HNHacker News
TopNewBestAskShowJobs

delta_p_delta_x

5,063 karma · joined July 15, 2021

submissionscomments
delta_p_delta_x··on Astronomers may have found the first exomoon
There's really a lot to unpack here. (I rewrote this response thrice...) To spoil the answers straight away, and perhaps to address some misconceptions upfront:

> If a brown dwarf is right on the edge of stardom would it start protium fusion in only the part of itself that is the right pressure

The nuclear cross-section of protium fusion in astronomical bodies is determined by temperature, pressure, and density. These variables are in turn dictated by the total mass of the object in question. Brown dwarfs never have sufficient mass for protium fusion, so they never undergo any protium fusion whatsoever. This is a hard-and-fast boundary for stardom.

> or would that ignition precipitate a pressure wave through the entire body forcing fusion to begin everywhere

The three variables above are not uniformly distributed throughout the body; there is a maximum at the centre, and fusion only happens here. Even in the Sun, about 99% of the fusion happens within about a quarter of the radius from the centre. The temperature drops rapidly thereafter, reducing the nuclear cross-section of the proton-proton chain to essentially zero. At the photosphere (surface), the temperature is ~5777 K, which is a decidedly Earthly temperature (lightning bolts are ~ 30000 K). So no, fusion does not happen everywhere.

Now, we need to discuss star formation and why brown dwarfs have never experienced protium fusion at any time in their lives. Star formation is still a very active area of research, debate, and fitting models to empirical study, and this is especially true for the detailed interior and structure of protostars and pre-main-sequence stars; hence, this is going to be quite surface-level (pun not intended).

Collapsing molecular clouds form stars. The total mass of a given cloud (or a particular region of it) sets an upper bound on the resultant object, because the total mass dictates the gravitational potential energy and hence the terminal velocity of the matter, and hence the rate of matter infall at the centre before the cloud dissipates. If the mass is low enough, the central object will become a brown dwarf, or even a large gas giant and a 'rogue planet'.

As the gas cloud collapses, the central region increases in density, temperature, and pressure, but no fusion occurs yet. Conservation of angular momentum forms a circumstellar disc, and material continues to fall onto the central region. As long as this infall continues, the central region is called a protostar. At some point the mass of protostar crosses the boundary needed for deuterium fusion; if the infall stops here, the result is a brown dwarf. If this infall continues, the mass increases beyond the boundary (~80 Jupiter masses) needed for protium fusion, and protium fusion can begin. When the infall stops and the circumstellar disc largely dissipates, the result is a pre-main-sequence star.

Note that both these very young pre-stellar objects are not yet at hydrostatic equilibrium, and are still comparatively rarefied (or 'puffy') compared to main-sequence stars; they are still collapsing, and the temperature, density, and pressure at their cores continues to increase. Only when this equilibrium is achieved and gravitational collapse is halted do stars begin life on the main sequence.

Now, it should be evident why brown dwarfs never experience protium fusion: at no point in their lives have they ever had any region in their interior hot, dense, or hyperbaric enough to have a high enough nuclear cross-section for protium fusion. At their formations, they were simply not massive enough; they continue to collapse, which admittedly provides a considerable power output—surface temperatures are ~1000 K. The largest brown dwarfs experience deuterium/tritium/lithium fusion into helium, but this also stops over time.

(Side note: in my opinion the word 'brown dwarf' is a bit of a misnomer, because look at how bright molten iron (~ 1500 K) is even in broad daylight[1]; now imagine an object ten to twenty times the radius of Earth, emitting this much heat from every square millimetre. If you approached a 'new' brown dwarf it would cast a lot of light.)

[1]: https://commons.wikimedia.org/wiki/File:Scunthorpe_Molten_St...

delta_p_delta_x··on Astronomers may have found the first exomoon
Even within stars the layman labels are a bit iffy; it is better to use the stellar classification. 'Dwarf' is badly overused, as is 'giant'.
delta_p_delta_x··on Software rendering in 500 lines of bare C++
I'm clearly super old-school when it comes to rasterisation, and a lot of this has flown above my head. I have a ton of questions; I hope you can answer them.

> (1) discard (fast but not a great user experience)

What are we discarding here, and why is it fast but not a great user experience?

> (2) primitive synthesis

I assume this is retriangulating clipped triangles that are now no longer triangles?

> reverse transformed clipping rectangle

Which spaces does this reverse transformation map from and to? I assume the clipping rectangle here is the triangle's AABB in raster space (or as you say, barycentric space).

> integer fixed point, but that is historical — modern fixed point can be handled with careful control of the fp unit in the mantissa

So we are no longer doing 16.16 fixed point, but tweaking the FP representation itself?

> The major issue is regenerating the Z and the 1/Z values for the new vertices

Why is this a major issue?

> deferred attribute synthesis rasterizer

I assume this means attributes are perspective-correct interpolated in raster space.

delta_p_delta_x··on Astronomers may have found the first exomoon
Yes; I was under the impression that from context it was clear what I was talking about. In astrophysics, to avoid precisely this confusion, the convention is to use the adjective 'massive' when comparing, well, masses, and 'largest'/'smallest' for the dimension of length and its higher powers (radius/diameter/area/volume). For instance:

> the Sun is 3.33e5 as massive as Earth.

delta_p_delta_x··on Astronomers may have found the first exomoon
Worth noting that the artist's impression is... not accurate. Both CD-35 2722 b (the brown dwarf orbiting the primary star) and CD-35 2722 b I (the exomoon orbiting the secondary) should be much closer in size. It is estimated that Jupiter is essentially the largest any gas giant can get; adding more mass will simply increase density and interior temperature until deuterium and lithium fusion and brown dwarfdom, and then at around 80 Jupiter masses, protium fusion and stardom.

Look at Barnard's Star[1], which is actually a fusing red dwarf star: it is not much bigger than Jupiter.

[1]: https://en.wikipedia.org/wiki/Barnard%27s_Star

delta_p_delta_x··on Astronomers may have found the first exomoon
The definition of a brown dwarf is that there is enough heat and compression for deuterium and/or lithium fusion, but not protium fusion.

The exact model limit is unclear, but the physics modelling is relatively straightforward.

delta_p_delta_x··on LG to ban residential proxies from smart TV apps
The situation is very different on Macs. Apple control almost the entire hardware stack of the machines that macOS runs on. They make it an absolute pain in the neck to run any application not blessed by attestation. This attestation server is also frequently offline. They've also made it very difficult to write things like filesystem drivers, having completely pulled that functionality from more recent versions of macOS. In fact, if I recall correctly one cannot ship a third-party kext on more recent Macs, especially ARM ones. As such, hackintoshing has basically vanished as a hobby.

At least on Windows, the most you get is a scare screen saying 'this app is from an unidentified developer'. I know what I'd rather have.

delta_p_delta_x··on LG to ban residential proxies from smart TV apps
> There’s still a world of difference between a MacBook touchpad and a windows one

Oddly I've found otherwise (Dell Precision notebook from 2021, and Surface Laptop). It might be macOS's animations and smoothing interfering here, but I've found that my Windows touchpads are much more responsive, especially when swiping between desktops.

> The peripherals still need the addons though

Ugh, this is annoying. Hardware vendors need to be banned from writing lousy, inefficient, unsafe software.

delta_p_delta_x··on LG to ban residential proxies from smart TV apps
> It's Windows Update that's installing LG crapware upon seeing relevant hardware IDs

This is also a misrepresentation. Microsoft has provided LG with a certificate to sign its driver packages with, and allows LG to upload packages to Windows Update, which includes a feature to install sidecar applications. Now, the spirit of this feature is that any application is meant to provide genuine configuration functionality or some graphical front end to the driver's configuration knobs.

LG then abused that feature to provide adware. Now, there are millions of hardware vendors. One can't expect that every driver package submission from every vendor is thoroughly vetted every time, this is a matter of trust and respect.

Sometimes one actually wants the app that comes with the device, like AMD's and NVIDIA's configuration trays.

Even so, I fully expect that after this debacle MS will disable this stuff. There is precedent for this. Synaptics/ELAN/Alps touchpad tray applications largely disappeared after MS implemented precision touchpads somewhere in 2015, and suddenly Windows touchpads became as precise as those on Macs. Likewise with RGB peripherals. Even GPU driver applications have increasingly become less useful as the most important features—power control, adaptive sync, HDR, etc have moved into native Windows settings.

delta_p_delta_x··on LG to ban residential proxies from smart TV apps
> Another win for the Linux security model

I swear, OSs have become sports teams.

Linux's 'security model’ has plenty of holes. The very fact the kernel and much of its user-mode is written in C almost guarantees that its security model is worthless. The Linux ecosystem operates on trust and respect that can and has been easily abused by bad actors to provide supply-chain pwnage.

There have been so many zero-click local privilege escalation CVEs I've lost track.

Arch Linux AUR malware: https://lists.archlinux.org/archives/list/aur-general@lists....

delta_p_delta_x··on How proprietary formats have become Microsoft’s main tool for lock-in
> There are still differences between PDFs in different viewers

Most obvious when viewing PDFs in Adobe Reader versus in Chrome's built-in viewer. The former has subpixel anti-aliasing. The latter doesn't.

delta_p_delta_x··on LG monitors silently install software through Windows Update without consent
I should've specified USB audio interfaces, which do have gain and stuff like loopback. I'm very glad the Realtek tray program has fallen by the wayside, what an annoying bit of crap.
delta_p_delta_x··on LG monitors silently install software through Windows Update without consent
Windows does plenty of 'advanced work'. Almost all video games are written primarily on and for Windows.

> In my experience, most of the advanced work is done on Linux nowadays

That's because your experience probably hasn't ever included work on Windows internals. Take it from someone who has—the complexity and 'advancedness' of the stuff running on Windows is at least equal to that of Linux or any other OS. The fact that Windows can so thoroughly abstract the computer away from the user is in itself a massive feat that few other OSs have really managed.

> Just look at HN front page - how many posts are Windows-only

The overwhelming majority of posts on the HN front page are now LLM slop or web development. I seriously dislike this insinuation that work done on Windows is, as the grandparent claims, 'unserious' or less advanced.

delta_p_delta_x··on LG monitors silently install software through Windows Update without consent
You only need (5) (see the AWCCWINUI3APP thing there?). There is also a group policy equivalent to this:

  Computer Configuration\Administrative Templates\System\Device Installation\Device Installation Restrictions
  Prevent installation of devices that match any of these device IDs
  Set to Enabled
  Enable Also apply to matching devices that are already installed
  
Add the following two IDs:

  MONITOR\DELA246
  SWC\VID001&PID0001&AWCCWINUI3APP
IMO this is especially heinous as Dell have registered the AWCC.exe software component as a hardware 'device' within the device tree that needs its own 'driver'. Methinks Microsoft need to tighten the noose on these annoying OEMs.
delta_p_delta_x··on LG monitors silently install software through Windows Update without consent
> All the major tools for advanced work are Linux-based

No, they aren't. Linux hasn't yet got anything remotely close to PDB symbol servers and WinDbg's record-replay debugging. perf is... an attempt.

Source: worked on Windows and Linux drivers and user-mode applications. Windows tooling blows the competition out of the water in actually advanced developer experience. Vim is cool to the ricing hackerman types but not people who actually earn salaries. Windows doesn't need Docker because it has a stable user-mode ABI.

delta_p_delta_x··on LG monitors silently install software through Windows Update without consent
*.cpl, *.msc, etc are Windows sysadmins' (and developers') bread and butter; Microsoft will never get rid of them. I am betting (maybe hoping...?) that Windows 12 will undo the changes in Windows 8, 10, and 11 as bad experiments and return to what Windows has done best, which is discoverable GUI configuration. Let's see.
delta_p_delta_x··on LG monitors silently install software through Windows Update without consent
Use autounattend.xml[1] to pre-configure an image before installation. This is what corporate sysadmins have done in the past three decades to administer Windows NT fleets. Use PowerShell and various admin modules to configure an online installation.

Then, to get a better version of Windows, use MAS[2].

[1]: https://schneegans.de/windows/unattend-generator/

[2]: https://massgrave.dev/

delta_p_delta_x··on LG monitors silently install software through Windows Update without consent
> Why do we even have that lever

For plug-and-play devices with multiple configuration knobs. It is nice to be able to click through a printer wizard to configure how one wants to print their documents. Likewise with an audio interface: loopback settings, codec, sampling rate, gain and volume of channels, etc. Or consider a USB CNC mill; configuring things like milling revolution rate, setting which bit is installed, what lubricant is used, etc. Or consider the Nvidia/AMD control panels for their GPUs; things like colour depth and space, resolution, scaling, anti-aliasing, vertical synch, power settings, etc.

Some of these settings are device- and even manufacturer-specific; one might argue these are more than a driver or the platform can or should provide. That being said, this stuff should go into a user-mode driver...

That LG have exploited this functionality to install adware is on them.

delta_p_delta_x··on LG monitors silently install software through Windows Update without consent
> Linux won't ever do this BS.

Arch Linux's AUR was recently hit by an actual malware supply-chain attack[1], which I would claim is arguably worse than adware. NPM is regularly in the news for supply-chain attacks. And then there was the XZ utils debacle in 2024. I concede that Microsoft is in part responsible for facilitating something like this, but just because something is free and open-source or based on Linux doesn't make it a universal panacea for malware or supply-chain pwnage.

[1]: https://lists.archlinux.org/archives/list/aur-general@lists....

delta_p_delta_x··on LG monitors silently install software through Windows Update without consent
Edited with another method.
delta_p_delta_x··on LG monitors silently install software through Windows Update without consent
Workaround:

  gpedit.msc
  Computer Configuration > Administrative Templates > System > Device Installation
  Prevent automatic download of applications associated with device metadata
  Set to enabled
  OK
On home editions sans gpedit.msc:

  sysdm.cpl
  Hardware tab
  Click Device Installation Settings
  Under 'Do you want to automatically download manufacturers' apps for your devices?', select 'No'
  Save Changes
delta_p_delta_x··on Vāgdhenu: A Sanskrit Chanting TTS System
> In some ways, the schwa sound in English the worst as there is no symbol which is committed even partially towards it.

> English imports spellings from other cultures and adds its own layer of pronunciations. Other languages probably don't do it as often.

English is almost unique in how inconsistent its orthography is especially with respect to vowel sounds, and loanwords are only the latest manifestation of this problem. Even consonants aren't spared. Consider the digraph 'th'. This can be a voiced or voiceless dental fricatives. English used to have letters for each: ð and þ. 'ough' has nine pronunciations—plough, though, through, thorough, cough, rough, bought, lough, hiccough.

English has experienced contact with such a wide variety of unrelated foreign languages to an extent few others ever have.

delta_p_delta_x··on Vāgdhenu: A Sanskrit Chanting TTS System
I wouldn't say 'neither', because by your own example, 'a' in English can represent ⟨ə⟩. That every other vowel letter also can is English's low level of phoneticism with respect to the script it uses. Many other languages using the Latin script (German, Italian, Finnish) don't have this problem, what you see is what get.

Oddly enough amongst the Indic languages, Tamil has precisely the same problem: க, ப, ட may be voiced or voiceless plosives depending on context. Sanskrit loanwords in Tamil are typically pronounced very differently compared to their original in Sanskrit, or even in Telugu.

Personally when transliterating any Indic language I tend to use ISO-15919, and in that scheme, it is Rāma, and 'a' represents the schwa. Or Rāmaḥ, Rāmam, Rāmē, or Rāmasya, whichever grammatical attachment is appropriate.

delta_p_delta_x··on Vāgdhenu: A Sanskrit Chanting TTS System
It's pretty funny reading comments on social media by north Indians whenever something on Sanskrit is posted. 'It's Ram, not Rama, it's yog, not yoga'... And they have no idea what schwa deletion is.
delta_p_delta_x··on Almost Always Unsigned
I have a better solution: address the root cause of unsafe semantics by not using raw indexed for-loops, unless one absolutely needs an index, in which case one should generate it with std::views::enumerate. To reverse it, use std::views::enumerate | std::views::reverse. Ditto for languages with similar semantics.

I almost never write a raw for-i loop any more, especially since 99% of the time I want to enumerate through the entire array or vector, and I can just use a ranged for-loop to do that. It allows me to redesign my code around the data, express things at a higher level, and my code looks far more SIMDable and reminiscent of array programming languages. And yet it is safer, I will never see under/overflow or any of these old-hat problems.

If you are using C, then too bad, you're stuck with a language that doesn't allow the programmer to more meaningfully and more clearly express intent at a higher level of abstraction without paying additional runtime costs.

This stuff compiles to broadly the same assembly.

delta_p_delta_x··on TypeScript 7
This is called automatic type inference, and it is a big feature of functional programming languages, many of which are very strongly typed. Also, for the record, C++ gained type inference about a decade and a half ago.

In C++ one can declare a completely typeless lambda:

  auto callsAdd = [](auto x, auto y) { return x + y; }
And the programmer need never specify what x and y are, as long as there exists a reachable declaration of operator+ that has two arguments that accepts whatever x and y resolve to, at instantiation time (which is compile time).
delta_p_delta_x··on TypeScript 7
That is what the parent author means. Static vs dynamic typing is along the dimension of when the type is checked, and strong vs weak typing is a matter of how strongly bound names adhere to types. JS, for instance, is super weak here, you can assign a numerical value to something and in the next line re-assign it to a string, an array, or even a function object.
delta_p_delta_x··on Nintendo announces new product revisions in Europe with replaceable batteries
Brussels effect, please do your magic; thanks.
delta_p_delta_x··on Microsoft GDID telemetry includes full browsing and gaming history
Some context conveniently left out of here, the Reddit post, and the Twitter cross-post:

> Scattered Spider has been involved in hacks resulting in more than $100m (£75m) in ransom payments, according to the DoJ.

> The charges come after two young men pleaded guilty to offences connected to the £39m 2024 Transport for London cyber-attack, which the UK National Crime Agency (NCA) believed was conducted by the group.

> Investigators from the NCA said they believed the 2024 "network intrusion" - in which around 10 million people's data was stolen - was carried out by Scattered Spider.

https://www.bbc.co.uk/news/articles/cwy0we4yw1lo

delta_p_delta_x··on This blog is written in en-GB
Not to mention the fact that English basically everywhere else but the US is essentially en-GB with a few choice changes and anachronisms. Consider en-IN, en-IE, en-SG, en-MY, en-AU, en-NZ, etc.
← PreviousPage 4 of 34Next →