the problem with that is that training can’t be done “immediately” it takes tons of compute
i should be writing
the problem with that is that training can’t be done “immediately” it takes tons of compute
Copyright and IP concerns disappear with an open dataset.
i don’t think i’d agree with that, doesn’t matter if dataset goes open if content went there without consideration for authors
also even things like thispersondoesnotexist were used to mass-create fake identities and such
i’m not. just because he’s an underdog here means that you’re gonna ignore all the harms of generative ai up to this day? it’s like complaining that big oil stole the idea of adding tetraethyllead to gasoline from you and you got no profits from that as a result
You might be semi-comfortably running linux mint cinnamon on these (assuming 4gb ram) with xfce you’re trading clunkiness and ancient looks for lower memory usage
no idea about the usual suspects, wifi, bt, graphics probably will require tinkering as is tradition
how low we’re talking
xubuntu is fine if your box is a potato or if you’re coming from windows vista
i’m here to remind you that for last 20ish years half of the time chemistry nobel goes to biologists, and now they doubled down on ai wankery with giving it to alphafold
because these end up generating most of electricity. older plants matter less specifically because these are less efficient - operating them means more fuel costs per MWh. normally, you can see new flashy plants generating all the time it’s practical, because these are more efficient, have less maintenance downtime etc and when demand grows, progressively less efficient units start generating coming from spinning reserve. the two exceptions are NPPs which are best operated at constant high power because of their neutron physics and renewables that are literal free energy so everything they do is taken in. the only place where you can improve efficiency of NPPs is in turbine, and that probably is pretty well optimized unless turbine is very old, because increasing steam temperature would mean changed conditions in reactor in way that could happen to be out of spec. we have figured out wind power pretty well, and perovskites aren’t a thing, and won’t be a thing until they become more durable, which they won’t. in all cases, upgrades would have to make sense both economically and/or in emission costs. this includes CHP and laying municipal heating grids, and good luck with that with how dysfunctional american local govts are (where probably biggest emission gains from CHP could be made)
you can redo this for other types of thermal powerplants and come to the same conclusion. if you say that saltman&co and his assemblage of lying machines can outsmart thousands of turbine engineers, you might be a shill for making other people believe that or a moron for believing that yourself
CCGTs have efficiencies in the range of 60%-ish percent. bringing it up to 85% would mean that now these run almost at carnot efficiency taking adiabatic flame temperature of methane burning in air as upper and practical temperatures of heatsink (60C) as lower. this is not happening, because other cycles with lower efficiencies are used in practice
if you want to improve efficiency of power generation, just replace old junk with new kit, or better yet, build nuclear and renewables where efficiency matters less when considering emissions. you know what, damn i do think that lying box burning enough electricity to power a small country (like macedonia) could come up with this
not gonna happen, this would break thermodynamics. he made that up on the spot, he’s full of shit and he knows it
you are part of the problem
Nobody should trust LLMs with anything
ftfy
also any inputs are probably scrapped and used for training, and none of these people get GDPR
oh no, it would be so sad if all these libertarian assholes lost their livelihood in form of money laundering nerd tokens
don’t you know, it’s hate speech against corporate profits
this is not how conspiracy theories work. these start and end with need for feeling special for “having” some secret knowledge. it’s all elaborated nicely if you have an hour of unnecessary time https://www.youtube.com/watch?v=JTfhYyTuT44
easy high power generation from hydrogen would be in gas turbines, but this will have horrendous roundtrip efficiency. which is why it’d be better to soak up peak power in hydrogen and use it for non-power uses, like ammonia and then fertilizers, or direct reduced iron, or various hydrogenations in fine chemicals segment. these things take a solid chunk of energy to make. it’s net positive because it replaces gas https://en.wikipedia.org/wiki/Steam_reforming while storing hydrogen is pain it’s easier than electricity, and some intermediate can be stored too if hydrogen consumption can be surged
gas turbines are also fantastically versatile. any petroleum fraction lighter than grease, ethanol, biogas, syngas, hydrogen, ammonia, really anything that burns and can get through nozzle can be used as a fuel. if you have a carbon-neutral source of liquid fuel that can be stored, you have carbon neutral peaker plant
i thought sodium batteries need low hundreds C for ceramic electrolyte to work. i stand corrected
e: CATL made sodium-ion battery, i was thinking of sodium-sulfur battery
nvidia executive called saltman a podcasting bro