two hardcoded presets which are selected based on whether or not the device is "docked".

For Switch 1 games, docked means "we're running as fast as the internal cooling can handle and outputting 1080p, assume the user can't use the touch screen and such". A dongle is fine here as long as you don't block the air inlets or exhausts, or place the Switch on something heat sensitive.

For Switch 2 games, docked means "we're running as fast as the externally boosted cooling can handle and cranking out 4k, and assume you can't use the touch screen and such". Assuming the fan does useful work here, then with a dongle you'll be thermal throttling and have a bad gaming experience - on a PC you'd turn down the settings, here those are hardcoded.

So, what about using the switch 2 handheld mode on a switch 1 dock, seeing it's the same resolution? Well, if you tell it to run in handheld mode it might assume you have access to the touchscreen, always-on VRR and HDR, is using internal speakers, etc.

It's totally doable, but to do this right the stack needs to be prepared for it. As such, I understand why they did it.

Heck, this was the same issue for the Switch 1, and it took a while for third-party docks to pop up. It'll take all but a moment for third party docks to support this.

arghwhat··on Nintendo Switch 2 Dock USB-C Compatibility
There's a lot of misunderstandings about USB PD communication, in particular Vendor Defined Messages. The LTT video kept making this mistake, and mixing in misunderstandings around messages to the eMarker chip itself (SOP'). It was a painful watch.

Vendor Defined Messages have is part of any normal PD exchange, as they're simply anything that isn't defined by the PD spec itself. You'll see VDMs when connecting any device supporting more than just dumb charging, as it's used for all sorts of things like DisplayPort, Thunderbolt, eMarker identification, etc. - stuff we'd expect ehre.

The quote from Antak refers to just a single, possibly/likely proprietary, message. This could be to ID the dock on the basis of e.g. rejecting the switch 1 dock should one cram it in, or to reject switch 1 dongles. Maybe it's Nintendo speak for "dock capabilities: cooling", with the switch having no mode for docked gaming with reduced performance without cooling.

Intentional incompatibility, yes, but it's 1 message of an unknown type within bog standard USB-PD, not a "vendor defined lanugage" or "over 30 proprietary messages".

arghwhat··on Toyota is recycling old EV batteries to help power Mazda's production line
> Generally speaking the cells that are welded on are designed to be welded on in the areas were you do the welding.

... by automated spot welder programmed to the specified timing and temperature control from the cell spec sheet, in a controlled environment with suitable protection and fire suppression for a battery manufacturing line. Not by a hobbyist's first try with a homemade spot welder and a safety squint.

I have made such spot welder and done such spot welding. Sure it's fun to do stupid things, but it remains stupid and unnecessary. For a homebrew battery bank, this is the wrong tool, wrong cell and wrong chemistry.

Buy premade, or if you must, buy boltable prismatic lifepo cells. They can dump a lot of power if your short them, but you can drill straight through them and they'll remain stable. The random 18650 li-ion cells... Not so much.

arghwhat··on Toyota is recycling old EV batteries to help power Mazda's production line
If you don't know what you're doing, don't do this. Even if you know, probably don't. It works, and is a regular industrial process, sure, but you're trying to perform controlled melting of the protective housing of one of the many, tightly packed chemical energy bombs you're sticking together. Doesn't take too much of a mistake to go very wrong.

If you're building a battery pack in this day and age, use something like LiFePo prismatic cells and bolt-on busbars instead - way less dangerous chemistry, way less spicy process - but realistically speaking just buy the premade packs. For normal sizes, they're not more expensive (but don't buy the "too good to be true" ones) and means not having to deal with entirely unprotected battery terminals eager to give you a Very Bad Time.

arghwhat··on Toyota is recycling old EV batteries to help power Mazda's production line
A degraded battery bank does not mean a bank with outright "bad" cells. The cells will probably be way more off than they used to be, but there can still be plenty of effective capacity in the bank. Heck, if space isn't an issue, it's productive as long as it isn't self-discharging too fast.

You can still have a working battery even with some "bad" (i.e., way out of spec) cells, depending on the BMS. All the thresholds are configurable, just that a regular EV setup would lean towards safety.

arghwhat··on Temporary suspension of acceptance of mail to the United States
> Expats often chose their location based on such factors.

You're not answering the question I wrote. Recreational or investment properties in foreign countries, i.e. not a peron's only property in their home country.

Expats pick a country to live and work in. They don't get to pick different countries for income tax, for property tax, for company tax, etc., like a billionaire being a US citizen with a swiss bank account, luxemburg company, luxury vacation homes in e.g., gated communities in Morocco.

(Also note that this isn't a discussion of just property tax, but billionaires avoiding tax in general.)

> Good luck finding someone who actually pays zero property tax on US land they own

Observe the point above - US citizens owning land in foreign countries.

Also, US citizens paying way less by severely misreporting and paying people to look the other way, like the earlier example.

> Comparing wealthy taxation is only meaningful based on percentages.

All taxes should be fixed, unavoidable percentages yes, that is my original argument. Progressive tax doesn't fix things because it doesn't hit the ultra wealthy like people think, as those evade the flippin' thing anyway. Deal with the corruption and evasion, get some proper (and fixed) rates to apply to their business, and you won't need complex progressive tax systems.

arghwhat··on Temporary suspension of acceptance of mail to the United States
> Regular people also commit fraud which can make some differences here.

Yeah, but it's hard for a regular person to under/overreport a property by, say, 200 million USD, and most people do not have the power to negotiate with or bribe municipalities to get different tax treatment like big companies and the rich people owning them do.

> Unpaid property taxes result in forfeiture

Avoided taxes are not unpaid taxes - we're not speaking of people on the run from debt collectors, but people who cheat their way out of having to pay anything.

> So do regular people

What regular people own properties in foreign countries selected for their tax benefits, for investment or recreational uses?

Most regular people own between zero and one property. Granted, they could move, but it takes extra riches to be able to casually straddle multiple borders.

arghwhat··on Chinese astronauts make rocket fuel and oxygen in space
No no, you integrate the pump directly into the tree. Skip the farms, just plant the trees at the station in place of the current pumps.
arghwhat··on We regret but have to temporary suspend the shipments to USA
Shock and uncertainty causes large dips in the stock market. As as long as the economy isn't completely toast by the time you backpedal completely, it'll recover somewhat.

Just imagine the money you could make if you could induce such events on demand, with only you and your friends all being prepared for it!

Just a thought. Purely hypothetical. No one would do that. Surely.

arghwhat··on Temporary suspension of acceptance of mail to the United States
They don't really. They misreport the value for tax purposes, they own it through some other legal entity that is all negative, they own it through a company that negotiated tax rates to have some other business happen there, they own it in countries that don't to property taxes, etc. etc.

You can be damn sure they're paying at most a tiny fraction if not zero of any tax rate you'd be paying for that asset as they'd much rather pay very good lawyers, accountants, and most importantly, politicians to not have to pay tax.

You don't need progressive taxes. You just need everyone to pay it equally, without loopholes. Fewer less complicated tax rates are harder to work around.

arghwhat··on Temporary suspension of acceptance of mail to the United States
Oh they do, and you probably have several of them installed...
arghwhat··on Temporary suspension of acceptance of mail to the United States
The shops you're thinking of are less of a sweat shop than, say, Amazon in the US. Theres no good wages, but factory jobs don't really pay well anywhere.

They're mostly just a lot more efficient at scale, with a few plants managing close to the whole worlds supply of random shit. Almost all microwaves by all international brands are made by the same Chinese company, all prismatic LiFePo battery cells come out of one of two factories in China, and so forth. Economy of scale on turbo steroids.

Imagine having to compete with Ford by making cars in a garage. Now imagine Ford as the garage shop vs. these factories.

The sweat shops you're thinking of is stuff like clothes manufacturing in other third world countries than the usual suspect. That shit is nasty - breathing and handling acid with naked skin nasty.

arghwhat··on Temporary suspension of acceptance of mail to the United States
Most wealthy people are not billionaire wealthy.

Billionaire wealthy pretty much manage to avoid all taxation, progressive or not, so the comparison is moot. They just trample on everyone else.

arghwhat··on Temporary suspension of acceptance of mail to the United States
Tariffs like this is a market regulation that the people pays for.

It doesn't "benefit the wealthy" because it's not progressive, it benefits the wealthy that have investments in the tariffed industry by distorting the market to their advantage instead of having to be competitive on a level playing field.

The rest of the wealthy are equally annoyed by the tariffs as everyone else, possibly more so as they see their investments tank.

arghwhat··on 'Reading crisis' prompts Denmark to end 25% tax on books
I find this logic extremely flawed. If you save half, you save half. If someone saves a lot more than you as a result, it mains they paid a lot more before, and is still paying a lot more than you now. As fair as it gets. Those that complain are usually imagining a robin-hood esque system where the tax progresses towards 100% as income goes up.

Trying to heavily tax billionaires is one thing, but the issue with them is tax avoidance by virtue of these complicated systems, and a lot of the incremental taxes land on people just plain working their ass off, getting no sleep, high stress and high blood pressure as a result. If someone has more because they worked more, they're entitled to exactly that.

Incremental tax also means that if you have a good year and a bad year you pay way more tax than if you just had two average years. Not to mention that such complicated tax is what enables tax optimization whereby those higher up can end up paying less tax. It's stupid.

arghwhat··on 'Reading crisis' prompts Denmark to end 25% tax on books
Note that while you have 25% VAT on things, actual VAT accounting is fully dynamic and banking and accounting systems handle this just fine.

On the incoming side, bills can come from other countries with an entirely arbitrary VAT value, so there the VAT value is recorded from the bill.

On the VAT refund side, arbitrary values are used even for Danish VAT, as companies can get full, fixed fraction or even entirely variable fraction if VAT refunded. For example, a company dinner can only be partially VAT refunded as you had private benefit of food, and VAT refund of an asset like a van is the set by the ratio of intended company vs private usage.

On the outgoing side, most banking and accounting systems would be prepared for other EU countries.

There's definitely going to be something hardcoded somewhere (including over 9000 times in the tax systems themselves), but the point is that VAT is already a dynamic size.

arghwhat··on Walkie-Textie Wireless Communicator
For any radio system, receiving continuously also kills battery, just not as fast as transmitting does.

Low-power requires you to turn the receiver off for extended periods of time, but what you can do there is limited by how interactive the device needs to be, and how much the power the transmitter is willing to waste on retries/longer preambles.

For proper low-power (e.g., devices with ≥ 1 year battery life on small batteries), you're likely to need sleep periods of minutes to hours, or only waking up on physical interaction.

arghwhat··on A gigantic jet caught on camera: A spritacular moment for NASA astronaut
Not to mention flat-earthers, climate change deniers, 5G-causes-vaccines activists burning down 4G towers as they have no idea what 5G even is, etc., etc.

It's all too easy for the less skeptical to be misguided. :/

arghwhat··on Big Tech's A.I. Data Centers Are Driving Up Electricity Bills for Everyone
That's really doesn't apply to water vapor where the natural evaporation is many orders if magnitude grater than what were dealing with, and is a required part of the natural cycle to replenish water reservoirs, with the natural cycle being incredibly short and increased evaporation shortening that cycle.

The issue with greenhouse gasses is their potent impact combined with long lifetime. CO2 has to be slowly absorbed by e.g. plants, and adding more does not speed up the process unlike for water vapor.

You also cannot evaporate past 100% relative humidity, an amount of water saturation that naturally occurs. This is unlike gas emissions, where the sky is the limit.

> BTW we kind of boycott sunlight, it’s called sunscreen

Unless you're applying it to the sun, that's not boycotting sunlight, it's adapting to it. Not to mention that your skin requires sunlight, sunscreen just avoids overexposure.

arghwhat··on Big Tech's A.I. Data Centers Are Driving Up Electricity Bills for Everyone
And sunlight causes skin cancer, but we don't want to boycott neither water vapor nor sunlight lest we will have no precipitation and no... life.

Water evaporates constantly in from soil, plants, and water bodies - most notably from the ocean itself, which is how ~babby~ precipitation is formed. Evaporation from a datacenter is unlikely to make any notable impact through water vapor. What it could impact is highly localized potable water availability and humidity in places with e.g. low precipitation and limited reservoir capacity.

And, again, note that other cooling methods are also in use, and that other things also use evaporative cooling.

Heck, even when evaporative cooling is used, it doesn't necessarily mean that the water escapes into the atmosphere! Both heat pipes and refrigeration cycles are forms of evaporative cooling where the gas is allowed to condense, cool and evaporate again.

arghwhat··on Big Tech's A.I. Data Centers Are Driving Up Electricity Bills for Everyone
The water cycle refills drained reservoirs - the "carbon cycle" is not one we want to be part of.

Also note that there's other cooling solutions than evaporative cooling, such as closed loop water cooling with chillers, cooling with sea water, using heat pumps to redirect the heat to district heating loops (making use of the heat!), just building in colder places requiring less cooling, etc.

← PreviousPage 8 of 34Next →