Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
kllrnohj
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
51 ms
·
901.
▲
by
kllrnohj
3y ago
Personally I think comptime is a questionably design and is something both Rust & C++ did much better. In C++ branches that don't have to "compile" are explicitly marked as such with `if constexpr`. In Rust you've go
902.
▲
by
kllrnohj
3y ago
The other workaround is to reboot at least once every 3 years, which surely most users are doing anyway to pick up on security patches & similar. Exceptions definitely exist, but the workarounds are both pretty straightforward and you c
903.
▲
by
kllrnohj
3y ago
> You did not take into account my point about code operating on primitives definitely not having that kind of overhead. Because it's not really interesting to debate. Nobody writes Java code like that, and even when they do there&#
904.
▲
by
kllrnohj
3y ago
The most recent 10 years and that's exactly the point - we're plateauing on hardware advancements, so software inefficiencies are more of an issue now than ever before. 2x is now a huge change in performance. You're not get
905.
▲
by
kllrnohj
3y ago
> having almost native speed But it doesn't unless your "almost" is very generous. Java is pretty consistently 2-10x slower than the major performance-focused AOT offerings (C, C++, Rust) Now maybe you call 2x "almo
906.
▲
by
kllrnohj
3y ago
AOTs consistently do much more inlining & unrolling than JITs do. JITs often have preset heuristics for figuring out where to split the code that's practical to implement rather than being the most performant possible. After all, t
907.
▲
by
kllrnohj
3y ago
Nearly every consumer Intel CPU sold for the last 10 years has had unified memory between the CPU & GPU. Similarly nearly every Qualcomm, ARM, etc... SoC for the last 10+ years has had unified memory. This isn't a rare or new archi
908.
▲
by
kllrnohj
3y ago
Most zen4 consumer reviews didn't even have any Java benchmarks at all much less any that would have significantly changed the narrative around those CPUs. Epyc vs. Xeon is what this is about as the datacenter is the last holdout for O
909.
▲
by
kllrnohj
3y ago
A violent uncontrolled stop like that would be hard to get your bearings for the next burst of acceleration. Also a continuous acceleration along a curve is likely going to be faster regardless.
910.
▲
by
kllrnohj
3y ago
Of course it can. Just extract the qsort implementation to a static library pulled in by the kernel and by libc.
911.
▲
by
kllrnohj
3y ago
Well hopefully at some point we'll stop writing C at all
912.
▲
Android's UltraHDR Image Format
(developer.android.com)
1 points
by
kllrnohj
3y ago
|
0 comments
913.
▲
by
kllrnohj
3y ago
Uh, Pixel C? Google did the magnetically attached keyboard that turns it most of the way into a laptop long before Apple did. And the keyboard part worked great.
914.
▲
by
kllrnohj
3y ago
Seems extreme since app updates regularly still happen?
915.
▲
by
kllrnohj
4y ago
Do a search for Apple & Khronos and you'll find a lot more examples of spats between them over the last decade than anything else. Including an ongoing legal dispute. You also won't find any hint of involvement from Apple in a
916.
▲
by
kllrnohj
4y ago
AMD's Mantle is the direct precursor to Vulkan. AMD donated it to Khronos.
917.
▲
by
kllrnohj
4y ago
How is that evidence that Apple ever tried to engage with Khronos on a Metal-like API, though? Seeing as Apple had already stopped updating OpenGL versions about 4 years before the release of Metal, it seems more likely that Apple never pla
918.
▲
by
kllrnohj
4y ago
I think your doing your own revionist history here. The rumors are all that Apple is the one that blocked Khronos' initial attempts with the "Long Peaks" proposal that was supposed to become OpenGL 3. And while Metal was rele
919.
▲
by
kllrnohj
4y ago
Not quite. The main problem with Vulkan is, as the blog post goes into, one of usability . Vulkan isn't designed for end developers to use, it's designed for middleware vendors to use. Vulkan is already being pushed by many major
920.
▲
by
kllrnohj
4y ago
A battery bank at your house and that same battery bank at the power station are effectively equivalent. There's no benefit to distributing batteries here since transmission costs are relatively small. So while yes water storage system
921.
▲
by
kllrnohj
4y ago
When your heat pump is using emergency heat yes it'd be "better" to burn natural gas to do so directly that to burn gas to make electricity to do resistive heating. However you need to look more at the average scenario to d
922.
▲
by
kllrnohj
4y ago
If those batteries are too old/weak to effectively power a car how useful are they going to be at powering a household stove? Especially when why would you need such a thing anyway when you have 220v+ just itching to get the job done d
923.
▲
by
kllrnohj
4y ago
How do you effectively mandate that those building owners are maintaining their gas lines and ensuring they aren't leaking? This is where it stops being "personal responsibility" Also you personally can still have gas anyway,
924.
▲
by
kllrnohj
4y ago
GFCI was already made mandatory in all kitchens nationally. The 2023 code updates expanded and strengthened the requirements here
925.
▲
by
kllrnohj
4y ago
> Plus electric goes out often. Gas doesn’t. That's because when gas lines fail, which they obviously do, they leak instead. And leaking natural gas into the air of densely populated areas is of course a bad idea. And since those
926.
▲
by
kllrnohj
4y ago
The dangers of lead were known before leaded gas was even invented much less popularized. But immediate profits and convenience are much stronger human motivators than long term health consequences ever are. See also PFOA and similar. Healt
927.
▲
by
kllrnohj
4y ago
https://godbolt.org/z/4WxGj7c65 Sure looks to me like GCC figures it out. Clang doesn't seem to, though.
928.
▲
by
kllrnohj
4y ago
Firefox has support it's just behind a flag as the spec is still a work in progress draft. Which is what chrome is supposed to do, too, but these days chrome doesn't really care at all about standards and compatibility, instead
929.
▲
by
kllrnohj
4y ago
More of Twitter is funded by Saudi Arabia than NPR is by the federal government, should Twitter also have a state-affiliated label? The stated purpose of the label was to identify state-operated news (ie, literal government propaganda outle
930.
▲
by
kllrnohj
4y ago
WASM was designed before spectre happened, so no it's not the exception. In-process sandboxing cannot be securely designed anymore. You can endlessly chase security exploits with increasingly expensive and convoluted workarounds, but t
More ›