4 ms·
> Shouldn't OceanGate have recorded the exact coordinates where it was released and then been able to localize its possible location to a small area around tha
by eigenspace 3y ago
> Shouldn't OceanGate have recorded the exact coordinates where it was released and then been able to localize its possible location to a small area around that by modeling ocean currents or something?
Modelling and knowing surface currents is one thing, but this submersible was thousands of meters deep. Deep ocean currents can be very fast, change often, and we have way less data on them.
- anigbrowl 3y ago'Telemetry? who needs it' - Oceangate, probably
- BurningFrog 3y agoThere is no way to transmit telemetry to the surface anyway.
- bg46z 3y agoTelemetry doesn’t work very well under water. There’s work to be done on telemetry that is capable underwater, but it seems a long ways off.
- nameless912 3y agoNot to be crass, but wouldn't any low bandwidth sound-based protocol be easy to rig up for underwater? Obviously there's a bit of adaptation required, but it shouldn't be _that_ hard to set up a 28.8k modem with a directional speaker and microphone rigged up to point vaguely toward each other (i.e. at the surface pointing down and on the sub pointing up). Hell, even automated Morse code sending literal text messages at 3 sps is better than nothing.
- anigbrowl 3y agoWhy are you assuming it's wireless? ROVs use data tethers with the spool kept on the base ship. Or you could have a black box recorder attached to the sub by a cable that blows its connection and floats to the surface if it loses contact for more than a few minutes. I can think of 10 different ways to maintain some contact with the surface and store some data about dive conditions. Honestly perplexed at all the people who are still buying into Oceangate's demonstrably bad way of doing things and just saying it's impossible to do any better.
- gregoriol 3y agoWhen communication has been lost, what kind of telemetry would you expect?