i nerd sniped myself tonight and i imagine the NSA operative who is assigned to me is very confused by the increasingly erratic and frustrated google searches for HOW DO I CALCULATE THE MOON WHERE IS THE MOON

[aeva@mastodon.gamedev.place]

i nerd sniped myself tonight and i imagine the NSA operative who is assigned to me is very confused by the increasingly erratic and frustrated google searches for HOW DO I CALCULATE THE MOON WHERE IS THE MOON

[aeva@mastodon.gamedev.place]

I just want to calculate the current direction of tidal force from the observer's earth-relative position in space and time is that so wrong

[aeva@mastodon.gamedev.place]

wikipedia: even the ancient babylonians knew how to use math to figure out what the moon is up to

me: great how do i do that

wikipedia: 😏

[jon_valdes@mastodon.gamedev.place]

@aeva oh boy. I went down the rabbit hole of calculating star positions a few years ago, and the whole thing is... convoluted. Step one was calculating the Earth Rotation Angle, and that was... urgg

[aeva@mastodon.gamedev.place]

@jon_valdes my hope was that the moon's position relative to the earth would be the one situation where geocentrism still holds enough water for a useful approximation, but I'm starting to think that this is a problem space where people don't typically seek useful approximations

[jon_valdes@mastodon.gamedev.place]

@aeva what I found is most methods for calculating these things are approximations that break down if you deviate from the year 2000 by more than 5 millenia or something. It just depends on what you consider a "useful" approximation 😅

[aeva@mastodon.gamedev.place]

@jon_valdes oh that's great. I figure for my use case it only has to be moderately accurate for the remainder of my lifespan however short or long that may be, and it's not really my problem if it accumulates a bit of drift after that

[jon_valdes@mastodon.gamedev.place]

@aeva perhaps the easiest way would be to look at how other programs do it. I spent a bunch of time reading code for Stellarium, for example. Perhaps that, or a website like this could help you?
https://eyes.nasa.gov/apps/orrery/#/moon?time=2025-09-10T04:57:12.041+00:00

[aeva@mastodon.gamedev.place]

@jon_valdes i found this a few hours ago http://www.geoastro.de/elevazmoon/basics/index.htm haven't had a chance to look at it in detail but it looked promising at first skim

[dotstdy@mastodon.social]

@aeva @jon_valdes I'm trying to remember what program we use to generate location accurate star charts in Avatar and Division... Maybe it can do the moon as well, will look it up later.

[dotstdy@mastodon.social]

@aeva @jon_valdes alright I think we just use data from sdss and reproject it for alpha centuri, blindly extrapolating for the correct date using the star velocity. so no complex orbital mechanics. However for the planetary bodies in solar system we use code derived from https://stjarnhimlen.se/comp/ppcomp.html

[jon_valdes@mastodon.gamedev.place]

@dotstdy @aeva oh, nice!

Word of caution about calculating these things on the GPU: implementations of trigonometric functions on the GPU are less accurate than you'd hope, and if you calculate a bunch of sin() and cos() for your view direction per pixel, they turn out not to be accurate enough for stable movement across a 60° fov on a 4K screen. Particularly Intel, which seems to barely pass the D3D precision requirements there...

[dotstdy@mastodon.social]

@jon_valdes @aeva right, yeah I should also mention all that is done on the cpu, and with double precision. I'm not really sure if you can even squeeze it to work with f32.

[jon_valdes@mastodon.gamedev.place]

@dotstdy @aeva at least for star positioning f32 was barely enough for me... as long as I rolled my own trig functions

[aeva@mastodon.gamedev.place]

@jon_valdes @dotstdy blessedly I'm planning on doing this on the cpu, but D:

[jon_valdes@mastodon.gamedev.place]

@aeva @dotstdy if you want to see this in action, open this shadertoy on a few different GPUs and compare the output on each. Particularly fun if you have a pre-Arc Intel GPU...

https://www.shadertoy.com/view/dtlyD8

[aeva@mastodon.gamedev.place]

@jon_valdes @dotstdy my laptop has a recent-ish Xe gpu

[aeva@mastodon.gamedev.place]

@jon_valdes @dotstdy it's bizarre seeing my 10 year old telephone's mali do better at... anything at all

[aeva@mastodon.gamedev.place]

@jon_valdes @dotstdy though I guess the corner the intel case is cutting is where the slope starts to level out so I guess that makes sense. are CPU trig functions just as bad?

[aeva@mastodon.gamedev.place]

@jon_valdes @dotstdy would mollytime get a warmer, fuller sound if i switched to expensive software sine waves :3