@i2cjak Old coworker had the idea that optical feedback would be the solution to heavy, rigid machinery. "Just work at it till it looks right" instead of "tool move here"
@OwenBenjamin the only part of the moon where the earth would be at high noon would be right in the middle of the face of the moon we stare at from earth. if you are at the edge of the disk we're looking at, youd be seeing the moon along the horizon. its "static" in the sky per location
@OwenBenjamin@grok its not accurate because its not "rising" thats just where it sits in the sky on that part of the moon according to the model in question
@haydendevs better to have the copper there and not need it than need it and not have the copper.
sometimes its for debug. sometimes its for shipping different flavors/functionality.
@AshtonForbes do you have evidence outside of the same video you tweet every time? do you have any new developments or are you just pumping the same dry well every time?
@jwt0625 they sell on board optics and have for like 15 years. avago had the micropods/minipods. you dont do OSFPs, you terminate the optics to something intermediate like MPT, then take it the rest of the way out of the system.
nobody does this because its expensive and power hungry
@AshtonForbes Can you leave the space just if you never actually get around to demonstrating the science you talk about? like with math or simulations or other stuff? instead of just showing the same old video and making claims?
@realwillreil Might try a different flux. Lotta "no clean" flux is inert after reflow and not meant to be cleaned but is also near impossible to do so.
@michael_chomsky@OpenAI@asciidotdev Retail on those GPUs is well into 5 figures. You probably have 8 ARM cores on your phone. CPU time is very, very cheap comparatively.
Also note they're not giving VMs to every customer, just the paying ones.