archi42

71 ms·

661.
by archi42 7y ago
Actually, PCB milling[1] is a thing. I suppose a MPCNC can be used for that. But it still has the huge drawback that you're limited to one layer if you're lazy (like me) or two if you fell like spending the time to add manual vias
662.
by archi42 7y ago
I suppose the average HN commentator is used to indexing arrays using the postfix []-operator, thus starting at 0, while the tools mentioned by GP are aimed at everyone - and most "normal" people start counting at 1. For reference
663.
by archi42 7y ago
Chill please. I used the exact number from the paper as quoted somewhere else in this discussion. Which seems adequate when saying that the authors final number (sixhundredsomething) is most likely correct even if the intermediate value &qu
664.
by archi42 7y ago
> Crypto doesn't discriminate on whether you have a warrant or not. Actually, backdoors could be implemented to require a cryptographically signed warrant. Think e.g. lawful intercept on mobile network routers, US (or any other nati
665.
by archi42 7y ago
Interesting list. Besides Palantir, I know these: GitLab, Databricks, MemSQL and mongoDB. I don't think "they" are using these to exfiltrate data "Crypto AG" style - I'd be surprised if "big data"&#x2
666.
by archi42 7y ago
The total 620 is actually correct again: 0.0072 times 0.0003 times 311591917 is 673.04. I suppose the 53 person discrepancy is due to rounding towards 0.03% but using a more accurate number for the actual computation (e.g. 0.0277%).
667.
by archi42 7y ago
At least our Thai Airways A380 only had a singular, front-facing camera at the top of the vertical stabilizer. No other view was possible. But this may depend on the plane's configuration.
668.
by archi42 7y ago
I doubt it: For VR, latency and jitter are much more important than for other games. My GPU (old R9 290) has ~15ms delay in the encoder when using parsec for streaming. Add a few ms in the decoder (1ms), plus whatever the network/Inter
669.
by archi42 7y ago
Brute force attempts in servers are pretty old news. If you're running a public facing ssh server, it will be found and it will be tried. My solution is to disable passwords and put the server on a non-standard port (yeah, users could
670.
by archi42 7y ago
Depends a lot on the compiler and target arch. You'll miss out on a lot of stack accesses, or add too many. You don't get around looking at the final executable if you want good results. And for more complex targets, in the end yo
671.
by archi42 7y ago
> Hell of a lot has to be done for Habana, too. Yeah "barely more complex than NUMA" was a slight exaggeration and only meant in comparison to the other ;-) Judging by the architecurial differences layed out in the link, Habana
672.
by archi42 7y ago
Uhm, I'd suppose the service runs on one or more dedicated nodes - so there should be no competition for RAM (or if a node runs multiple services, the I'd expect a fixed memory amount to be available). In such an environment, each
673.
by archi42 7y ago
Not surprised: 100GBe interconnects vs proprietary interconnect. Complex architecture for which a hell of a lot work has to be done in the compiler (anyone remember Itanium?) vs something that's barely more complex than NUMA. Lower TDP
674.
by archi42 7y ago
We have {}[], but for me they're a pain to type since it breaks my typing pace. When I was gifted an old laptop with US layout I saw the light and changed my desktop keyboard to US as well. When I started working my employer asked whet
675.
by archi42 7y ago
Oh, I did expect a higher geekbench score TBH: My dual Xeon 2687W (v1) with 128GB DDR3 (2*4 channel) gives me 830 SC and 6000/10000 MC (1/2 CPUs). And that's with thermal limitation due 1U cooling solution. OS is Arch Linux o
676.
by archi42 7y ago
Ah, I think I have a hugely different approach to data processing: For my work I often have a very good idea what the output should look like, and what transformations are required on the input to get there. E.g. when processing log files t
677.
by archi42 7y ago
As a computer scientist/software engineer, please allow me the question: Why would I prefer a notebook over e.g. equivalent python script(s) in a git? I first saw jupyter notebooks when my sister (physicist, non-programmer) used it for
678.
by archi42 7y ago
At home: I'm running a RAID1 btrfs on my 12 disk cold storage (rackmount, SAS backplane, JBOD SAS controller). It has two new 4TB 24/7-rated SATA disks I got for that NAS, the rest is mostly salvaged from work (old drives, 500GB t
679.
by archi42 7y ago
Hello fellow non native speaker. You don't mean "insensitive[s]", but " incentive[s]". On topic: My university offers "Ethics for Nerds [=CompSci]" lecture. The lecturers have degrees in both CS and Philos
680.
by archi42 7y ago
> competing talents were denied their rightful places Absolutely. What are the odds he (and/or others in his team?) did this just for this one competition? It's possible, but unlikely he/they invented this (or other) code
681.
by archi42 7y ago
Similar for me: If I want to look up a doctors phone number or the location/opening hours of a shop I still fall back to Google.
682.
by archi42 7y ago
Correct :-)
683.
by archi42 7y ago
I second this: My alma mater (in Germany) has a compiler group, and, as far as I'm aware, at least some of its alumni went to the companies mentioned by parent.
684.
by archi42 7y ago
Also: Aerospace, I think.
685.
by archi42 7y ago
Other dev: "I wish I could just tell an AI to do $implementation for me", me: "You know what our boss just did?". ;-)
686.
by archi42 7y ago
This! Our company OpenSUSE Tumbleweeds were still vulnerable on Friday. On Arch, the package was already available for two days then. The thing is, the distro maintainers need to decide whether they want a fast moving "latest and great
687.
by archi42 7y ago
The guy who is called after it's too late (he's a friend, I didn't call ;-)). So it's bad advice to use a work factor and instead you recommend using a work factor? Uh yeah, still thanks for pointing out how that's
688.
by archi42 7y ago
I was told it's best practice these days to hash several times: If you hash the $salt.$pw string several hundred times (e.g. 512 or 65536) that only takes a few ms on you server (each time), but for a cracker the constant factor is rat
689.
by archi42 7y ago
Two ideas to lessen the trade-off: 1. Use a browser extension that throws a warning on all unicode domains (maybe even with unicode highlighting). Drawback: Needs to be done per-device. 2. Let your pihole MitM all https traffic with a certi
690.
by archi42 7y ago
Exactly this. If I woke up with a million-$ idea tomorrow, chances are good I could not just get up and just pursue it. Why not let someone else take the chance? Add to this that, as time progresses, more and more ideas that are doable by a

More ›