Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
berkut
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
35 ms
·
361.
▲
by
berkut
7y ago
OS X Lion was the first version to use an x64 kernel, and there were definitely x86 (32-bit) versions of the OS and apps. In particular, the quicktime runtime / codecs (until quicktime x) was only x86, and various VFX apps on OS X ha
362.
▲
by
berkut
7y ago
Most C++ build systems have support for moc generation - CMake even has CMAKE_AUTOMOC which can automatically do it for you with no further intervention with some caveats.
363.
▲
by
berkut
7y ago
So that it's grep-able?
364.
▲
by
berkut
7y ago
std::shared_ptr uses CAS atomics, which in heavy multithreaded code (multiple threads operating on the same pointers), can have surprising overhead in some situations.
365.
▲
by
berkut
7y ago
Yeah, HP Z4 or Z8 (or the Dell equivalents) seem to be used a lot in the film VFX industry, which at the high-end is mostly Linux...
366.
▲
by
berkut
7y ago
I'm in NZ, and Vultr has been awesome for me, although no block storage in Sydney... AWS is available (with Lightsail as well) in Sydney as well (albeit with reduced bandwidth compared to other regions for the pricing).
367.
▲
by
berkut
7y ago
For highly-tuned code, it's fairly trivial to find cases where data alignment and the amount of loop unrolling compilers do can have significant differences between microarchs. Whether it's to do with cache associativity, differen
368.
▲
by
berkut
7y ago
Yes, but if you're interested in performance, it's very possible that using the wrong scheduling of instructions for those particular optimisations (very particular scheduling ones) will result in extremely bad performance due to
369.
▲
by
berkut
7y ago
I totally buy that Intel should use the feature flags instead of checking the CPUid name for stuff like AVX, FMA, etc... However, ICC does support several advanced optimisations whereby it literally schedules instructions based on expert kn
370.
▲
by
berkut
7y ago
TBB has support for PowerPC and Sparc: Intel accepted patches. It's not that difficult to port Embree to other architectures, at least for the 4-wide stuff: you just need intrinsics wrappers, and a bit of emulation for the instructio
371.
▲
by
berkut
7y ago
Which would surely need to be GPL given it seems to be an auto-converted version of BusyBox, so I would have thought that makes it a derivative?
372.
▲
by
berkut
7y ago
Historically, yes. There was a reason previously the CAA and EASA were historically happy accept the FAA's certification for new aircraft, on the assumption the FAA were very good at its job. That no longer seems to be the case (given
373.
▲
by
berkut
7y ago
Erm, it's pretty obvious something was wrong at the FAA, in that it seems fairly clear they let Boeing dictate terms to them on multiple occasions, and effectively overrule them. https://www.seattletimes.com/business&#x
374.
▲
by
berkut
7y ago
HEIF support in Safari would be nice...
375.
▲
by
berkut
7y ago
Indeed. And rumour is they've stopped doing them now. On the VFX side of the industry we were often jealous of Pixar and Dreamworks being able to do shorts, in terms of having the scope and ability to test out new tech / pipelines
376.
▲
by
berkut
7y ago
Yep. Especially on a big screen, where jpegs out of the iPhone (and other brand phones) often look like water colour paintings due to the aggressive denoising that is done, even in bright sunlight where the ISO should be fairly low for the
377.
▲
by
berkut
7y ago
McDonald's ones have yellow "stuff", although I'm not sure if that can be construed as "cheese".
378.
▲
by
berkut
7y ago
No fires for me either in Firefox and Chrome on both MacOS and Linux (left the page several minutes).
379.
▲
by
berkut
7y ago
I've noticed on MacOS (on two MBPs) that Firefox 69 (with ublock origin as only plugin) is noticeably slower to render / lay out sites and maybe download content than Chrome (with ublock origin as only plugin) on same machines. Ba
380.
▲
by
berkut
7y ago
The EU have a corresponding case against the US and Boeing as well. They can also apply tariffs based on the fact they didn't do so last time they won a case against the US and Boeing: https://leehamnews.com/2019/0
381.
▲
by
berkut
7y ago
Out of interest, what would you use instead these days?
382.
▲
by
berkut
7y ago
If North Korea did have a means of delivering nukes with ICBMs, and this previous type of thing: https://twitter.com/realdonaldtrump/status/94835555702242099... had escalated more, I don't think it's
383.
▲
by
berkut
7y ago
I've been seeing pretty bad upload failures (probably around 30%) for uploading hundreds of 30-40 MB files per month to B2 from New Zealand since I started using B2 over a year ago. And I'm not convinced it's connectivity iss
384.
▲
by
berkut
7y ago
I think saying it's far from being on life support is really pushing it... Maya's USD support so far is effectively just Animal Logic's open source stuff bundled with Maya, Bifrost is indeed the one place they're actuall
385.
▲
by
berkut
7y ago
It's quite possibly a silent bug on the other platforms as well (i.e. relying on undefined or compiler-defined behaviour that just happens to work currently). A new compiler / OS update / graphics card driver in the future mi
386.
▲
by
berkut
7y ago
This has been my experience as well writing cross-platform software (some of it proprietary in the past). Because different compilers and different stdlibs do things in subtlety different ways, that may just work on one platform by chance
387.
▲
by
berkut
7y ago
If the soname is the same between the duplicate copies, and another copy is in $LD_LIBRARY_PATH it shouldn't matter about location.
388.
▲
by
berkut
7y ago
Only if they're the same version (which using the older built-in package system all packages are likely built against the same version)... If the containerised versions of apps all have different versions of dependencies (quite likely
389.
▲
by
berkut
7y ago
I'm pretty certain all 64-bit x86 (so amd64) systems support SSE 2, so FP calcs are done with these registers.
390.
▲
by
berkut
7y ago
It's a lower-end part, but regardless, POWER9 isn't really anything special in terms of instruction throughput / calculations (it's only got 4-wide SIMD for example), and its FP performance is not great. Where it reall
More ›