VHRanger

48 ms·

691.
by VHRanger 4y ago
You should think more of vulkan as an IR endpoint than the actually usable API here. Vulkan is well supported by most GPUs because it's so low level. Performance tends to be good everywhere. What would make vulkan succesful is having A
692.
by VHRanger 4y ago
There hasn't been reliable benchmarks to that effect. In fact, Oracle and MS SQL specifically disallow posting benchmark comparisons publicly
693.
by VHRanger 4y ago
Why is anyone still giving that grifter any attention?
694.
by VHRanger 4y ago
yes, rugpulls are known to be very uncommon in the crypto world
695.
by VHRanger 4y ago
Wikipedia has always had an issue on some pages with tankies whitewashing some brutal monsters. Similarly to corporations/politicians editing their pages FWIW.
696.
by VHRanger 4y ago
Nonsense. Presumably any private equity firm would also have came to a conclusion of reducing headcount. They wouldn't have done so in a single week based on ridiculous analysis. And wouldn't have done so in a way to destroy the m
697.
by VHRanger 4y ago
Its more common in heads up, yes. In multiplayer you see it where ranges are narrowed, like 3bet pots or on turn/river It matters less than you'd think because overbets imply you have a polarized range (nuts or air). You generally
698.
by VHRanger 4y ago
Correct, but there are likely applications of AI in strategic reasoning that are used in the real world (outside the obvious finance bots) by engineering a "translation layer" from real world constraints to AI-compatible value fun
699.
by VHRanger 4y ago
All Poker AIs developped as yet approaches Nash Equilibrium -- it's just a "perfect" strategy that wins by default because it makes no mistakes. Since you make mistakes against the AI strategy, and the sum of the game of poke
700.
by VHRanger 4y ago
FWIW one of the big things poker AI taught humans is massive overbets (eg. going all in for $200 over a $15 pot). This is scary to do well in practice, because the mathematically optimal bluff frequency approaches 50% as you increase the ov
701.
by VHRanger 4y ago
AI remains better than humans at anything that has well defined rewards and small time gap between action and feedback mechanism (either naturally, like poker, or by value function engineering, like Go or Chess) The problem here is that it&
702.
by VHRanger 4y ago
Nonsense. If the throughput is lower, the latency is necessarily higher because you'll be waiting on the dock longer for a car to pick you up.
703.
by VHRanger 4y ago
Putting trains in the tunnel would be pretty exciting! Imagine how much more throughput you could get out of the system this way!
704.
by VHRanger 4y ago
If you're writing a new project, it's a good idea to try to have it compile on all major compilers (and platforms, if possible) from the get-go. You catch a lot of these errors and UB. It will also force you to code more idiomatic
705.
by VHRanger 4y ago
$65B in alleged assets is what they should have said.
706.
Digital Currency Grift(dirtybubblemedia.substack.com)
4 points by VHRanger 4y ago|1 comments
707.
by VHRanger 4y ago
Moving the goalposts to something vague and further in the future wont help you convince anyone about usefulness
708.
by VHRanger 4y ago
Nobel winner James Heckman did a lot of work on this [1]. The takeaway is that IQ doesn't matter as much as other attributes (character, learned skills, etc.) in life outcomes. HN style crows way overvalue IQ style skills when intelli
709.
by VHRanger 4y ago
More like bribery But according to the US supreme court, money is speech so it doesn't matter
710.
by VHRanger 4y ago
Maybe taking money from crypto gamblers and giving it to progressive political causes is the true effective altruism!
711.
by VHRanger 4y ago
Go to any casino at 4am and you'll find the same mentality. He's just angry he got kicked off the blackjack table because now he can't make it back up by gambling more
712.
by VHRanger 4y ago
So there's no other conclusion than this website was run by drugged up children. However bad you think FTX was run, it's worse. They basically didn't keep track of anything, disbursed money to buy themselves houses in the Bah
713.
by VHRanger 4y ago
The answer is that they almost certainly are securities, but the precedent isn't established. White collar prosecution is expensive and complex, and prosecutors care about their careers. They want certain wins, preferably fast. So the
714.
by VHRanger 4y ago
I think he meant theyd stop trading but keep market making Or he just lied yeah
715.
by VHRanger 4y ago
Bitcoin failed, though? All the proposed usecases except speculation and censorship resistant wealth transfers (eg. getting money out of Russia) have been abandoned.
716.
by VHRanger 4y ago
They used FTT as collateral for loans used on trading activities that boosted FTT value It's not literally a Ponzi, but it's Ponzi adjacent behavior, like what Bill Hwang did.
717.
by VHRanger 4y ago
It's a race for who was most deluded in 2021!
718.
by VHRanger 4y ago
Russians paint "Z" on their combat vehicles
719.
by VHRanger 4y ago
Yes, which is why compiling on different optimization settings will have bounds checking on or off in C++
720.
by VHRanger 4y ago
Yes, and security has a large performance impact. Just look at the performance costs of bounds-checking array access in C++ code. Or more macro, the performance impacs of AV tools or Windows Defender on your system

More ›