Anyone know if there is any method to import an accurate milky way starfield into blender? I would think as some sort of point cloud or collection of vertices maybe? any leads/places I could check for such a thing?
Thanks,
-Trev
Do you mean a picture of it ?
Go at NASA site or Hubble site you’ll find lots of pictures to put in the background
Salutations
Well, you could try faking it by using halos or something. Each halo would represent a star, and if you make them different sizes and colors you should get a good effect.
the point cloud model is available here: http://astro.uchicago.edu/cosmus/projects/milkyway/
but i am not sure how to import this into blender.
There’s an amazing particle build that appeared on the news & discussion forum on the 9th . I played around with the settings and made a milky way by mistake a few days ago. Hope this helps.
Ummmm, I fear that this would not work very well.
They are using the rather nice HYG dataset of stars. It is nice, but alas it only includes stars within 50 parsecs (160 light years). Unfortunately our galaxy is 30,000 parsecs in diameter (100,000 light years).
Put it this way: if our galaxy was the size of a 12 inch dinner plate, the area covered by the HYG dataset would be about the size of a grain of salt.
Sorry to get all technical but I’ve sort of studied this.
ok nyrath, you are right! I have no idea what the webpage is about and I just googled for something that MIGHT be what trevj was looking for.
I don’t think it’s possible to get a scientifically accurate milky way, because so much of it is obscured from us. We can only get a certain amount of information about the centre, because of all the dust and the density of stars there. Anything on the other side of the centre is a complete mystery. Scientists have worked out that it’s a barred spiral, and how tightly wound the arms are, but that’s about it.
For our planetarium show, I’ve made a few spiral galaxies and I’m using one of them as the milky way. It took me a long time to do it, and I used a lot of python scripting.
The central bulge is basically an elliptical galaxy. I wrote a script using a monte carlo type method to generate a point cloud in an elliptical shape. I made a few of these so I can control the colour and brightness of the centre. They are all old, yellowish stars in there.
I used the same script with much fewer particles to make a cloud of globular clusters that surrounds most of the galaxy. They’re yellow as well.
The bright blue areas are star forming regions. I grabbed a photo of a spiral galaxy, selected just the blueish areas using a colour range selection in photoshop, indexed the colours of it, and textured it on a plane. Using python, I took random samples of pixels from that image and the size of the plane, and created points that have the right position (plus a randomised “height” from the plane that I fudged certain ways to give it thickness) and a material with the colour sampled from that photo.
I did the same thing for the dust lanes, but just made them one object with one material.
Layered in there as well are some randomish clouds of blue and red stuff to give it more volume. I used the nebula script I got off the forums somewhere for that (I have to talk to that guy about using his script in a commercial project!).
All of these things are on separate layers and put together with compositing nodes. The only exception is the dust, which is on the same layer as the brightest part of the central bulge so that it occludes the centre nicely when you look at the galaxy from side on.
I’m not happy with the barred shape of it. I used lattices to try and change the shape to what I wanted, but it hasn’t worked as well as I wanted.
Hope this all helps!
Attachments
lisae,
That’s very likely the best CG galaxy I’ve ever seen! I especially like the globulars and the dust. How well does it hold up when viewed from other angles?
-waystar
Anybody ever wonder how scientists made that “you are here” photo of the Milky Way, when we’ve never been outside of our own solar system?
Our Galaxis has rougly 200E6 Stars
so i would forget trying to make an exact 3D model for that
unless you have a cray computer with 100 GB of memory and a few months for rendering
Sometimes thing cannot be done so scientifically accurate - you also have to take into consideration the PC’s power - Memory and time to render
Salutations
Hm, I would be very interested in other views or even an animation as well (you mentioned it was done for a planetarium show). I’m not sure if you can us show your work but if so, it would be great 
By the way, I’ve got to agree with lisae’s points. What do you want when you say a “scientifically accurate” Milky Way? Even if you had more than the rather “simple” pointcloud which was posted before (and which is very awesome though, in my eyes), what would you do with it (see also the reply above…).
So, something like the method above would be the way to go, I think…
most likely drawings
It’s true that we’ve no way to know exactly how our galaxy looks from the outside. But from what can be observed and from observations of other galaxies that share the same characteristics, a pretty good idea of how it looks can be formed.
-waystar
i think the original poster is looking for a ‘visually correct’ milky way.
lisae, is it possible for you share your method and maybe post a bigger version of the picture? the galaxy looks wonderful.
i thought the little grays took some poloroids for us on their way in…?
In any case what we called the milky way is only a small white band in the sky
and to see this you truly need a dark sky far away from large cities light on a very clear night
or if you can fly at 30000 to 40000 Feet at night you might see it too
but i think if you on NASA or Hubble you should be able to find something
Sa.lutations
Sorry I took so long getting back to this thread! I’d like to post an animation of the galaxy when I have the chance. At the moment I have an animation of a quasar evolving into this galaxy, and I am working on a version now where we zoom out on it from the level of the solar system to the galactic level (and later zoom back in again). This is really challenging, but I think I’ll get there. But I don’t have an animation where we just fly around it. I’ll do one if I get some time this week.
I also used a few different versions of this galaxy and another one to create a whole bunch of random galaxy images to billboard onto planes so that we can have a good background for the scenes that are at the galactic level. These are also used in my animation of the big bang, which was very challenging too! But I really can’t share any of this stuff here because it’ll be a commercial show. I will definitely share the trailer when we get to that stage, though.
Sounds fantastic, Lisa!
It will be a few years before we have a full dome system here (we’re in the planning stages now). We may or may not be able to afford to purchase commercial shows. If we can, you’d better believe I’ll vote for a show that involved Blender at some stage in the workflow.
Hopefully I’ll get to see a demo at a planetarium conference sometime. That would be extremely cool.
-waystar
Most of our galaxy is invisible to us, but clouds of hydrogen moving at a large enough velocity relative to us can be mapped by radio waves. Presumably the stars are associated with the clouds. The velocity constraint means there are two triangular shaped blind spots, with their apex on the Sun, one aimed at the galactic core, the other in the exact opposite direction.
The best recent approximation map is here:
http://www.news.wisc.edu/newsphotos/milkyway.html