Rocket engine Tyche test day yesterday. We did all procedures necessary to fire (without firing) to work out the kinks in the process. Since real propellants are expensive, we substituted the combustible liquid oxygen and kerosene with relatively inert liquid nitrogen and water.
In the first picture I'm adjusting the valve actuator pressure (Muscle Pressure Manual Valve/"Muscle Man ๐ช"). The second picture is a head-on of our engine's injector plate. You'll notice some holes are bigger than others, that's because we did our flow math wrong (note to selves: d=2r, not 1r) and had to redrill to achieve the intended total orifice area without drastically increasing the number of holes (and as a result, adding potential failure points). Live and learn.
@SawyerMerritt Magnesium, huh. It could be that they're manufacturing this themselves and using it as a testbed for magnesium casting in prep for Optimus. Just a thought.
n=3 but this is starting to paint a picture
in all of the pillar/pole collisions we've seen the following is always true:
1. empty context / start of drive (aka fsd doesn't have memory of objects/pillars/pole entering a blindspot)
2. pillar/pole is in a camera blindspot
in all of them it appears that the objects do exit blindspots, at least momentarily, within seconds of starting, so in theory this a solvable problem without additional memory/compaction methods for these specific cases
surely, though, there are cases though where objects will not exit a blind spot and better memory mechanisms are going to be needed as long as blindspots exist
in the meantime, i'm happy to use smart summon when not parked around any pillars
People have been hating on cars so much that they apply the same opposition to robotaxis, even though robotaxis solve a lot of the problems that transit also solves:
-no need to drive or own a car
-cheap
-no need for parking
-safe
For all but dense city cores itโs perfect.