Remote
Bart
@bartreardon@aus.social
I work with mac and linux - I also write the odd app or two.
274
Followers
206
Following
18
Posts
Joined April 28, 2022
Photos (sometimes):
He:
Him
Some of my projects:
Posts
Replying to
@ckeen@social.vernunftzentrum.de
@ckeen@social.vernunftzentrum.de @NanoRaptor@bitbang.social "By the grace of Shaddam IV of House Corrino, ascendant to the Golden Lion Throne of Padishah Emperor of the Known Universe, I stand before you as Herald of the Change."
Replying to
@NanoRaptor@bitbang.social
@NanoRaptor Physically mark and identify compulsively curious people with this one weird trick.
I would have a yellow index finger so quick.
Replying to
@NanoRaptor@bitbang.social
@NanoRaptor@bitbang.social Don't eat the yellow snow white design.
Replying to
@mattgrayyes@chaos.social
Replying to
@decryption@aus.social
@decryption The hype about them is it's not a mac - with a 9B model my studio (M1 Ultra) is faster using ollama. The GB10 is about 5% faster on a 16-20B model and the Studio dies with a 30B model which eats about 70Gb of RAM on the GB10. An M5 Ultra Studio would shit all over these. Even a handful of the M3 Ultras chained together beat out a GB300 and for way less than half the cost. (still expensive - like $80k vs $180k)
But the GB10/300 is not a mac if you're not into that and I think you can hook up as many as you want over QSFP but I only have one to play with so can't test that out :). It's basically an nvidia gpu in a box with an arm cpu attached to run an OS on.
Replying to
@decryption@aus.social
@decryption I was playing around with some local models today on a mac studio with 64Gb and a dell GB10 with 128Gb. Managed to crash the studio because it didnt have enough ram. The models I could run on the GB10 pretty much suck compared to claude or whatever. Those are $6000-$7000 devices. Local LLM's are better than they used to be but are barely better than curiosities on anything within reasonable reach of mortals.
Replying to
@talkingmoose@mastodon.social
@talkingmoose@mastodon.social Still waiting for the day I wake up to hear he's been turfed out or dead. Either works for me.
Replying to
@siracusa@mastodon.social
@siracusa@mastodon.social Somewhere in Apple Park is an unopened package an intern neglected to deliver. In it, a t-shirt that could have inspired and made all the difference.
That said, they could backtrack today and announce an M5 Ultra mac pro with upgradable GPU and RAM and I would know 0 people that would buy one. I think it's an important product to have existed but market reality has spoken. As nice as it is, the negative ROI on that case design is probably second only to project titan.
Replying to
@decryption@aus.social
@decryption@aus.social @NanoRaptor@bitbang.social Anything can happen. NMA has a 1955 FJ Holden. Someone drove it around for 25 years and now it's in a museum. Beats trying to flog it off on facebook marketplace I guess.
Replying to
@NanoRaptor__dup_1737@bitbang.social
@NanoRaptor@bitbang.social It has a retina display but only once.
swiftDialog has a (version) birthday today
3.0 is out now with new stuff and better old stuff
https://github.com/swiftDialog/swiftDialog/releases/tag/v3.0.0
Replying to
@siracusa@mastodon.social
@siracusa@mastodon.social Mr Incredible Meme: "5 is 5"
But yes, it's quite different 🙂
Also Perl 6 aka Raku aka wtf is raku?
Replying to
@caseyliss@mastodon.social
@caseyliss@mastodon.social @siracusa@mastodon.social To be fair, the python people care about is 3 which was released in 2008 and perl 5, the one people usually associate with "perl" was 1994.
Replying to
@standupmaths@mathstodon.xyz