zrav

> So you would think that it would make sense to break the signal down into stuff with frequency and time. Some kind of wavelet probably. Maybe something that very accurately models what a human hears.

Interestingly wavelet based compression went nowhere because although they have nice mathematical properties, when applied in a lossy compression scheme, they did not fit well with how humans perceive detail/quality, both in terms of psychoaccoustics and psychovisuals, i.e. PSNR vs subjective quality diverged more than with other systems. Not surprisingly none of the state of the art lossy compression algorithms use wavelets.

Does any of humanities current resource consumption seem sustainable for an unlimited time to you, even assuming no further population increase?

You say it's about control over the resource, which can be understood as the profit that can be derived from it, but I'd go further and say its just as much about leverage over the regions dependent on that oil or gas.

The US has been vehemently against projects that deliver fossil fuels to Europe by countries that are not its allies, like the Nordstream 2 pipeline or the Iran-Iraq-Syria pipeline. On the other side, Syria has blocked the US-backed Qatar-Turkey pipeline (interestingly drawing gas from the same South Pars/North Dome gas field), sparking US-Syria tensions around 2010.

> For Confluence, I think of Media Wiki.

Having used and looked inside both Media Wiki and its enterprise fork BlueSpice, I can't really recommend either if you need anything more serious than just "minimum viable wiki features".

A few years back we temporarily switched from using Confluence to Xwiki, which worked okay for a while, but then different small and medium issues kept adding up until we had to go back to Confluence and swallow the increased price (sixfold in a span of 3 or so years). This was with Xwiki 8 though, the current version 12 might have improved things, I should try it again really.

Mostly embedded, industrial systems, medical devices, set-top boxes and home routers. You connect to the serial port or gain access by other means and most of the time you are greeted by ancient versions. It gives me the chills.

Early versions of 2.6 and 2.4, release >= 15 years ago.

> It is hard to know why anybody thought it so urgent to go to 64-bit seconds counters for internal use in the kernel.

I routinely encounter systems running 15-20 year old kernels. Given that, NOW is the time to fix the 2038 problem.

I recently transferred my collection of MD recordings, but not using NetMD, getting a MD Deck with SPIDF output off ebay for cheap, and feeding the digital signal to a Focusrite Scarlett audio interface. This seems to ignore any copy protection bits. Resold deck for same price afterwards.

Indeed, seeing that there's also Radeons with 2048 bit bus for HBM.

Yes, since Sandy Bridge (2011) each new generation of Intel CPUs brought a single digit percent increase in single-thread performance only.

We're now at LXC/LXD version 3.x. Using LXD (LXC container orchestrator) is a breeze, with better and more polished UX than Docker, IMHO. Then again, LXC/LXD is more of an alternative to VMs than to Docker.

A similar problem I've seen a few times are the consultants that participate only in the first few months of a multi-year software project as architects, so never get to experience the long-term implications of their design and the mess they prescribe. Not only is this bad for the project, but also breaks the feedback loop for their own learning.

According to the x264 developers (can't find the reference anymore, sorry), wavelet based lossy compression is good for optimizing for PSNR, not for perceptual image/audio quality, being prone to loss of detail. x264 beats Dirac and modern image compression formats like HEIF (based on HEVC) easily beat JPEG2000. Modern codecs usually use transforms derived from the classic DCT and DST.

40 for 50 is not so bad. Some regional ISPs installing FTTB/FTTH offer 100mbps + voice for ~40€, which is a pretty good deal.

Due to heavy oversubscribing cable is an inferior product in Germany (on weekends actual speeds drop to a fraction of the advertised line speed), and the biggest cable provider by far (Kabel Deutschland/Vodafone) has serious user support and service quality issues, not to mention they only offer ipv6 dual-stack lite connectivity.

We do have the UN for that. Unfortunately, all permanent members (except possibly France) are guilty of state sponsored cybercrimes, so nothing comes of it, as with many other issues.

After PlaysForSure and Zune this is the third time MS shafts buyers of DRM'd content.

torrent.ru still is excellent for music, as long as you manage to register despite the cyrillic text.

Meanwhile, the F-35's fully mission capable rate remains miserable. https://www.pogo.org/investigation/2019/03/f-35-far-from-rea...

I don't think that "individualistic" is a strong enough word. European society, for instance, is also very individualistic, yet doesn't have anywhere near the levels of indoctrination against "socialism".

It seems Americans (generalizing here) have been let to believe that someones gain can only happen at someone else's loss, thinking there can't be a win-win situation.

I'd be quite surprised if it bested zstd.

The impact of audio hardware quality specifically on the noticeability/ABXability of compression artifacts is greatly exagerated. Going above mid-range speaker setups I have not found them to be more revealing in this specific regard. And of course, as you mentioned, even inexpensive but decent headphones beat very expensive speaker setups for critical listening.

I completely missed the Hydrogenaudio multiformat test in 2014. I wonder how much things progressed since when I participated in the early tests in 2006.

Compared to the many anti-scientific audiophile articles, this one was quite decent.

A bit that bothered me was the description of how the encoders work. It mentions the discarding of frequencies, which is not entirely correct. The only real discarding of frequencies happens with the lowpass filter that is typically applied as a pre-processing step, not the actual encoding. The encoding of the frequency components happens with coefficients, which are quantized (coarsely encoded), not zeroed. This introduces noise to the signal, which is weighted against the computed masking threshold, driving bit allocation to the different frequency bands.

As for the format recommendation for people ripping their own CDs, I'd say stick to a lossless format for storage at home. That way you're future-proof. For mobile use, make encodings from those files using whatever best lossy format is available at the time.

We likely have much less than 2 billion years before the earth becomes uninhabitable: https://en.wikipedia.org/wiki/Timeline_of_the_far_future

Statistics with representative sample size disagree. https://www.backblaze.com/blog/2018-hard-drive-failure-rates...

Since WW2 the US has been involved in actual or attempted regime change on average every two and a half years. Interestingly, or rather significantly, the last two attempts, Syria and Turkey (confirmation pending but likely), failed.

Until recently China ran a foreign policy of stability. But they have changed gears now, pursuing unilateral goals with increasing aggressiveness, also filling the power vacuum left by the waning US.

There is ongoing research about what to use as chamber wall material. The difficulty is that the material needs to be able to withstand high temperatures, minimize the impurities released when hit by a particles from the fusion plasma and ideally produce short lived and/or harmless isotopes when activated by the fusion radiation. Unfortunately, the metals most used and best known in engineering have the tendency to produce pretty nasty isotopes. The current best candidates are tungsten based alloys.

Under optimal conditions (rendering) enabling SMT on Intel gives just ~30% extra performance, ~40% on AMD (usually much less), so of course you can't compare the additional thread provided by SMT to a thread running on another, otherwise idle core.

Scheduling only threads of the same process on a SMT group at a time is the easiest portsmash mitigation while still retaining some of the SMT peformance benefit.

On CRT monitors the difference between 120 and 200Hz was very noticeable in fast arena shooters. On LCDs with their inherent motion blur there's probably less of a difference between 144 and 240Hz, but I'm not going to call it placebo due to my previous experience with high frequency CRTs.

Granted, the benchmark only covers training, but for a chip that spends significant die space on dedicated AI circuitry the performance gain over the previous generation is disappointing.