5 ms·

The thing is, you use phrases like: > may not have been possible > incredibly challenging > a total guess Which, yeah, rocket science is hard! But the abov

The thing is, you use phrases like: > may not have been possible > incredibly challenging > a total guess Which, yeah, rocket science is hard! But the above is not difficult because it's impossible to solve for, it's difficult because we never tried! Nobody, government or otherwise, thought that rapidly iterating on rocket technology would be a good idea. This above all was the ticket to SpaceX's success - when you can throw rockets away for science, you can learn a lot in a short amount of time. Did some things need to be invented before SpaceX could use them? Obviously, although I have doubts that the computations being performed by these "GPU solvers" really require a modern custom-built GPU and couldn't have been done on FPGAs or arrays of microcontrollers from the 80s. If we understand the math being crunched, it's just a matter of throwing more computing power at the problem.

hwillis 5y ago
To put those phrases in context, I think they would have required more effort than the Apollo project. > Obviously, although I have doubts that the computations being performed by these "GPU solvers" really require a modern custom-built GPU and couldn't have been done on FPGAs or arrays of microcontrollers from the 80s. Here's a talk by the guys: https://www.youtube.com/watch?v=vYA0f6R5KAI https://www.youtube.com/watch?v=vYA0f6R5KAI The Cray 2 could do <2 gigaflops. So a single simulation with 10,000 points per side at 1 microsecond timesteps for .1 seconds with ~50 flops per cell-step would have taken 79 years to run, assuming the Cray had 50 cycle fetch/store times. Engineers need to run lots of simulations to come up with a working design. And that's just simulating something around the scale of a single nozzle- you'll need to be doing sims for a 40 meter tall stage as it flips around. Also, the cray had a 256 Mword memory. So you could only store a single timestep of a grid of 400x400x400 cells anyway. The supercomputers of the 80s were really just not very useful for things like this.
- nexuist 5y ago
  Going to watch the video now. Thank you for the explanation! Graphics isn't my primary field so admittedly I don't know what I'm talking about. One thing I love about SpaceX is how engineers are allowed to do talks like this, I barely hear anything from the engineering division at BO or ULA.

hwillis 5y ago

To put those phrases in context, I think they would have required more effort than the Apollo project. > Obviously, although I have doubts that the computations being performed by these "GPU solvers" really require a modern custom-built GPU and couldn't have been done on FPGAs or arrays of microcontrollers from the 80s. Here's a talk by the guys: https://www.youtube.com/watch?v=vYA0f6R5KAI https://www.youtube.com/watch?v=vYA0f6R5KAI The Cray 2 could do <2 gigaflops. So a single simulation with 10,000 points per side at 1 microsecond timesteps for .1 seconds with ~50 flops per cell-step would have taken 79 years to run, assuming the Cray had 50 cycle fetch/store times. Engineers need to run lots of simulations to come up with a working design. And that's just simulating something around the scale of a single nozzle- you'll need to be doing sims for a 40 meter tall stage as it flips around. Also, the cray had a 256 Mword memory. So you could only store a single timestep of a grid of 400x400x400 cells anyway. The supercomputers of the 80s were really just not very useful for things like this.

nexuist 5y ago
Going to watch the video now. Thank you for the explanation! Graphics isn't my primary field so admittedly I don't know what I'm talking about. One thing I love about SpaceX is how engineers are allowed to do talks like this, I barely hear anything from the engineering division at BO or ULA.