ignaloidas

@ignaloidas@not.acu.lt

I do in fact exist

I'm an information sponge, so if you have some question that you think I might have an answer to, feel free to ask! Even if I won't have it off my head, I know how to look up things fast.

This profile is from a federated server and may be incomplete. Browse more on the original instance.

jon, to random
@jon@gruene.social avatar

Welcome to today's thread - South East Europe Day 02 31 May 2024 - Tallinn - Lelle - Pärnu - Häädemeeste

Crossing these borders:
None – only crossing Estonia, getting close to Latvia

These borders on the borders map:
https://umap.openstreetmap.fr/en/map/crossborderrail-all-the-borders_935041#8/58.744/24.192

Today's routes on the routes map:
https://umap.openstreetmap.fr/en/map/crossborderrail-all-the-borders_935041#8/58.744/24.192

ignaloidas,

@pony @DiegoBeghin @jon I will note that this one is planned to just have platforms on the main line, without any kind of supporting tracks

and probably a shack or two for cover on the platforms (not that familiar with detailed Estonian station plans, but I don't see how that location would justify anything more)

jon, to random
@jon@gruene.social avatar

They have a handy map of where I’m going tomorrow 🙂

ignaloidas,

@jon Depending on how you count, Pasraučiai in Lithuania is smaller, 28 inhabitants in the village the station is named after, but there's also a village with a bit more than 100 inhabitants just as close.

ignaloidas,

@jon I think there should be some construction happening at the site already? not sure though

it is quite in the middle of nowhere though, like, unfortunately, many stops in Kaunas-LT/LV border section

ignaloidas,

@jon tbh it feels like most of the RB route has been made with minimizing interactions with people on smaller towns in mind

which I guess is one way for making sure you don't get boycotted to hell due to some people needing to sell their houses, but it also doesn't bring that much to people next to the path of the project (because there aren't any).

ignaloidas,

@panda @jon There are some plans for Kaunas-Berlin connection (without changing at Mockava), but in like 2028...

ignaloidas,

@panda @jon I think Šeštokai was only used quite a long time ago, Mockava has been used since 2021 AFAIK.

ignaloidas,

@jon @panda I'm pretty sure it's because Šeštokai only has a 200mm high platforms, and none of them are between both gauges, while Mockava has a 550mm platform with both gauges at the sides, which is more suitable for changing trains.

There is some suitable platforms further down, like in say Marijampolė, but because the section from Šeštokai onward isn't currently signaled (though that's being changed AFAIK), it's not that viable to changeover anywhere else.

ignaloidas,

@jon @panda Could be both to be honest. AFAIK it's planned that it'll be a PKP train to Kaunas once the signaling will be done, so I guess the schedule will be changing somewhat.

whitequark, to random
@whitequark@mastodon.social avatar

flipper zero: has an assembly/production issue requiring you to disassemble it and reseat the battery connector https://www.reddit.com/r/flipperzero/comments/zavpjs/flipper_zero_battery_issues/
also flipper zero: uses awful PH0 screws that strip if you look at them wrong

good job folks :)

ignaloidas,

@whitequark @jacqueline Torx Plus is still somewhat hard to get, even though the patent expired more than 10 years ago...

Basic Torx is still pretty damn good for consumer electronics use cases thankfully

whitequark, to random
@whitequark@mastodon.social avatar

so, house heating saga continues

as far as i can tell from the measurements i took, it works like this: running the heating system for 5 minutes raises the temperature at the desk where I sit by almost exactly 0.2 degrees centigrade

so... I could write a script that looks at the current temperature and runs heating for 5*(20-t)/0.2? and then put it into cron

ignaloidas,

@whitequark I'd say for heating that responds that slowly, I'd go for a PD controller, with D dominating - check every hour (or maybe less), and see how much the temperature changed - if it increased, drop the heating period length, if decreased, raise the heating period length, with some absolute control so that it stays near the set-point, something like 10 extra minutes for every degree under setpoint, -10 minutes for every degree over setpoint should work.

The longer the response time the more useful the D is in my (admittedly limited) experience.

ignaloidas,

@whitequark oh, within a single degree is a serious challenge I'd say. This approach could probably still work, but would probably take a lot of tuning to get there, good luck.

gsuberland, to random
@gsuberland@chaos.social avatar

SD cards are the literal worst.

they've expanded to be the size of small hard drives, and devices like the rpi keep using them as boot media, but they:

  • use garbage tier low endurance flash cells internally
  • have little to no overprovisioning for wear
  • perform only the most basic wear levelling
  • have no protocol level integrity checking
  • have few internal error correction features, if any
  • decay comparatively quickly without patrol scrubs
  • do not perform patrol scrubs
  • cannot do PLP
ignaloidas,

@gsuberland what's really annoying is that they could do a lot of that, but don't because it's marginally cheaper and for most consumers, it's difficult to notice the regularity of their failures.

eMMC is essentially the same stuff but soldered on, but because it's always used in a large scale, companies will notice all of that stuff, and therefore money will be spent on reliability.

realistically, the only thing SD cards couldn't do is PLP

azonenberg, to random
@azonenberg@ioc.exchange avatar

What's the go-to option for RGB status indicator LEDs that contain an integrated controller so I can control a lot of them from a handful of pins? Are people still using the WS2812 or are there better options these days?

ignaloidas,

@azonenberg There are some other series, some with the same protocol, some with a different ones like SPI, depends on what package, colors and protocol do you want.

Miaourt, to random
@Miaourt@raru.re avatar

Wait wtf wasi no don't also take this path :neocat_sad:

https://kerkour.com/webassembly-wasi-preview2

ignaloidas,

@lanodan @Miaourt if anything, restricting ourselves to things that C can easily represent is a mistake, because C can't represent shit.

You can't get any kind of guarantees on memory management, behavior, etc in C - that shit has to be done manually in C, and it fucking sucks.

I cannot throw a Python object into C and just trust that it's refcount will stay correct - even though that could be done with Rust or Ruby.

C is the shit underpinnings that keep language interop so shit that the industry recognized way for language interop is HTTP - and it's not good language interop, but at least it will fucking work with C.

azonenberg, to random
@azonenberg@ioc.exchange avatar

First light on the Thunderscope beta after grabbing an upstream driver fix. There's definitely some rough edges I'm going to need to fix before it hits production, but I'm really excited - it's FAST!

With ngscopeclient running directly on the laptop the scope is plugged into (i9-10885H, Debian Bookworm, Quadro RTX 3000) I get 17 WFM/s for 4 channels @ 250 Msps with 10M points memory depth. That's 680 Ms/s (out of a theoretical 1000 coming off the ADC). At 8 bit sample size that's 5.44 Gbps, around double my previous record (PicoScope 6824E).

With the TS.NET bridge server on the laptop and a 10GbE Thunderbolt dongle connecting the laptop to my LAN, then ngscopeclient running on my main bench workstation, I "only" get 7-10 WFM/s - about half speed and competitive with the PicoScope.

There's likely room to optimize all of these numbers, this is just an initial sanity check.

ignaloidas,

@azonenberg Cool stuff

Do you know if it's using the "Legacy" TB3 communications or was it "upgraded" to USB4 (slightly nicer protocol to deal with I'd say, less Intel proprietariness).

ignaloidas,

@azonenberg yeah, 10GbE would make sense for larger labs

I can't find how they're doing the TBT3 in the repo, was wondering if maybe it could be done with just a simple code update

GossiTheDog, to random
@GossiTheDog@cyberplace.social avatar

For those who aren’t aware, Microsoft have decided to bake essentially an infostealer into base Windows OS and enable by default.

From the Microsoft FAQ: “Note that Recall does not perform content moderation. It will not hide information such as passwords or financial account numbers."

Info is stored locally - but rather than something like Redline stealing your local browser password vault, now they can just steal the last 3 months of everything you’ve typed and viewed in one database.

video/mp4

ignaloidas,

@GossiTheDog "Organizations that aren't ready to use AI for historical analysis" is such a wonderful phrase for "anyone who thinks even a little about the consequences of this"

lanodan, to random
@lanodan@queer.hacktivis.me avatar

Lol a USB → M.2-NVMe case can cost more than an M.2-NVMe SSD.

ignaloidas,

@lanodan yeah, high speed ones definitely. Throwing bits around over a wet string can get pretty expensive.

niconiconi, to random

After a whole day and night of debugging I've finally found why my simulation results were all garbage...

ignaloidas,

@niconiconi how the hell in the pit of random features that is C++ there is no newtype equivalent? Like, that is one of the better features in type systems that are not awfully anemic, and it's surprising that C++ being the language feature katamari that it is hasn't picked it up yet.

ignaloidas,

@niconiconi does it? all the solutions I've seen were essentially "wrap it in a struct or something"

I get you on the fact that it wouldn't catch everything, but it can certainly help.

lanodan, to random
@lanodan@queer.hacktivis.me avatar

Random size fruits/vegs the best.
screenshot_mpv:[SubsPlease] Yuru Camp S3 - 07 (720p) [CE7E2E00]@00:07:30.993.png

ignaloidas,

@lanodan @00 osm doesn't even truly map topographical info IIRC

there is opentopomap, but that uses NASA data for elevation. OSM only really stores vector data, where you kinda need rasters for elevation.

ignaloidas,

@lanodan IDK, I think for some deals could be made (e.g. it Lithuania OSM is mapped purely using official 1:10000 ortophoto maps after some talking with national land department, and they also do topographical maps, so definitely a possibility), but it's just that it's hard to map that into OSM and low-res openly accessible elevation info is ok enough (see opentopomap)

lanodan, to random
@lanodan@queer.hacktivis.me avatar

Windows getting dtrace before Linux. :whyMelt:

ignaloidas,

@lanodan IIRC there is a new dtrace on linux version that is all on ebpf without any custom kernel patches from Oracle of all things

ignaloidas,

@lanodan Yeah https://lore.kernel.org/all/ZhBRSM2j0v7cOLn%2F@oracle.com/

ignaloidas,

@lanodan I think it just depends on what kind of hooks you allow for eBPF? I'm fairly certain plenty of them cannot affect the system.

  • All
  • Subscribed
  • Moderated
  • Favorites
  • •
  • megavids
  • InstantRegret
  • magazineikmin
  • thenastyranch
  • modclub
  • everett
  • rosin
  • Youngstown
  • slotface
  • ethstaker
  • mdbf
  • kavyap
  • osvaldo12
  • DreamBathrooms
  • anitta
  • Durango
  • ngwrru68w68
  • tester
  • khanakhh
  • love
  • tacticalgear
  • cubers
  • GTA5RPClips
  • Leos
  • normalnudes
  • provamag3
  • cisconetworking
  • JUstTest
  • All magazines
    •