LHACK4142

3/10 though I may be associating you with @Souptime

"goal is to lower the minimum requirements over the course of EA." -Dakota WOOOOOOOOOOOO!!! HOPE FOR MY 1650!!!

I'VE BEEN ASSIGNED TO PROGRAM A SWERVE DRIVE FOR OUR FRC ROBOT AND I DON'T KNOW WHAT I'M DOING AND I'VE GOT TO DO CALCULUS AND CS AND EVERYTHING IS ON FIRE AND IM BEHIND ON EVERYTHING AND ITS TAKING TOO LONG AND HELP HELP BUT IT'S COOL AAA WHAT DO I DO HOLY [REDACTED] [REDACTED] AAAAAAAAAAAAAAAAAAAAAA sorry had to get it out

When you begin to notice how frequent the word "swerve" is because you keep thinking about swerve drives.

WOOHOO!!! I'M ON TIME!!!

One of my teammates in my FTC robotics team suggested using AI to drive the robot instead of painstakingly pre-programming instructions. I thought this was kind of far-fetched, but as I researched the idea, it seemed more and more plausible, so we're going to give it a go! If we can get it working, it could honestly be a bit of a game-changer; basically nobody else is doing it, and it could save so much time and accelerate development dramatically. I want to document the process so that we can write about it in our essays and stuff, and also so that we can maybe make a tutorial, so I'll be posting whatever stuff I found here. So far, I've decided that I want to use a reinforcement learning model because they use agents in environments, which is easily adapted to a robot in a playing field. I'm not sure Q-learning would work because from what I gather, it fills out a table of all possible states and decides the best thing to do based on that table, and we'd have around 2^140 states. Therefore, I think I'll use deep learning. Since I need to use Java for FTC, I think it's time to start learning deeplearning4j!

My robotics team named our robot Eugene Grabs after Mr. Krabs from SpongeBob. I've never watched that show, so I figured I should read up on the lore behind our robot's name. Well, it turns out that my entire personality is literally just a slightly toned-down Squidward. The resemblance is actually uncanny- rather glum, overly factual, always tired of funny/annoying things that others do...

n45

I really second this. Of all the things I've seen in KSP2 footage, this has somehow been one of the things that bugged me the most.

"Opinions among writers and editors differ on whether to use the serial comma..." -Wikipedia yes, I'm being annoying... sorry

As a fellow extreme-KSP-1-bug-experiencer-and-hater who hasn't gotten the time to play KSP2, is it in fact better than KSP1 in terms of bugs, @Bej Kerman? (Assuming you've tried KSP2)

Necropost, but why couldn't the robot feel anything? It was mecanumb.

Come to the Android side...

I'm excited for the polish, new parts, good stock graphics, and cool explosions!

Been listening to this a lot after my robotics team got eliminated lol. We've experiments to run, there is research to be done, for the seasons that will soon arrive.

Looking at gameplay, it seems like a good part of the loading time is spent "pumping the sim once". I wonder what could this mean? @K^2

I had a really interesting one: I had broken my cello bow, so I somehow manifested the KSP revert menu into real life and reverted back to Thursday. After doing so, however, I immediately regretted my decision because I had created a new timeline, which really didn't feel like my main, real one. I really wanted to go back to my old timeline where I had broken the bow, because being in this timeline I had created felt so fake, contrived, and superficial. However, I couldn't go back, having forgotten to save the original timeline. Very, very weird dream, and it makes me feel weird about time travel.

As one of our in-house teenagers, I have to say that I find the VA rather annoying, and the odds are that I'd not use the tutorials and therefore have a higher chance of not playing KSP2 if I were new. I'd also like to say that IMO tutorials have a massive impact on how many people go on to become active players of a game. The only reason I don't play DCS is that the tutorials are complete garbage. You need to fly each mission perfectly, and if you don't, you need to wait through 10 minutes of some guy saying the same thing inaudibly. Similarly, if people boot up the game and have to listen to those tutorials, I feel like a good amount of them will drop the game. I'd also like to add that I always hated these types of voices as a kid, perhaps even more than I do now. They always struck me as ridiculously contrived, and not at all factual or weighty. That said, I can tolerate it, and it's decent.

Mods, can you move this thread to the KSP2 Discussion subforum?? /j

I'm 1000% with Bej here. Every one of their complaints about KSP1, I feel it. I wanted to make some cool landing legs for a reusable rocket. My plan was to have a beam on a hinge, then a piston connected to the rocket and the beam to actuate the leg. Basically the simplest design possible. Guess what happened when I tried it design out? First, the legs started flopping all over the place. Next, when I tried to retract them, they somehow started spinning on the hinge at ridiculous speeds, leading to an inevitable RUD. The last mission I tried was a Duna return mission. I had a rover from the mod Planetside, and I had landed my ship far from it. I had driven that rover a long ways, so I figured I'd be fine. For some reason, though, the rover just wouldn't respond to any input at all. I tried everything, but nothing worked. KSP1's main gameplay loop is to build a thing, test it, fix it, then repeat. This becomes really tiring if there's a 3-minute loading screen between each 3-minute step. When I run KSP, it inevitably crashes a few hours in. It's actually corrupted my user registry and forced me to re-install my entire operating system. It's just unbearable. I hope KSP2 does better, but given these requirements and the footage we've seen, it doesn't look too promising. Maybe I'll have to try SimpleRockets2 again.

I think that visual languages do help in making things easier. system.out.println("Hello World!") will not run in Java, and it's really hard to tell why ("system" has to be capitalised, and a semicolon is missing). In fact, I nearly didn't learn coding at all because I kept running into that type of error. Visual languages, on the other hand, make everything like that much clearer. I do really hope that there's some kind of scripting ability in KSP2. Programming actual physical things is so much fun, much more fun than programming some application IMO, and it'd be awesome to bring that experience to people who may not have done so.

I personally like the noodle rockets. It can be kind of overt at times, but I think it's really cool to see your massive rocket or station bend and (imaginarily) creak. I tried SimpleRockets, and found the ultra-rigid-stick rockets kind of boring to watch.

It would be nice if I could run it, and it would be nice if it handles really complicated ships better than KSP1. I haven't watched all the preview videos and stuff, but it looks like neither of those will be the case. However, I'm still excited to try it out on my computer, and see how it evolves.

FTC (First Robotics Challenge, a robotics competition) Inter-league Competition. Today was one hell of a day- the culmination of like 5 months or something of hard work. We got sixth place out of twenty, and ended up not advancing to the next level. We're all pretty disappointed, but I personally feel like we honestly did really well. I mean, we're a fricken massive team (30 members, the usual is like 5), and we're ALL ROOKIES. Our team is very, very open to everybody regardless of how enthusiastic they are about robotics, or how skilled they are. This is of course great because we recruit lots of new people to FTC and create a really fun environment, but it's also bad because, well, we have a massive crowd of extremely chaotic kids who don't know anything, mess around 90% of the time, and may not be particularly dedicated. Contrast that to some other elite teams (who we had to go up against), that basically only recruit the best and most commited students, and I think what we accomplished is pretty damn decent. At the beginning of the day, we all gathered up, got our robot inspected, and went for our interview. The way FTC works is that you can advance to the next round of competitions through both the actual performance of your robot and awards given for best robot design. This is so that students learn to write about their robot, get good at interviewing, and also in case some team gets horrendously bad luck or something. Anyhow, the interview went OK, and we gave the judges our portfolio to read over. Next were the actual matches. We did pretty well, winning four out of five matches. We also had to do a frantic fix in the middle of the competition because our programs would actually give us penalties. After the matches, we were in sixth place, and qualified for the semifinals (which happened today also). As luck would have it, we were up against the number one team in the semifinals. They of course vaporised us. Now our only hope was to win with awards. The chance of doing this was pretty slim, given that our portfolio and presentation were made very last-minute. We didn't win enough awards to advance, but we got a few cool ones, most notably (for me, the head programmer, at least) 3rd place for the programming award. So yeah. Overall I think this season was a triumph, if not quite a HUGE SUCCESS, and I'm tired.