@TheMattbat999 Yeah, most likely all of the above, and much more...

Gotta be grateful(?) when visiting these countries.

@AlexandreVaillancourt always need to be grateful for what ya got. Things could be worse.

Yep :)

So how was your day?

It was good, got a couple of bug fixes in, got the work laid out for the rest of the week, I'll be working from home tomorrow and vacation Tue/Fri

Good

I'll have to play programmer hero tomorrow though... that's fun and not fun at the same time...

How so?

Well when you know how to fix all of the tricky bugs, it's nice to feel needed, but at the same time, you feel some pressure because you need to get things fixed ASAP, and well, if you get hit by a bus, your team is in a bad situation. I don't consider this situation as super-professional behaviour, even if I can't do much about it.

@AlexandreVaillancourt ah.

@TheMattbat999 What about you?

@AlexandreVaillancourt good. Been busy with band practice, school, and goats so not as much time to program as I would like

School has started again already?

Atleast here. Most schools started a few days earlier or later.

Oh really? Here school starts later, like by the end of August..

Our goes from early August to mid May

Ah, that's it. Here it finishes just before June 24.

Oh. Just off-set of each other.

And now I am seeing if I fixed my Daylight cycle bug

Yep. I know why we stop by just before June 24, but I can't see why you guys go from August to May. Is it because just after mid-May is when the summer is the most bearable outside?

I guess.... not sure really. Just tradition?

That works too ;)

Summer, late Spring, and early Fall are really hot here tho

Oh, ok! It may also be that the kids were needed in the fields at some point in time (I guess it's still the case for some).

Reallt hot is like... 100°F, not including humidity, in which makes it feel like 115°F :|

Yeah, that's real hot!

Yep. Good night.

Good night!

@TheMaskedRebel I recall a while ago you were looking for methods to compute minimal bounding spheres. I was writing an answer about Welzl's algorithm today and coded up a C# implementation, in case it's of interest.

Kind of hoping the author of this question drops by the chat. I'd love to geek out about systems for structuring planet-shaped environments. :D

@DMGregory Yeah, there are some interesting questions out there, but not for the site

@DMGregory I dont think they are actual giant spheres right ?

you just load a chunk of flat terrain

and loop trough them ?

I mean you can calculate what chunk you should land on based on the point of a sphere that you hit

Typically it wouldn't be all one mesh, you're right. Though it could be if you were making a "Little Prince" / Mario Galaxy game with tiny planets. ;)

well if the planets are tiny then yes :D

I mean we know from bugs in No mans sky that its just a skybox with normal terrain chunks like in any game. The sun was not a real object ;(

But there is a question of what space those chunks occupy. Are they fitted segments of a sphere, or tiles of a cube that's been inflated, etc.

Nothing in games is a "real" object. ;) As long as you get the pixels where they need to be, the abstraction is doing its job.

well if its for games... The game just needs to fool the player he is on a spherical planet right? I doubt that you would need any curvature on the chunk itself.

no I mean that they said the sun you see in the sky of any planet is the actual game object sun of the solar system that planet is in

It depends if the player needs to be able to circumnavigate the planet, as we do in Starlink. Then it needs to wrap around somehow.

it was not. It was just a directional light with a sun disk

you can loop trough a 2d array of chunks and send him to the other side

right ?

Getting that to be topologically equivalent to a sphere is trickier than it may seem.

so how do you do it then ?

A torus topology is easy. Someone recently pitched a clever topology for a non-orientable 4-dimensional embedding but spheres are weird. :/

In Starlink, we do treat the gameplay space of the planet as literally spherical. So "down" points in a different cartesian direction depending on what part of the planet you're on.

huh.. that seems more complicated than doing the math to flatten a sphere

so you have curved planet chunks in a spherical space ? or the chunks themselves are not curved ?

It turns out to be complications pretty much no matter which way you go. It's just a matter of which complications are more palatable for what you want to do.

As our physics prof said once, "there's a sort of conservation of math"

this was with regard to Hamiltonian mechanics

sure, the equations are dead simple once you get the normal coordinates... but you gotta do a ton of work to get those.

I like that expression. :D That's exactly how it feels sometimes!

it's just that you get a usable expression when you'd get a fucking mess with Newtonian mechanics.

:D

For us, one of the biggest pinches was raycasts. Now a raycast that's "flat" in the player's conception of "along the surface" actually bends around the curved surface.

oh fun

Myself, I want to try working with a planet that's an icosahedral net. On the surface it's completely flat, with teleporation portal planes at the seams. Away from a seam, you can pretend there's nothing special happening at all. And you just need an extra check for seam crossings to break a raycast into a pre- and post-portal hop.

bending raycasts that must have been fun to do

Somehow that sounds like a smart thing to do

I think it sounds promising. But the devil's always in the details, so there might be even more painful trade-offs hiding in the parts I haven't implemented yet.

The really rough bit is physics bodies straddling a portal. You can arrange the world so this can happen only in a few places, but you still need to solve it somewhere. And it's harder to handle multiple players in one consistent world (eg. local COOP), since you can't re-arrange the seams to keep them away from both players simultaneously.

Versus working on a sphere embedded in regular cartesian space, where even if the physics are a bit complicated, they're complicated in the same way for everyone / everywhere. There's no viewpoint-dependent game logic or seams between terrain tiles to fix up the same way.

Those are clearly unsolvable and unintuitive problems which proves that the earth is flat.

@nwp Oh gosh, wouldn't that be great? XD So much would get easier if the world would just stop being a darn ball.

Someone clearly screwed up in the early design phases :P

On the plus side, it gives me like 300% more opportunities to cite the Hairy Ball Theorem, and that always makes my day. ;)

You log out? o.O

Oh, they fixed the consecutive days thing.

@Tyyppi_77 nooooooooo!!

It used to only count logging out for a day as non-consecutive. I got that 1 year consecutive online every year for a while even though I was offline for weeks sometimes.

Less than 2 hours left!

Dammit... i already have about 50 tables in my mysql workbench for all sort of stuff... its a real mess... and theres no way to group them

@genaray There is always a way to group them. But you'd duplicate data and make it messier ;)

I've written a little script that parses the gamedev.se posts, and well, we have a lot of link rot :P

And hello @ChrisMcFarland!

@AlexandreVaillancourt Yikes. Looking for common causes to add to the meta post?

@DMGregory Yeah, well there are a couple more file/image hosting sites to add. It includes files on wikipedia :P

I'll add them later tonight or later this week.

Ahh, I'd have thought those would be reasonably stable. Good catch!

But even if we get rid of all of those listed on that post, there will be a lot of links left that people have used the files directly from the original site, and sometimes it's down. If we'd like to make things clean we'd have to parse through all of them and make sure they're on stack's imgur. But that's "monk's work"..

Maybe what could be more valuable for now would be a kind of bot that checks all the new posts and see where the image is hosted...