Posts by funkylab@mastodon.social
(DIR) Post #B5x1QV6PYODjg2Gvb6 by funkylab@mastodon.social
0 likes, 0 repeats
@azonenberg I doubt your pockets will be deep enough for NVlink things involving anything but GPUs :)
(DIR) Post #B5x1w5bHuAKPqnmEYy by funkylab@mastodon.social
0 likes, 0 repeats
@azonenberg I was assuming that you'd probably (assuming infinite money) could buy an Nvidia server platform that has network->VRAM piping (I assume this because I presume that's what nvidia bought mellanox for)
(DIR) Post #B5xHXk3fweQ9M6T248 by funkylab@mastodon.social
0 likes, 0 repeats
@CliffsEsport @azonenberg oh, that's not very similar to the Ethernet you know. The MAC mechanism is *completely* different.
(DIR) Post #B5xJEXayG2SR6n0rTM by funkylab@mastodon.social
0 likes, 0 repeats
@azonenberg @CliffsEsport waitwaitwait, you're referring to something that's not ×Base-T1S, right? Because that has PHY-Level Collision avoidance, with a master controlling transmission opportunities based on Node IDs (and methods to keep the system running should the master fail, by another node taking over that job).
(DIR) Post #B5xJkJa1twFCuLivVg by funkylab@mastodon.social
0 likes, 0 repeats
@azonenberg @CliffsEsport "Just" running full-duplex on a single fiber is still a bit cool to me, but yeah, that thing (maybe aside from the "master/slave" phase during link training) a lot more classical.100/1000-T1S is indeed a lot more like what happens when CAN, someone who hates CAN, and Ethernet raise a protocol child together (lovingly, but competitively). (aunt Firewire comes by once in a while to tell stories of address arbitration)
(DIR) Post #B6SDc7p59cLyInEYAi by funkylab@mastodon.social
0 likes, 0 repeats
hey, @poleguy , that's not cool: check out the pieces @azonenberg commissioned from the same series, done in a very similar style.I get the "let's not boost AI art-simulacra", but this universal "let's doubt everything" that people are getting into lately really hurts real people.
(DIR) Post #B6SIDuw9uZKl9rotiS by funkylab@mastodon.social
0 likes, 0 repeats
@poleguy @azonenberg I just realized they might not have survived all social media shifts, so, as a start:https://ioc.exchange/@azonenberg/115497634614768332
(DIR) Post #B6SILRz9RAD7gNoNiS by funkylab@mastodon.social
0 likes, 0 repeats
@azonenberg @poleguy hm that post I linked to above doesn't work if viewed from my "home" instance, but works from yours: "bucket not found" error in a JSON object. Maybe this fares better:https://ioc.exchange/@azonenberg/116535645863180873
(DIR) Post #B6SKxrFGAav4w9mvFg by funkylab@mastodon.social
0 likes, 0 repeats
@azonenberg @poleguy I should probably link to https://ioc.exchange/@azonenberg/114051170225875530 , though
(DIR) Post #B6XCEO5XtoU3dQqpt2 by funkylab@mastodon.social
0 likes, 0 repeats
@dtl @azonenberg huh, my LEDs at least according to their datasheet should emit a spectrum pretty much centered on 370nm, that shouldn't make much of a differnce. Need to work on better extraction of chlorophyll, it seems.What I use my UV source for: excite little plastic phosphorescent seal figurines, tbh. (and also: prototype for multispectral honey analysis)
(DIR) Post #B6XHLQX9hnov16wQrI by funkylab@mastodon.social
0 likes, 0 repeats
@azonenberg @dtl PLCC 2835. Haha, nowhere 1 W Popt, this is safe without protective eyewear.Xinglight XL-2835UVA-02. (and I found ny mistake – these are 390nm, not 370 nm)
(DIR) Post #B6XHX783g5WkJDm7tY by funkylab@mastodon.social
0 likes, 0 repeats
@azonenberg @dtl "close to toy safety" was actually more important than any other optical spec on this.
(DIR) Post #B6XHaoIFeMkT39J04m by funkylab@mastodon.social
0 likes, 0 repeats
@azonenberg Well we've never met and you might just be really really tiny?
(DIR) Post #B6deP7etHxtvWr3BBo by funkylab@mastodon.social
0 likes, 0 repeats
@azonenberg it's proprietary display-over-USB-Ports thing
(DIR) Post #B6jnomcQgDU3hgigiG by funkylab@mastodon.social
0 likes, 0 repeats
@azonenberg I'd really be curious how well the quadcore RISC-V lends itself to high speed data shuffling
(DIR) Post #B6qpwDDZk8syF1U5NQ by funkylab@mastodon.social
0 likes, 0 repeats
@azonenberg @datenwolf yeah and Apple M-series has good RAM bandwidths. An RPi, is judging from the single channel and CPU speeds, probably very lucky if what it can cumulatively do between memory and CPU is 10 GB/s, which would be around a factor of 6 less than the old-ish apple M1, and 25 times less than the M4 Pro. Also, how on earth would you get any data into a set-top box SoC like the one in the RPi5? it's maximum peripheral bandwidth is limited by, soldering included, two separate PCIe2x1
(DIR) Post #B6tTBij7A6F0ptXfSC by funkylab@mastodon.social
0 likes, 0 repeats
@azonenberg yeah, never keep time in floating point; it's typically not a thing where that makes a lot of sense (this was learned by SDR software *several* times). Your timestamps should just be int64 multiples of some pre-defined time quantum, and you'll be fine. You don't need sub-attosecond accuracy, and neither do you need to be able to measure in millions of years; the fixed 64 bit range suffices.
(DIR) Post #B6tUY5LDrUIZIbawxU by funkylab@mastodon.social
0 likes, 0 repeats
@azonenberg yep, need to shift the time values to a "local environment" if you will (extreme example: the interpolator doesn't care about absolute time, you can always shift all times by an integer value so they're zero-centered, locally)
(DIR) Post #B6tVzUKVACH8ZWt8tc by funkylab@mastodon.social
0 likes, 0 repeats
@azonenberg exactly! (and that's, surprisingly, something that the vernerable and deservedly-so universally disliked libc `struct timespec` gets right, in principle.
(DIR) Post #B8It1O7F5IOvkYwuYK by funkylab@mastodon.social
0 likes, 0 repeats
@mntmn Wider der Bugs' Verdruss / ich kconfigs lesen muss