3 ms·
Nice find, only 4.5x slower is nothing to sneeze at. I doubt that WASM is optimal for interpretation and you could probably gain another 2x through a bytecode c
by ridiculous_fish 7y ago
Nice find, only 4.5x slower is nothing to sneeze at. I doubt that WASM is optimal for interpretation and you could probably gain another 2x through a bytecode compilation target optimized for interpreting.
- vshymanskyy 7y agoActually, it's not interpreted directly. Wasm is converted to a very interpreter-friendly representation.