• 0 Posts
  • 165 Comments
Joined 2 years ago
cake
Cake day: June 17th, 2023

help-circle

  • I’ve got 1 dream check, that’s fairly reliable, when I need it. I check my back pocket for “heavy weapons”. Basically, think cartoon “hammer space”. It’s an almost unnoticeable check when awake, that doesn’t do anything. In a dream state however, an ak47, or a bazooka is to hand.

    This is particularly effective against nightmares. My subconscious happily accepts that I can pull whatever cartoon doodad I need out of my back pocket. This let’s me jam nightmares. I’ve not had one since I trained myself to do this a couple of decades back.

    The 1 trick to note, you need to “believe” on some level that it will work. It’s akin to accepting a film. You know they are just actors and CGI, but you accept it as real. This belief gives it power in your dreams.


  • It’s interesting to look at what is actually required to be a technological species (assuming they develop it themselves).

    • Dextrus Manipulators.

    To make technology, you need something to manipulate the world reliably. Hands are the most obvious method, but not the only ones. Octopus tenticles could also likely fill the roll.

    • Social groups and communication.

    Developments are useless, if they can’t be passed on to the next generation, or shared around. Technology requires building on the work of others.

    • Brain development.

    There needs to be something to drive early brain development. With humans, it was likely sexual preferences. It could otherwise become a chicken and the egg type problem.

    • Generalist.

    A specialist species will tend to lean into their strengths. There’s far less need for intelligence when you have big claws, or heavy armour already. This also applies to size. Too big, or too small tends to specialise in a why the precludes other developments.

    There are several species on earth that hit some of these points, but not all. E.g. Dolphins hit all but the manipulator issue. Octopus are completely solitary. Many mammals hit all but brain development, and crabs overspecialise.

    I could easily see a small tweak leading to a radically non human technological intelligence. That is also based only on what has already developed and stabilised in the earth’s biome. The cambrian explosion showed that far more body forms are at least viable.


  • cynar@lemmy.worldtoProgrammer Humor@programming.devAI in reality
    link
    fedilink
    English
    arrow-up
    43
    arrow-down
    1
    ·
    16 days ago

    LLMs can’t become AGIs. They have no ability to actually reason. What they can do is use predigested reasoning to fake it. It’s particularly obvious with certain classes of proble., when they fall down. I think the fact it fakes so well tells us more about human intelligence than AI.

    That being said, LLMs will likely be a critical part of a future AGI. Right now, they are a lobotomised speech centre. Different groups are already starting to tie them to other forms of AI. If we can crack building a reasoning engine, then a full AGI is possible. An LLM might even form its internal communication method, akin to our internal monologue.


  • cynar@lemmy.worldtoSelfhosted@lemmy.worldWhat access points do you use?
    link
    fedilink
    English
    arrow-up
    2
    arrow-down
    2
    ·
    1 month ago

    The dream router is an excellent base to build upon. It provides all the normal functions (ethernet, wifi, router etc) as well as hosting the control software.

    Unifi’s real power is when you expand it. The access points make extending WiFi coverage easy. You dont even need a wired link. It will link over WiFi, either as a primary or as a fall back. The flex mini is also quite handy. It’s a little poe powered switch. I have a couple tucked away providing extra ports around the house.

    With my setup, I have detailed monitoring and control down to the port or wifi device. I can monitor and control things in detail, or get a high level view of my network.


  • While expensive, UniFi hardware is just a huge step beyond the rest of the consumer market.

    I’ve had literally 10x the range (5x vs 50m), in congested environments, compared to ‘gaming’ hardware. I actually did a side by side to test. I was shocked at the difference.

    The bridging function is also a life saver. 2 LR units can get a reliable signal between each other, at ridiculous ranges.




  • Most of the problem with lasers come from focusing them. The eye is incredibly good at it. This means even a small laser pen can reach MW/m^2 ranges by the time it hits the retina.

    IR is a different story (at longer wavelengths). Without the ability to see it, our eye will not attempt to focus on it. Also, our eyes lenses are not particularly transparent to it. 3rd, the ultra short pulses mean that there is no time to focus.

    As for the mosquito, the laser is tuned to a frequency that is strongly absorbed by their wings. Given their size and how delicate their wings are, a tiny amount of energy can cause significant damage. Conversely, the same energy on our eye will just cause a slight amount of heating. The bulk mass of the eye will absorb this fine, with no damage









  • There’s a particular particle, the kaon, which can be created. This particle is highly unstable, and so, decays rapidly into other particles. Ever so often, it doesn’t decay down the normal route but instead decays into a pion. This is extremely rare (6 in a billion).

    In physics, we have what’s called the “standard model”. It’s our best guess as to how physics works at the fundamental level. It’s incomplete, however, with multiple slight variations. This decay pathway is interesting because it is quite sensitive to differences between these models. By measuring the energy and ratio of the resulting mess, we can disguard some variants of the model (their predicted energy is too high or too low).

    By using a large number of little measurements, like this, scientists can home in on the most accurate “standard model” variant. This, in turn, informs work on a deeper understanding of physics.

    Basically, a decade’s work to put a single new point onto a graph. A point that only theoretical physicists care about, and might, or might not be useful down the line. Welcome to modern physics.


  • It’s a combination of both, I believe.

    The initial conditions had a definite rotational bias. This is preserved in the current orbital plane and direction.

    On top of that, anything massively off that plane is liable to hit or interact with the material in the plane, given enough time. It will be flung around, eventually either out of the system or into the plane.

    Stuff orbiting relatively close to the plane will have a biased pull towards the “average” plane. This will slowly flatten the orbits out.

    All these processes take a lot of time. The solar system, in general, has had enough time to settle. The ort cloud and other outer bodies are still quite chaotic. We see a lot more off plane than within the traditional solar system. They experience the latter effects far less, and so take longer to equalise. They still have a bias towards the initial spin however.