Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

> it's easier to do that estimate on something following basic physics

Cameras and microphones follow the exact same basic physics as the op-amps in in the OP's circuit.



Don't many use some kind of pre-processing or analog tricks in their products? It's what A/V people often told me.


Sure, but so what? That's my whole point: none of that is necessary, or even helpful. All you need is a good lower bound on your noise quality estimate, multiply by 10 or 100 or 1000 depending on how you want to trade off speed for paranoia, and you're done. Anything else is a distraction.


"Sure, but so what? "

Point being that a person with a baseline of electronics training knows exactly what one or two basic components are doing right next to each other. Products like A/V that use extra tricks can make the analysis more difficult or just take more knowledge. I lean toward the simple ones. A mic or camera sensor with nothing messing with its input might be fine, too.

"All you need is a good lower bound on your noise quality estimate, multiply by 10 or 100 or 1000 depending on how you want to trade off speed for paranoia, and you're done."

That part sounds right.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: