Roy wrote:

Wanna see J1407b

That would be insane, also I would like to see Uranus and Neptune's 

Khorrah wrote:

also I would like to see Uranus and Neptune's

You lol, they are already there.

SpaceEngineer wrote:

Source of the post Bunch of a new screenshots.

What and outstanding job Vladimir. This code is a piece of art. There are many things that might be of interest here that I don't know where to start. Kudos to you and the entire SE team.

Rings shader doesn't uses doubles, it uses emulation through two floats.

Roy wrote:

Khorrah wrote:

also I would like to see Uranus and Neptune's

You lol, they are already there.

I meant their rings xD

Khorrah wrote:

Roy wrote:

Khorrah wrote:

also I would like to see Uranus and Neptune's

You lol, they are already there.

I meant their rings xD

Once again I tell ya "U LOL". Because:
viewtopic.php?t=536&start=135#p35942

Roy wrote:

Khorrah wrote:

Roy wrote:

You lol, they are already there.

I meant their rings xD

Once again I tell ya "U LOL". Because:
viewtopic.php?t=536&start=135#p35942

I DIDN't SAW THEM SORRY LOL they look beautiful ♥ 

FastFourierTransform wrote:

SpaceEngineer wrote:

Source of the post Bunch of a new screenshots.

What and outstanding job Vladimir. This code is a piece of art. There are many things that might be of interest here that I don't know where to start. Kudos to you and the entire SE team.

This is a dream come true!  I wonder if it would be possible to ring "jump" from rock/particle to rock/particle?  That sounds like a fun game within the game!  I cant wait to see what the highly complex rings of Saturn look like with the shepherding moons and ringlets crossing each other and all that.I remember seeing the photographs from Cassini and this looks even better than that especially since it is immersive!

SpaceEngineer wrote:

Source of the post Bunch of a new screenshots.

wow!!!

SpaceEngineer wrote:

Rings shader doesn't uses doubles, it uses emulation through two floats.

Ah got it.  So doubling up on FP32s and not native FP64.
Does it take twice a long to compute like this compared to normal single FP32 operations?

Yes, performance is almost halved by using emulated FP64.

Sounds like another good reason to keep the rings as thin as possible  (10-20m rather than 100s-1000s of m)

SpaceEngineer wrote:

Yes, performance is almost halved by using emulated FP64.

Still a lot faster than native FP64 since that is 1:64 rate on nvidia and a still slow 1:16 rate on AMD. 

evildrganymede wrote:

Sounds like another good reason to keep the rings as thin as possible  (10-20m rather than 100s-1000s of m)

No, thin rings are the problem, because then you have to make the particles smaller, which causes the aforementioned precision issues and requires fading out the particle rendering at a closer distance to avoid a massive performance hit (since there are way more particles).

could something like fog be used to simulate the effect of smaller particles? we don't really need to model every particle down to the centimetre scale (and smaller) here!