I take my shitposts very seriously.

  • 16 Posts
  • 1.2K Comments
Joined 2 years ago
cake
Cake day: June 24th, 2023

help-circle



  • AI would be chronically incapable of implementing actually surprising plot twists that are both unexpected and consistent with the rest of the plot (and not somehow someone back into existence). If it hadn’t been written before, an AI would never make Darth Vader be Luke’s father unless specifically prompted, at which point, why even.

    (I’ve just finished a hexalogy marathon, my head is full of jedi.)



  • I meant games that had tutorial videos built into them. Stuff like Syphon Filter; a rushed, poorly voiced video that lists your controls and tosses you into the mission. The player is told what does what and isn’t given a chance to learn how to interact with the world.

    Soul Reaver 1’s first 20 minutes is what every game should be aiming for. You learn how to navigate the world, how health and the spectral/material realms work, how to solve the combat puzzles, and more importantly, how those systems interact; then you’re on your own. If a game needs the help of extrenal resources to convey such basic information, it’s a failure of game design. Not necessarily out of incompetence but because game design principles hadn’t evolved to that point.

    I’m not against external (including physical) resources, iff they’re used in a clever way. Shenzhen IO has a thirty-page manual themed as actual technical documentation about the electronics used in the game. Through this, the manual becomes part of the game. Same for Keep talking and nobody explodes. Volo’s Guide to Baldur’s Gate is a fantastic example of presenting supplemental information that is good to know but isn’t a roadblock in its absence. If a manual improves the game experience, it’s good material. If it’s necessary to make a game playable, it’s bad design.



  • When your eyes are open and unobscured, light is coming in from every direction. The lens is shaped in such a way that light rays parallel to the eye’s axis are focused on the macula, the center of your sharp vision. A near-sighted (myopic) eye focuses those parallel rays in front of the retina, and a far-sighted (hypermetropic) eye focuses them behind. The farther away the ray is from the eye’s axis, the more it is refracted by the lens, and the more obvious its out-of-focus-ness becomes if the lens has an incorrect shape.

    Corrective eyewear works by refracting the light before it enters the eye and essentially cancelling out the lens’ imperfections.

    A pinhole works by obscuring light rays that are farther from the axis and contribute to the blurry image, only letting through light rays that are near the axis, already aligned more or less with the macula, don’t have to be refracted as sharply, and don’t contribute as much to the blurry image. This is why the camera obscura works, and why apertures in modern photography are used to control both the image’s exposure and the strength of the depth of field.











  • No.

    The local machine boots using PXE. Clonezilla itself is transferred from a TFTP server as a squashfs and loaded into memory. When that OS boots, it mounts a network share using CIFS that contains the image to be installed. All of the local SATA disks are named sda, sdb, etc. A script determines which SATA disk is the correct one (must be non-rotational, must be a specific size and type), deletes every SCSI device (which includes ATA devices too), then mounts only the chosen disk to make sure it’s named sda.

    Clonezilla will not allow an image cloned from a device named sda to be written to a device with a different name – this is why I had to make sure that sda is always the correct SSD.