looks like a doom level
HN user
natas
For sure, I'm glad we have those emulators, real hardware is hard to find or expensive or faulty. Having emulators helps a lot.
don't they still give some hobbyist licenses?
Would be amazing to implement a JIT in hercules (would make it 10x faster potentially as fast as z/PDT which implements a JIT); good timing as IBM is retiring z/PDT!
I wish this could make it to qemu's main branch; always fun to preserve new systems. I wonder if it runs openvms/ia64 too!
shocking, who could have predicted that?
Windows source has leaked (so many times) and someone ran claude code on it to find vulns, as expected; few million more patches coming; those exploits sell well $$$
Hence, there is a huge monetary incentive to leak window's source; especially if your RSUs are not really bringing money home.
I won't make the mistake to buy any other sony product.
I bought it yesterday, not too hard to get running on linux, and played for an hour, it's immense fun.
it's a bit skewed, I suspect 25% are bots and the like that don't present an identifiable user-agent. Also, a lot of bots probably identify themselves as windows to "blend-in".
fortunately I'm old, I drive a 2007 car, and won't live in this new world for much much longer!
I'm done with sony, forever. Cameras, consoles, whatever, I'm just done. Goodbye!
I feel like we've already had this discussion with respect to Crypto; crypto used more energy than banks (which proved to be false, but whatever); and water. Now AI uses magnitude more energy than crypto (or humans) and somehow they're now saying it's okay?
Napkin maths.
Alphabet: ~$4.5T value / ~$403B revenue ≈ 11× revenue
Microsoft: ~$2.9T value / ~$282B revenue ≈ 10× revenue
OpenAI: ~$850B value / ~$13B revenue ≈ 65× revenue
Can someone explains that logic?
pretty cool!
How long before github deletes all repos? or make private repos public? and such and such?
I wish you a fast and full recovery.
100% vibe coded.
100% vibe coded
lipstick on a pig
No thanks, I'd rather use linux.
Stocks aren't prediction markets, but options and futures are.
Do you want to hear what the execs are doing about it from a trusted source instead of a marketing person: https://news.ycombinator.com/item?id=47216047#47223342
I'd take a dm42n over any Ti anyday.
I don't have a prime, but agreed on 48gx + dm42n without a doubt. I'd add the 50g as another contender, doesn't depend so much on external memory cards (which have the tendency to fail) and is fairly aligned with the 48/49 line.
Soon there will be ads on Github, you'll see.
@haileys I would love to see that on ArcaOS! I bet it can't be done.
OpenAI is a few years behind Anthropic, and it's unlikely they'll catch up at this point.
Give it a year or two, and apple or any other hardware will have unified memory OR AMD will have a good offering to run all that stuff locally. It won't be as good as Claude, but it'll do for 90% of the things. It will be expensive as first, just like the first mainframes, then give it another 5 years or so, and it'll be affordable.