Видео

there are links to blueprints in the description on this video: ruclips.net/video/p9S68L-ucIg/видео.html

Nice! The py discord is pretty active and friendly if you want more interaction. PySEx is also being developed!

derp. Good eye!

haha i have a notepad with the todos

It's a planned speedrun

imagine this. you make something because you like it, not because you're trying to get famous.

For sure! Yeah transport drones is a fun mod!

check out episode 0

@@paxtorio1495 In this ep, you show the mods. I had requested a list, when you show the mods it is not very detailed. It goes up and down and takes several minutes to show them all.

Yes, I spend a lot of time in Helmod planning. I basically have a plan for 6 complex circuits per min, but thats just to make sure that I have the minimum. When I actually build a particular ingredient I usually build more, depending on what is a good ratio or based on a throughput limitation -- hopefully designs that could be copied if they need to be scaled up later. I'm also getting help from @simonplays314, he has some great videos: ruclips.net/video/L-nLlr_7CPM/видео.html&ab_channel=SimonPlays

@@paxtorio1495 Thanks for the link, that style of design and video editing was super interesting. I think I might try PyShort now that I've finished exotic industries. I've stalled out on at least 7 PyProper attempts.

check out episode 0

Thanks! Sexy base design is what this channel is all about :-D

deadlock's stacking beltboxes and loaders

Thanks! Maybe Ill do a few recordings then of the planning process.

@@paxtorio1495 awesome sauce

lol yeah that was silly! I was so confused haha.

thanks! yeah py isn't for everyone.

heart breaking tragedy

sushi is OP!

what if i told you that its like having an 80 lane bus?

hey, thanks! Yes, train throughput can be an issue. But I've been developing this train system for a long time and I'm not worried about train throughput. Having the hexagonal grid and three-way junctions allows for a lot of scalability with the trains. Py is also in general very big and relatively low through-put, and LTN can be pretty smart which also mitigates the train issues compared to vanilla. I also have the 4-lane variation of the 3-way junction if I ever run into issues, but I think this design can scale easily to 500 trains. Py also has some higher tier trains, so those might come in handy too.

i added a link to blueprints in the py science 2 episode: ruclips.net/video/p9S68L-ucIg/видео.html&ab_channel=Paxtorio

Carter does repeatedly mention how much math goes into the dialing...

haha yeah, getting it on the third try was sooo lucky! I had messed up on that recording a few times, so that was like the third recording, and I was like "can't mess up now!" Although it didn't really save much time getting lucky, as like you, still had to do all the automation, and then the queen will just sit there for a while before I scale it up. I almost was too slow in scaling the arqads, might have even been better to have a few more modules built...

No! I didn't even realize that as a possibility. Is that what you did?

@@paxtorio1495 I didn't have to. I was setting up arqads before the mod update removed the empty honeycomb from a bunch of random easy stuff. So when I updated the mod I already had enough combs to start the loop.

Thanks! Indeed those mods help so much! Definitely a lot of the mods I use are practically cheating compared to your playthrough, but "My speedrun, My rules" ;-). I generally play ahead about 10-15 hours and develop plans for 2-3 episodes, probably like 7 hours of planning for 1 hour of recording. Very often I have to do the recorded playthrough 2 or 3 times. Many times I start the recording and then realized the base isn't ready with the resources for the planned build because I get there so much faster, so I have to reroute.

Also a bit of time watching your videos!

good eye, haha! well it gets fixed in the next episode, but I didn't even realize the mistake.

oh wow so you tried to account for the spherical shape? I just assumed it was because of that. Hmm... maybe its related to the actual center of the triangle, as they are isosceles. I approximated the distance from the center to the top as twice the distance from the center to the bottom, but I dont think thats exactly correct. Let me know if you ever figure it out!

@@paxtorio1495 I figured out where my error term was coming from. For the math to work perfectly, it is important that you first normalize the corners of the primary triangle (as you did). For every subsequent triangle, it is important that you do _not_ normalize again. It will fail if you get either one wrong. The fact that normalizing subsequent triangle will introduce errors means that measuring any of the subsequent triangles (which I also did) is challenging. After all, the game computes the answer by iterating 7 layers down flat triangles and _then_ normalizing. You need to project that answer back onto your original triangle to figure out a coordinate you can use. Assuming you normalized at each level (I can't follow all your code), then your assumption that the spherical shape was at fault is basically correct. Subdividing a triangle into 64 sub-triangles gives you a different result if you project the corners onto the sphere first. If you do every level of subdivision on the same flat triangle and only normalize at the end, it works. :) Your "center to top as twice the distance from center to bottom" math is flawless. For an isosceles triangle, that will always be true.

@@kpsmith67 awesome! good to know.

You're welcome! You can guess/brute force the puzzle, once you generally get the idea of how it works. The starmapping only gives you clues about the pentakis dodecahedron, which is only the first coordinate. The second clue in the temples will help you put together the triangle face map, which is needed for the 2-8 coordinates. You can figure out the triangle face map from measuring with the gate. I went to about 30 of 60 temples. Getting some prod 9 modules is kinda worth it too.

ah he realised quite quickly haha