tgtweak

9 ms·

91.
by tgtweak 8mo ago
Yes I've run many here in Canada - but you still need to keep intake air very clean and they will absolutely rust out or short from dirt ingestion if filters and humidity are out of check. They're also designed with higher tolera
92.
by tgtweak 8mo ago
Tiktok ads, Youtube Ads, Instagram/Meta ads - there's just a huge influx of scams and obviously fake sites on them. AI generated copy, AI generated landing pages... My honest take on it is that it's the payment companies tha
93.
by tgtweak 8mo ago
The simple fact the back button while on the main screen doesn't exit the app is something that honestly should be illegal and is not permitted in the app stores.
94.
by tgtweak 8mo ago
Or quicker, or more comprehensively for the same price.
95.
by tgtweak 8mo ago
It's not a static number as it's also based on ambient air temperature in the form of dew point - 45% RH at low temps can be far more dangerous than 65% RH at warm ambient. Likewise the impact on server longevity is not a finite b
96.
by tgtweak 8mo ago
>San Diego has a mild climate and we opted for pure outside air cooling. This gives us less control of the temperature and humidity, but uses only a couple dozen kW. We have dual 48” intake fans and dual 48” exhaust fans to keep the air
97.
by tgtweak 8mo ago
Claude itself (outside of code workflows) actually works very well for general purpose chat. I have a few non-technical friends that have moved over from chatgpt after some side-by-side testing and I've yet to see one go back - which
98.
by tgtweak 8mo ago
plot twist, it's all claude code instances submitting bug reports on behalf of end users.
99.
by tgtweak 8mo ago
I don't think there is calculus that makes the cost of getting it to space worth putting it there - Elon of all people knows this - if you scale up the ground-based solar and add battery storage, the costs are still far lower than tryi
100.
by tgtweak 8mo ago
They actually need the entirety of the backside of the panels to cool them - if not they would literally burn out from the accumulated heat from being exposed to the sun.
101.
by tgtweak 8mo ago
Added some math on my comment that outlines the boundaries of economics of this considering most of what you mentioned.
102.
by tgtweak 8mo ago
The crux of it is that radiative cooling in space is dependent on exposed surface, not total surface area like a radiator or heatsink on earth - this means if you just put an existing finned heatsink in space it would radiate most of it&#x2
103.
by tgtweak 9mo ago
You can extend this by looking at the IP route for the reverse path, I've found it's usually accurate to the state at least on the last hop before destination - added benefit that there's usually an airport or city code on th
104.
by tgtweak 9mo ago
is a wasm sandbox as secure as a container or vm?
105.
by tgtweak 9mo ago
https://www.digitimes.com/news/a20260128PD213/tsmc-intel-nvi...
106.
by tgtweak 9mo ago
gpt-oss is pretty great tbh - one of the better all-around local models for knowledge and grounding.
107.
by tgtweak 9mo ago
I find Gemini does the most searching (and the quickest... regularly pulls 70+ search results on a query in a matter of seconds - likely due to googlebot's cache of pretty much every page). Chatgpt seems to only search if you have it
108.
by tgtweak 9mo ago
Very curious for what use cases you're finding gemini unusable.
109.
by tgtweak 9mo ago
5.2 is back to being a sycophantic hallucinating mess for most use cases - I've anecdotally caught it out on many of the sessions I've had where it apologizes "You're absolutely right... that used to be the case but as o
110.
by tgtweak 9mo ago
Much like talking to your doctor - you need to ask/prompt the right questions. I've seen chatgpt and gemini make one false assumption that was never mentioned and run with it and continue referencing it down the line as if it wer
111.
by tgtweak 9mo ago
Literally just hit the MPO bug yesterday - what a shitshow that things like display scaling can fail with a mainstream gpu/cpu and windows 11 - like do they not have testing rigs for this in microsoft? if insiders preview is fine but p
112.
Using Claude Code to analyze your genome for interpretation [video](youtube.com)
1 points by tgtweak 9mo ago|1 comments
113.
by tgtweak 9mo ago
Caveat: usual disclosures about using 23andme, and public LLMs for your entire genome.
114.
by tgtweak 9mo ago
I think it's impressive for what it is: this level of complexity being reached by an ai-only workflow. Previously, anything of modest complexity required a lot of human guidance - and even with that had some serious shortcomings and c
115.
by tgtweak 9mo ago
Definitely have some memories from 3 years old - some people claim earlier and I wouldn't doubt that, although it's very rare for memories before 2 to be recalled episodically.
116.
by tgtweak 9mo ago
Seems most major players are not only fabless but also fab-agnostic - as I noted they switch from one supplier to another even for the same product line. I'm sure it is work but it doesn't seem to be an existential crisis for a h
117.
by tgtweak 9mo ago
Prognosis is then that work for m4/m5 should be relatively straight line now that refactoring is done?
118.
by tgtweak 9mo ago
Rate limits mostly - plus claude code is a relatively recent thing but sonnet api has been around for a while with 3rd party apps (like cline). In those scenarios, it was only api.
119.
by tgtweak 9mo ago
He hasn't offered any production services to third parties whatsoever (seems to end at patents/standards gifting) - I wouldn't expect a fab open to third parties at all - if anything it'll be reserved for tesla/spac
120.
by tgtweak 9mo ago
If Intel decides they're not going to continue foundry services after 14A - you can just shift back to TSMC like everyone did between Samsung and TSMC?

More ›