Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
sspiff
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
21 ms
·
241.
▲
by
sspiff
9y ago
I know this is about Power9, but these benchmarks really make me look forward to a time when companies will be ditching these first gen AMD EPYC processors on eBay for cheap. Good times are coming.
242.
▲
by
sspiff
9y ago
They used CSS to create the font (presumably as an art project?), but it is available in plain OTF/TTF format, so it can be used as a normal font and doesn't need to be a barrier to accessibility tools.
243.
▲
by
sspiff
9y ago
This seems like a classic case of annecdotal evidence being extrapolated into general truth. Two kids from an identical environment and sharing the same ancestors are hardly a significant sample.
244.
▲
by
sspiff
9y ago
Decelerating doesn't mean that it's no longer growing. It means that growth is declining over time, so growth today is less rapid than growth a year ago. That doesn't mean that growth will stop, though it might.
245.
▲
by
sspiff
9y ago
That's true, but the port work described in this post is older (predating the fork of LEDE and OpenWRT), and the latest work regarding these zsun devices are more recent (but they haven't been upstreamed into LEDE, and so haven&#x
246.
▲
by
sspiff
9y ago
I use one of these running LEDE. For the original porting work done on OpenWRT, check this link: https://wiki.hackerspace.pl/projects:zsun-wifi-card-reader For more recent work done to get up-to-date LEDE builds running on
247.
▲
by
sspiff
9y ago
> "Instead of paying hundreds of millions for a very big satellite in geostationary orbit, we now have new constellations of smaller — what we call nano satellites — in low earth orbit at much lower cost," he said. So, what abo
248.
▲
by
sspiff
9y ago
How does this work for two instructions in the pipeline at the same time that refer to the same cache line? If the second instruction executes the read phase before the first is retired/committed to cache, you would be hit by two memor
249.
▲
by
sspiff
9y ago
> For security reasons, the T2 is the iMac Pro hardware’s “root of trust,” and it validates the entire boot process when the power comes on. The T2 starts up, checks things out, loads its bootloader, verifies that it’s legitimate and cry
250.
▲
by
sspiff
9y ago
Sure, but then you still don't have V-Sync, which causes terrible and easily noticeable tearing on almost anything (at least in the case of Nvidia, it seems to be much better with AMD from my annecdotal testing). I doubt Apple would fi
251.
▲
by
sspiff
9y ago
Weird how they include an HD 630 on the die. This eats up a fair chunk of die area, and eats into the thermal budget. It also means that the displays will probably be wired to the HD 630, so the Radeon can power down to save power. Taking o
252.
▲
by
sspiff
9y ago
Sure, but it's still part of the outside world, so it's still opening a lot of extra interface surface to the platform.
253.
▲
by
sspiff
9y ago
I'm not sure I want this. I use WASM as a compilation target, but I rally enjoy the simplicity of the platform and how all connections from and to the Javascript world have to be explicit. Allowing more transparent access to the DOM wo
254.
▲
by
sspiff
9y ago
Actually, it is (1999): > This code is an original work by Paul Stoffregen, written in December 1999 The 2005 update just applied minor changes for a new board revision: > Status: This page is complete, code updated for rev 5 board,
255.
▲
by
sspiff
9y ago
Some public benchmark data: https://blog.cloudflare.com/arm-takes-wing/
256.
▲
by
sspiff
9y ago
This always manages to put a smile on my face. This is a pointless but awesome technical achievement, and in the spirit of the old definition of "hacker".
257.
▲
by
sspiff
9y ago
I've found that doing anything distracting can help. Commuting, listening to a podcast or audio book, playing a video game for 15-20 minutes, ... Whatever takes your mind away from matters at work.
258.
▲
by
sspiff
9y ago
rEFInd is the greatest little EFI bootloader!
259.
▲
by
sspiff
9y ago
I understand that there is experimentation and innovation inside and on top of the Linux kernel, but that still leads to some kind of local optimum, the best approach that fits within the design and implementation of the Linux kernel. I
260.
▲
by
sspiff
9y ago
For big metal rigs like supercomputers, there is almost no choice but Linux. Combined with Linux' dominance of the server market in general, there's a risk emerging to get stuck in a monoculture on the OS layer. Most companies, in
261.
▲
by
sspiff
9y ago
That's interesting, I'd not seen this comment before. So the good thing about System76 is that they can gather up the Linux users buying their products, and use the volume of those combined sales to give Linux users a voice with u
262.
▲
by
sspiff
9y ago
Sadly, no. I've tried to find out but couldn't find any clues. As far as I can tell, they're using their own custom built casing. It definitely looks much better built than the Clevo machines, but I've not yet have been
263.
▲
by
sspiff
9y ago
Sure, but implied in their Linux branding of the laptops is a promise of testing their configurations with Linux. So you are paying for a promise that the system will be compatible with Linux, I suppose. If these things interest you, you sh
264.
▲
by
sspiff
9y ago
No, but I can tell they're just rebranded Clevo builds. These machines have very decent specs and are easy to service and upgrade, but at the cost a of very low quality build with regards to the laptop case itself. There are plenty of
265.
▲
by
sspiff
9y ago
I used to work for Philips' TV branch, they used a similar setup. Instead of a camera they mounted a light sensor to the display and measured end-to-end delay for various color transitions (white-to-black, black-to-white, ...). All oth
266.
▲
by
sspiff
9y ago
I hope this leads to the revival the use of the truly awesome word "automaton" for autonomous robots!
267.
▲
by
sspiff
9y ago
Prior to ARMv8, all licensees could add additional specialized instructions or even entire instruction sets using the ARM coprocessor model. By changing a few registers, you could alter how instructions were interpreted. This was used by so
268.
▲
by
sspiff
9y ago
Both times the new CPU architecture was faster than the older one, granted. But still the emulated software would be slower on the new models with the emulator compared to bare metal execution on the old models. I assume Apple will make som
269.
▲
by
sspiff
9y ago
And both times they had a transitional period in which their toolchain was modified to produce "fat" binaries, including executables for both the old and new architectures. I find it stunning that people think Apple couldn't
270.
▲
by
sspiff
9y ago
What you need is a massive amount of parallel cores. Currently, the cheapest and most efficient way to achieve this is GPUs. It's true that some graphics-specific parts of a GPU are not needed for compute-only kernels (such as those fo
More ›