a-dub

12 ms·

211.
by a-dub 1y ago
rust is kinda ugly. i think i like zig better.
212.
by a-dub 1y ago
i recently worked with a macbook pro and it caused uncomfortable feelings of eyestrain. i had some app that was supposed to disable the temporal dithering but i'm not sure if it helped. i'm curious if there's anyone else on h
213.
by a-dub 1y ago
going all the way out to percept, the response of the system is non-linear: https://en.wikipedia.org/wiki/Mel_scale this is believed to come from the shape of the cochlea, which is often modeled as a filterbank that ca
214.
by a-dub 1y ago
> At high frequencies, frequency resolution is sacrificed for temporal resolution, and vice versa at low frequencies. this is the time-frequency uncertainty principle. intuitively it can be understood by thinking about wavelength. the m
215.
by a-dub 1y ago
the joke i like to make: it's a linked list with an O(exp(n)) append cost and a O(r*exp(n)) rewrite cost where r how many from the tip you wish to rewrite. other notable: the 32 bit nonce is exhausted very quickly with modern mining sy
216.
by a-dub 1y ago
i thought they were all built on the compression functions from secure hashes these days?
217.
by a-dub 1y ago
i see emojis in the comments. also no discussion of measured runtimes for the rendering code. (if it saves ~1.3ms that sounds cool, but how many ms is that from going over the supposed 16ms budget.)
218.
by a-dub 1y ago
" ... BUT most importantly, ... " i love the use of all capitals for emphasis for important instructions in the malicious prompt. it's almost like an enthusiastic leader of a criminal gang explaining the plot in a dingey dine
219.
by a-dub 1y ago
wouldn't it be like a crime against the crown to not have a cheri like thing in arm?
220.
by a-dub 1y ago
ahh man you just reminded me. the cd-roms on those machines didn't have the silly audio cable that pcs of the time did, rather they were just connected via scsi. playing audio cds would rip the redbook cd audio stream in real-time and
221.
by a-dub 1y ago
the hardware support was pretty bad. it ran on x86 with only the most basic of hardware with a vga graphics adapter in some basic vesa mode. it never felt serious, just more like a technology demo. it was also hard to really see the beauty
222.
by a-dub 1y ago
hah! i forgot all about quake c. in 1997 i had a windows nt 4.0 machine (might have been a beta) on my desk and a next turbo mono slab. visual studio wasn't terrible for development (and was a great place to get good at c++ and multith
223.
by a-dub 1y ago
lenovo p1 and x1 are pretty nice. lunar lake machines are posting macintarm style benchmarks for battery life for light tasks. the screens are matte by default. you can upgrade ram and disk rather than discard (all the framework hype has be
224.
by a-dub 1y ago
that's super cool! live and learn. even better would be the capability to drop caches from a supplied point in the filesystem hierarchy.
225.
by a-dub 1y ago
it would. it would just require pki and a secure enclave that lives directly on the imaging chip to support it.
226.
by a-dub 1y ago
> CSCI 2100: Unlearning Object-Oriented Programming but 90s fashion is all the rage these days!
227.
by a-dub 1y ago
> Local LLMs' speed can't be generalized, as the speed of each instance is entirely determined by its particular runtime environment. sure. today's on-device LLMs are either slower or less capable by orders of magnitude co
228.
by a-dub 1y ago
doing these sorts of benchmarks is actually quite tricky. you must clear the page cache by allocating >1x physical ram before each attempt. moreover, mmap by default will load lazy, where mmap with MAP_POPULATE will prefetch. in the form
229.
by a-dub 1y ago
i recently worked with a mac and was pretty unimpressed with all the visual clutter in the upper left corner of the screen. it was just kinda cluttered up with a junkyard of wasted space and needless controls. it didn't feel good, magi
230.
by a-dub 1y ago
detaching reporting from branded sources is a terrible idea.
231.
by a-dub 1y ago
local llms are slow. just pay for the service so they don't use your uploads. always read the outputs and don't ask for things you don't understand.
232.
by a-dub 1y ago
ok, excluding things they have bought and not yet destroyed. what's good? (we'll accept that xbox is good, distinct and unrelated to the rest of their offerings)
233.
by a-dub 1y ago
GALACTIC SCALE QMAIL that can run efficiently on a 486 AND survive a supernova!
234.
by a-dub 1y ago
the software is so bad it's literally a national security risk.
235.
by a-dub 1y ago
it's generally pretty remarkably bad. i think i agree. it sets a sort of psychological baseline culture that computers and their software should be shit, which is a pretty bad influence for people making software to be engaging with da
236.
by a-dub 1y ago
if you were able to normalize and quantokenize the distinct dct values in a consistent way, it could be an interesting approach. so yeah, undo the bit packing but keep the front end signal processing and compressed dct representation and vi
237.
by a-dub 1y ago
i am amused at how us-east-1 is basically in the same location as where aol kept its datacenters back in the day.
238.
by a-dub 1y ago
it's like the end scene in fight club. except instead of credit card company office towers it's the borrow checker and associated skyscrapers that symbolize the ascent of rust that are going down in flames as the x86 antiheroes hi
239.
by a-dub 1y ago
> This is the stupid part. You will burn CPU cycles waking up spuriously for timeouts with no work to do. Setting the flag won't wake up the event loop until the timeout hits, adding pointless delay. it's the smart part. waking
240.
by a-dub 1y ago
vim-plug with pinned hashes and manual reviews ftw!

More ›