Skybox Broken Near Galactic Center

[Cmdr ST4R F0X]

Has anyone else noticed this ? I've seen it both times ive headed out toward Sag A. At some point around 5000ly from the centre, the skybox starts getting weird, with enormous perfectly square sections of sky with no stars in them, right next to patches of sky with high star density. Also, as you approach Sag A even closer, the star density on the skybox drops to basically zero. All pictures ive seen from sag A just have the typical orangey hazey skybox, shouldnt the sky just be ablaze with stars ?

Either way here are pictures of the skybox bug i commonly experience, has anyone else seen this ?

CMDR Darklanky

 

[Leadfoot]

If those are two individual images, yes, I've seen that as well. Someone with more technical knowledge then I have will be along to explain the "why".

 

[AcademicDust894]

I was out by the core the other week. It was like your screenshots. Not good on the eye. It was never that bad on previous visits the other year.

 

[Dsign]

I reported this about 3 months ago. They said they'd look into it, maybe it'll be fixed in 2.4..

 

[Stupid_hippy]

Sadly,the galaxy has had this issue since the beginning of time.. or 3300 at least.

 

[killminster]

I had the same when I did my trip out to Sag A* over two years ago. Here.

I'm not normally one to cry about "mah immershun" but I was hoping for "OMG it's full of stars" rather than "that doesn't look right".

Kind of spoils the experience.

 

[CMDR Dreamstate]

Has anyone else noticed this ? I've seen it both times ive headed out toward Sag A. At some point around 5000ly from the centre, the skybox starts getting weird, with enormous perfectly square sections of sky with no stars in them, right next to patches of sky with high star density. Also, as you approach Sag A even closer, the star density on the skybox drops to basically zero. All pictures ive seen from sag A just have the typical orangey hazey skybox, shouldnt the sky just be ablaze with stars ?

Either way here are pictures of the skybox bug i commonly experience, has anyone else seen this ?

http://i.imgur.com/SaZRSFN.jpg

http://i.imgur.com/AUVd6tF.jpg

CMDR Darklanky

Must be some kind of limit to the amount of stars available to render - they really need to turn this into a graphical option in the menu so we can max it out.

 

[Darth Ender]

the stars in the galaxy are in many cases arranged by real star charts. This will do weird things like put stars in a straight line or create sharp edges where charts were stitched together. Some artifacts are created because not all the star charts were created the same way so their information is slightly incompatible.

I think it has more to do with that than any sort of limitation on star rendering. You see the results of these charts all over (ever been near Bernard's Loop?). These are not errors that can be fixed....they're a part of the stellar forge now.

 

[Zenith]

The galaxy in StellarForge is broken down into cubes of density for each star class.
Closer to the centre of the galaxy, all star classes get higher and higher density in a given cubic area. B stars being brighter show up disproportionately on the skybox. The cutoff you see is the change from one cube of high density to the next cube with lower density of the same star class.
Will Frontier do anything about it? All they can do is modify how the skybox is rendered. They can't go changing the galaxy stellar density or they'll mess up currently and previously discovered systems.

 

[tafkabb]

From another thread:

It's not really a bug, it's how the galaxy is procedurally generated. The cells overlap and in parts of the galaxy where the densities are similar it's not noticeable. It's only where the stars go from sparse to dense around the core do the anomalies show up.

edit: ninja'd by Zenith

 

[Riverside]

The galaxy in StellarForge is broken down into cubes of density for each star class.
Closer to the centre of the galaxy, all star classes get higher and higher density in a given cubic area. B stars being brighter show up disproportionately on the skybox. The cutoff you see is the change from one cube of high density to the next cube with lower density of the same star class.
Will Frontier do anything about it? All they can do is modify how the skybox is rendered. They can't go changing the galaxy stellar density or they'll mess up currently and previously discovered systems.

Good description.

Another glitch in stellar forge that explorers will come across is the narrow layer of non-scoopable stars around -30 on the Z axis (up/down). I think it was intended to be a small area around the start point but it covers the length & width of the galaxy.

The formula for the procedurally generated part of the galaxy was created once & while it can be tweaked to add & update real stars realistically we have to live with these quirks. It's still an impressive feat of coding to generate an entire galaxy of 400 billion stars, despite it's imperfections.

btw if you see a densely packed line of bright stars pointing towards Earth those are real stars inserted into the game, often as part of a nebula. O & B class stars are visible from thousands of Ly away so it can be hard to work out what you are looking at.

 

[NW3]

I noticed this over a year ago, on my way home from Beagle Point. It looks really wrong.

I wish FD would fix it.

 

[Quantis Trap]

It's not really a bug ... but it really is because it certainly an undesired feature.

I'm a programmer too and I know that given the time to spend (which is key), there's always a solution to any problem.

A germ of a thought which comes to mind for me is that Frontier may want to look into the possibility of an algorithm which will add smaller overlapping "sub-cells". These would be algorithmically added near the edges where cells with greatly differing star concentrations meet in order to defuse the seams. This would have the affect of adding new systems instead of removing any existing systems. In a old much simpler 2D since, it would be kind of like sub pixel rendering edge smoothing on a CRT or LCD. Not sure it's possible. It's just my first thought.

I hope Frontier gets a bit of time to figure this out in the future.

 

[Mengy]

I noticed this over a year ago, on my way home from Beagle Point. It looks really wrong.

I wish FD would fix it.

I doubt they will fix it, too many systems have been discovered and logged around there, and to "fix" it would mean actually rearranging star system locations. I just don't see Frontier going to that extreme now, it's too late. I do agree though it can look pretty silly when you see the separation of zones like that.

EDIT: Except for what Trap said there above me, Frontier could add new star systems to blend the zones better. That's a nice idea actually, but I don't think it would be easy to accomplish.

 

[NW3]

It's not clear if this is a deficiency in Stellar Forge or a rendering issue in ED.

There are limits to the total number of stars drawn, so it might just be an issue of needing to draw the closest stars first or something like that, rather than drawing cubic clumps of stars, one after another, and then hitting the drawing limit and stopping.

The question is, is that void actually present in the Stellar Forge database? If so, adding a few thousand stars in a gradient, to smooth the transition, is certainly possible, since we know that FD has made changes to the database previously.

 

[Riverside]

It's not clear if this is a deficiency in Stellar Forge or a rendering issue in ED.

There are limits to the total number of stars drawn, so it might just be an issue of needing to draw the closest stars first or something like that, rather than drawing cubic clumps of stars, one after another, and then hitting the drawing limit and stopping.

The question is, is that void actually present in the Stellar Forge database? If so, adding a few thousand stars in a gradient, to smooth the transition, is certainly possible, since we know that FD has made changes to the database previously.

My first trip to the core was around March-April 2015 & at that time the number of stars visible in the galmap was much greater (not the ball we have now), it was possible to see those cubes in galmap. I haven't been that way for a long time, the edges of the cubes may still be visible there if you are looking for them.

 

[CMDRvenerian]

I was there two weeks ago and had the same issue. I posted a bug report and Fdev said they were aware of the issue.

 

[Cmdr ST4R F0X]

Good response here thanks for this, glad it's not my installation being glitchy at least. But yeah it is pretty ugly. Has anyone else noticed the total lack of stars on the skybox when you are actually at Sag A ? Before i first went there i imagined the sky being ablaze with stars all super close together, is this not the case simply for technical reasons ? simply not being able to position and render that many objects ?

 

[Shinji Ikari]

the stars in the galaxy are in many cases arranged by real star charts. This will do weird things like put stars in a straight line or create sharp edges where charts were stitched together. Some artifacts are created because not all the star charts were created the same way so their information is slightly incompatible.

I think it has more to do with that than any sort of limitation on star rendering. You see the results of these charts all over (ever been near Bernard's Loop?). These are not errors that can be fixed....they're a part of the stellar forge now.

Most of the stars close to the core are proc-gen because in RL we can't see very well there because of the dust in the galaxy.

 

[Shinji Ikari]

I'd like to see some map screenshots that show those bright star cubes because I have never seen them even though I blocked all but B, O and unusual star systems when travelling in there.