4 ms·
How the hell do they do it! I can't even reduce 100 ms latency from my API calls.
by sidcool 3y ago
How the hell do they do it! I can't even reduce 100 ms latency from my API calls.
- jb1991 3y agoDon't be too hard on yourself. We can't all be Nobel laureates.
- quickthrower2 3y agoDon't tell the HFTs about these laser thingies!
- xxs 3y agoI know it's said in a jest. 100ms overall is extra high (not just reduce).
- spacemadness 3y agoSuperimposing light waves at various frequencies—-similar to how some sound canceling headphones work with sound waves, where sounds cancel if of different phases—where the top of one wave starts at the same time the bottom of the second wave does. It does this by creating overtones when the light hits a gas. It’s not just turning on light and turning it off again like a switch. Maybe try superimposing your API calls.
- alex_c 3y agoLinked elsewhere in this thread: https://www.nobelprize.org/uploads/2023/10/popular-physicsprize2023.pdf https://www.nobelprize.org/uploads/2023/10/popular-physicspr... I understand (at an undergrad level) superimposing light waves. But this blows my mind: At the same time, Ferenc Krausz and his research group in Austria were working on a technique that could select a single pulse – like a carriage being uncoupled from a train and switched to another track. The pulse they succeeded in isolating lasted 650 attoseconds
- spacemadness 3y agoUnderstanding it at a high-level certainly doesn’t take away from the achievement one bit. You still have to be incredibly precise and know the techniques inside and out. Obviously everyone agrees since they received a Nobel Prize.