My observations have been that image generation is especially challenged when asked to do things that are unusual. The fewer instances of something happening it has to train on, the worse it tends to be. Watch repair done in water fits that well - is there a single image on the internet of someone repairing a watch that is partially submerged in water? It also tends to be bad at reflections and consistency of two objects that should be the same.
I'm currently testing 4.7 with some reverse engineering stuff/Ghidra scripting and it hasn't refused anything so far, but I'm also doing it on a 20 year old video game, so maybe it doesn't think that's problematic.
I watched a starship launch live, in-person, and had the experience of driving up to the launch complex the night before and car camping right outside of it and looking out my car window in the middle of the night and seeing a massive rocket lit up with spotlights. It was the most "I live in the future" experience of my entire life. I can't wait to go back and see a chopsticks catch live.
If the physics were accurate enough, I don't think it'd be easy - you'd get constant elliptical orbits in most cases, right? making the timing much harder going forward
Technologies: Full stack web dev for over a decade, in recent years been focused on hobby game development projects using Babylon JS/WebGL and working with C++ and Go server-side (for multiplayer stuff). Many years of project and product management as well.
I have some cool projects! See my personal site for email and some of those projects:
Coolest recent one is probably https://www.idlequest.net/, which is an idle-game web-based version of classic EverQuest, complete with its original 3D graphics in-browser and a real-time MMO server!
Looking largely for product engineering positions, or project/product management style work. Have done both and enjoy both a lot.
I'm fairly ignorant on this but robots that are teleoperated seem completely capable of doing basically any household task and using tools like screw drivers. It may be slow, sure, but autonomy and speed seems like a solvable software problem.
There's also endless welding and assembly robots and have been for a long time. Sure they're huge and weight 3 tons or whatever, but it's not like we're building humanoid robots to do work like that anyway.
reply