Русский New site

Advanced search

[ New messages · Forum rules · Members ]
Page 25 of 69«1223242526276869»
Forum » SpaceEngine » Feedback and Suggestions » General suggestions (Post your suggestions here.)
General suggestions
DoctorOfSpaceDate: Friday, 05.09.2014, 23:55 | Message # 361
Galaxy Architect
Group: Global Moderators
Pirate
Messages: 3599
Status: Offline
Quote Watsisname ()
Perhaps introducing a black atmosphere effect could improve the appearance.


Or just wait for proper anti-aliasing to be implemented.

Quote Watsisname ()
And I'm not sure how a black atmosphere could be used to simulate gravitational redshift.


It couldn't be. The most you could do is adjust the atmosphere so you see a red color along the horizon and blue above. In principle this sounds like a good idea, but in practice its a terrible idea.

Here is a black hole surface with a custom atmosphere stuck on top


If I use a black atmosphere then you can't see anything. There is no distortion effect due to a shader edit.





Intel Core i7-5820K 4.2GHz 6-Core Processor
G.Skill Ripjaws V Series 32GB (4 x 8GB) DDR4-2400 Memory
EVGA GTX 980 Ti SC 6GB
 
WatsisnameDate: Saturday, 06.09.2014, 22:19 | Message # 362
Galaxy Architect
Group: Global Moderators
United States
Messages: 2610
Status: Offline
What I meant to say is I don't know how any kind of atmosphere could properly simulate gravitational redshift. It's just not the right approach even in theory. An atmosphere causes a glow effect by scattering of rays. This doesn't happen around black holes -- it's a vacuum. The color/intensity will also always be wrong because it depends only on mass of atmosphere through which the rays travels, and not the direction. For gravitational red/blue shift, it depends on difference in field strength between source and observer. These simply do not correlate.




 
SpaceEngineerDate: Saturday, 06.09.2014, 23:35 | Message # 363
Author of Space Engine
Group: Administrators
Russian Federation
Messages: 4796
Status: Offline
On the screenshots here you can see the "black atmosphere" effect. It is simply modification of the shader (I also removed black sphere because shader now can generate the black round without pixelization effect). I will replace it with the redshift/blueshift effect though.




 
FastFourierTransformDate: Sunday, 07.09.2014, 10:47 | Message # 364
Pioneer
Group: Local Moderators
Spain
Messages: 540
Status: Offline
Okey!, tree things that I want to put in consideration;

In my last message I was being kind of a devil's advocate. The truth is that I'm not sure about the atmosphere issue proposed by Salvo. I think it could be but I am not sure at all if this is what a black hole would look like.

My question is
Why the boundary of a black hole should be diffused?
I mean, it was something it could be because of redshift (to the infrared) in the boundary of the background light sources, but as Watsisname has said it could be a self canceling phenomenon:
Quote Watsisname ()
For an observer looking at a background object behind the hole, there is no shift at all because the two effects cancel each other.
.
I don't know, but surely this has also to be proven, They cancel exactly? (I mean that, from the star to the gravitational lense the light has been blueshifting (considerably in the last part of the voyage) and when it surpasses the vicinity of the black hole it starts to redshift until it arrives at the observer's eye, but since the observer is near the black hole dosen't the light would be less redshifted than blueshifted? they would cancel? In either cases I really don't understand why the soft gradient at the "event horizon" has to be implented (even if I one of the proponents wacko biggrin )).

PatrickPowers from PhysicsForums said this (not proven with real math):
Quote
If you were close to a black hole then light from distant stars would be blue shifted. If the light source is closer to the black hole than you are then the light you see will be red shifted.


If this is true, the effect should be quite awesome, and SE would be the only accurate depiction of it if it may exist.
I will explain the point: If we are far away from the black hole (say, from Earth) and the black hole is far away from the sources behind it then the doppler shift is cancelled out. This is simple. It's like an object in free fall to a point mass, first it accelerates gaining energy (getting blue shifted) and then when the point is surpassed it loses that energy until it stops (until it has no shift at all). But we can immagine a situation where the observer is near the black hole (lower in the gravity potential) so the light comes and blueshifts but it has not climbed to much of the potential (getting redshifted) when it hits your eye so only part of the blushift has been redshifted and the net result is that you can see the surce blueshifted. The other option is that the source is near the black hole so the light gets a "free fall from low altitude" (blueshifting) but has to overcome a huge amount of the gravity well to get to you because you are far away (very redshifted), so the net result is that the source gets redshifted.

SE it's accurate if we consider the black hole being seen through a telescope like in this classical depiction. But if we fly near it the relative distances of the souce-black hole-observer will change so much that when you could spot the lensing effect of the black hole (without zoom) you will be so near to maybe start to see the blueshifting of all the objects behind it (considering them extremly far away in comparison).
I have made some simplistic calculations. One of the most massive black holes known is APM 08279+5255 wich has nearly 23 billion solar masses. That means that the radius of the event horizon is 453 AU and because of that the diameter is 906 AU. If we are far enough from the black hole we could see the event horizon as big as the full moon, that angular diameter is reached when we are 103.821 AU away from singularity, or 1,64 light years away (the distorsions around and the Einstein ring would be much greater in angular size than the full moon at this distance so we ensure that the optical distortions can be totally seen from here). It's not so strange to immagine a star orbiting in a stable orbit at half a light year from the singularity. When the star is behing the black hole you will see it redshifted in this case and if you get closer to the black hole, the stars redshift would decrease and the background souces would start blueshifting quite a lot.

Is there any consideration that make this illogical? I haven't done the real computations so I can't know if this is a neglegible change in color or is evident. Maybe the effect is so extreme that the star show a gradient of color from blue to red, due to the differences in distance of every point of the surface sphere from the black hole.

An other thing that I want to ask to you is if SE takes into account multiple Einstein rings. When I see black holes on SE the Einstein ring angular size depends only on how far away from it I am but in real life the angular size of an Einstewin ring depends also from the distance of the light source:



where d_L is the angular diameter distance to the lens, d_S is the angular diameter distance to the source and d_LS is the angular diameter distance between the lens and the source. (Important to take into account that d_LS is not d_L - d_S at cosmological distances).
This mean that if there are two light sources behind the black hole at different distances one will create one ring and the other another completely separated ring (not the same as I see in SE). In fact double Einstein rings (like double rainbows cry ) have been spotted.
Is my reasoning correct?? In that case, Do you agree that SE should implement also this kind of thing (maybe it needs to much power to compute all the distances to the objects behind the black hole)?

The last thing I want to ask is about binary black hole's (not close ones by now) optical distortions. What could I see if I watch a black hole through the distortion shader of another one? if I encounter a binary black hole on SE (I haven't yet so I supose that it is not posible) would the "double distortion shader" be accurate enough?

Thank you so much and keep in mind I only want to create debate, I can't defend any position because I haven't studied spacetime geometry at university yet and all my assumptions are based on more or less reasoned intuition.

Attachments: 4540881.png(6Kb)


Edited by FastFourierTransform - Sunday, 07.09.2014, 11:00
 
SpaceEngineerDate: Sunday, 07.09.2014, 19:40 | Message # 365
Author of Space Engine
Group: Administrators
Russian Federation
Messages: 4796
Status: Offline
Quote FastFourierTransform ()
An other thing that I want to ask to you is if SE takes into account multiple Einstein rings. When I see black holes on SE the Einstein ring angular size depends only on how far away from it I am but in real life the angular size of an Einstewin ring depends also from the distance of the light source:


SE 0.971 don't have accurate gravity lens effect. It even don't use right formulas. And of course there no wavelength shift effect. This will change in 0.972. Did you saw this? http://en.spaceengine.org/forum/21-2070-43487-16-1409939160

But anyway, black hole distortion is a post-effect, ie it uses only rendered background image. No distance data is available, so it can't take it into account. In the formula you showed one can assume dLS >> dL and dS >> dL (this is true in 99% cases), so formula simplified greatly and can be used in post-process effect. Only black hole in the close binary systems would not accurately show the distortion effect (but they anyway should have the accretion disk that will mask the companion star). Even more, even when I change the rendering pipeline so the depth buffer would be accessible in the post-effect shaders, it will be impossible to accurately render the Einstein rings of two or more sources behind the black hole and behind each other: they simply overwrite each other in the depth buffer. It would be very hard to implement the correct algorithm (multi-pass is required).

Regarding to blue/redshift effects, I will try to implement them in the same post-effect shader, maybe like in those movies as a simple color saturation. But stars in SE are rendered using the conversion temperatre -> Plank thermal spectrum -> RGB value. So theoretically it is possible to model and render the real spectral shift for them, but this required a separate pass for star rendering (and stars certainly required a separate geometry pass, because simple image processing turn stars into a long arcs, which is incorrect because they actually are point sources). However, I may cheat and perform the reverse calculation in the distortion shader: RGB -> Plank spectrum -> temperature -> blueshift -> Plank spectrum -> RGB. This may give believable view for the stars, but obviously will be incorrect for the Milky Way and non-thermal-spectrum objects like planets. But in normal star systems I may ignore this, because planets can't approach close to the black hole. They may be observed behind the black hole in the telescope, but blueshift/redshift effect would cancel each other due to huge dL and dS.

And the last, current black hole rendering system is assumed that camera is far from the black hole. It will not render view close to it, view in direction parallel to the event horizon or in the backward direction. SE required the fast high-resolution skybox rendering to perform the accurate render of the distortion effect at any distance from the even horizon and in any view direction.





 
WatsisnameDate: Sunday, 07.09.2014, 21:00 | Message # 366
Galaxy Architect
Group: Global Moderators
United States
Messages: 2610
Status: Offline
Quote FastFourierTransform ()
I don't know, but surely this has also to be proven, They cancel exactly?


For a distant observer looking at a distant object behind the hole (i.e. the local gravitational field is same at both source and observer), then yes, they must cancel exactly. You just reasoned through it yourself by comparing with the situation of a particle moving through a gravitational well -- it has the exact same speed at radius R going out as it does at R going in. Gravity is a conservative force.

If this was not true, then we'd have to take it into account when measuring cosmological redshift of lensed galaxies.

If you like examples of things canceling out in relativistic physics, you might enjoy this beautiful example of the time dilation over the Earth's surface not being latitude dependent, counter to what intuition might initially suggest. At equator, rotational speed is faster (making clocks there tick slower), but the gravitational field is also weaker. Does one effect dominate? No, since Earth's shape very well approximates a rotating fluid, and it can be shown that such a surface is an equipotential surface, with no latitude-dependent time dilation. The only thing that does matter is altitude (clocks on top of a mountain tick more quickly).

Quote FastFourierTransform ()
PatrickPowers from PhysicsForums said this (not proven with real math):

What he says is correct, but you must be careful about motion of the observer. What he describes is for an observer at rest with respect to the hole. The view is very different if you are in free-fall.





 
DoctorOfSpaceDate: Monday, 08.09.2014, 00:22 | Message # 367
Galaxy Architect
Group: Global Moderators
Pirate
Messages: 3599
Status: Offline
Quote SpaceEngineer ()
I will replace it with the redshift/blueshift effect though.


Would it be possible to make a branch off of this and implement a red/blue shift for the camera/ships when traveling close to the speed of light, maybe as an optional cvar? Obviously rendering the other optical effects might be a bit more difficult, but an increasing FOV change and color shift doesn't seem like it would be too difficult to implement.





Intel Core i7-5820K 4.2GHz 6-Core Processor
G.Skill Ripjaws V Series 32GB (4 x 8GB) DDR4-2400 Memory
EVGA GTX 980 Ti SC 6GB
 
FastFourierTransformDate: Wednesday, 10.09.2014, 11:39 | Message # 368
Pioneer
Group: Local Moderators
Spain
Messages: 540
Status: Offline
Quote SpaceEngineer ()
This will change in 0.972. Did you saw this?


Yes, I saw it. Quite an amazing work there Vladimir!! You know, this is like when you comprehend more and more things, you develope the ability of knowing that you comprehend less and less of a vast natural world. Every time I see your development of the software I get so amazed that I make my little investigations and find more things that I want you to add to the programm smile
Now I'm on exams period so I can't search in the rigorous way I want, this things, but I am prepering some material I want to discuss with you and the SE community because it would be quite awesome if some of this ideas are implemented, and I'm doing the research (with my physics level in my second year at university) so the item can be precisely described.

Quote SpaceEngineer ()
So theoretically it is possible to model and render the real spectral shift for them, but this required a separate pass for star rendering (and stars certainly required a separate geometry pass, because simple image processing turn stars into a long arcs, which is incorrect because they actually are point sources).


Good point. I didn't notice until now! I am very happy you are aware of the fact that this depiction is not exactly how reality looks like (something very diffcoult to be aware of becouse movies, artists and scientist always render this as SE's does by now). If you have in mind this and more I know you would implement it one day (maybe in the far future ,it doesn't matter).
SE is the most well performed depiction of the universe I ever saw (even better than planetarium software and the artist concepts that NASA takes for images for the press). I know it's going towards perfection in that sense, and that is awesome, so thanks.

Quote SpaceEngineer ()
But in normal star systems I may ignore this, because planets can't approach close to the black hole.


That's true, but consider this when you are going to add disks around stars and black holes because aside the optical geometrical distortion you would need to take into account the red-blue shifts of the different parts of the disk. Here is an example that takes into account also the velocity of the spinning disk to render the image:



<a class="link" href="http://www.mssl.ucl.ac.uk/theory/bg/torus1.jpg" rel="nofollow" target="_blank">
</a>

Here you have the formulas and the algorithms to raytrace for rendering this images:
http://www.mssl.ucl.ac.uk/theory/pdf/bhdisk.pdf


Edited by FastFourierTransform - Wednesday, 10.09.2014, 11:40
 
JonasTheRomanDate: Friday, 19.09.2014, 19:10 | Message # 369
Observer
Group: Users
Germany
Messages: 18
Status: Offline
I don't know if it already was suggested, so here we go:
I think it would be nice to have an "automatic" modus for the planetarium. It would just randomly fly to different locations, stay there for a few moments (maybe circle around the object slowly) and then continue to fly to the next randomly chosen object, somehow like a space trip.

While this is not a very important addition, I think many people would sometimes enjoy to just let the camera fly around smoothly and look where it takes you, including me.
As an advanced comfort feature a settings dialogue could be implemented, where you can define things like flight speed, stay time, action performed at the object (like circling, going down to surface etc.), check and uncheck certain astronomic objects to visit and so on.

Thank you again for everything, SpaceEngineer! biggrin
 
HarbingerDawnDate: Friday, 19.09.2014, 21:54 | Message # 370
Cosmic Curator
Group: Administrators
United States
Messages: 8713
Status: Offline
Quote JonasTheRoman ()
I don't know if it already was suggested, so here we go:
I think it would be nice to have an "automatic" modus for the planetarium

Things like that have been suggested many, many times. I agree that some kind of automatic flying around the universe would be a nice feature, but it's low priority, so I wouldn't expect to see it for a long time.





All forum users, please read this!
My SE mods and addons
Phenom II X6 1090T 3.2 GHz, 16 GB DDR3 RAM, GTX 970 3584 MB VRAM
 
JonasTheRomanDate: Saturday, 20.09.2014, 09:51 | Message # 371
Observer
Group: Users
Germany
Messages: 18
Status: Offline
Quote HarbingerDawn ()
Things like that have been suggested many, many times. I agree that some kind of automatic flying around the universe would be a nice feature, but it's low priority, so I wouldn't expect to see it for a long time.


Thanks for the answer. Nice to hear that it is apparently planned, but yeah, there are more important things. smile
 
FastFourierTransformDate: Sunday, 21.09.2014, 09:59 | Message # 372
Pioneer
Group: Local Moderators
Spain
Messages: 540
Status: Offline
I don't know if this fits exactly on this thread but I'm going to put it here smile

The spanish Astrophysical Institute for Canary Islands (where we have the giant telescopes) with other institutions are making a documentary about the Solar System to be displayed on Planeteriums around Europe. The are showing demos of their animations and they are quite awesome. It looks a lot like the future Space Engine 1.0 biggrin





I like a lot the martian cave. Immagine that kind of things on SE!!!! The Venus surface looks totally non realistic but the amazing landscape could be from another planet on SE (if we had god rays, volumetric clouds, ect...). I like also the cometary tail development (geysers). The collisions are difficult to add to SE by now but seen this I realize that it would be a good implementation.
Immagine your computer procedurally creating those landscapes and animations in real time! One day... SOON... happy
 
SpaceEngineerDate: Sunday, 21.09.2014, 11:23 | Message # 373
Author of Space Engine
Group: Administrators
Russian Federation
Messages: 4796
Status: Offline
Why they render asteroid belts so unrealistic? People will look such movies "form official science" and will be sure that real asteroid belt look like this.
Also, the asteroid crashed on the Moon... lol
Explosions in a vacuum are looking very unlike explosions in an air. No dust trails after debris!

I'm curious, is there exist some software that can convert spherical panorama to Oculus Rift view? So you can feel yourself inside a planetarium dome and watch this movie.

EDIT: found the VR Player: http://vrplayer.codeplex.com
EDIT2: it's cool, despite on lack on depth and incorrect Oculus settings (I can't configure it to correct projection). I must record few such spherical videos with SE, but stereo, with depth.





 
WatsisnameDate: Sunday, 21.09.2014, 22:05 | Message # 374
Galaxy Architect
Group: Global Moderators
United States
Messages: 2610
Status: Offline
Quote SpaceEngineer ()
Also, the asteroid crashed on the Moon... lol


Not to mention its speed. Was the crater they were showing afterward supposed to be the result of that? wacko





 
FastFourierTransformDate: Monday, 22.09.2014, 01:52 | Message # 375
Pioneer
Group: Local Moderators
Spain
Messages: 540
Status: Offline
Quote SpaceEngineer ()
Why they render asteroid belts so unrealistic? People will look such movies "form official science" and will be sure that real asteroid belt look like this.

Totally agree. It appears that all those missconceptions are not erased so easily. Beside that, it's impossible to have something nearer to that depiction in a very short period in the protostellar disk? or it would be much more little objects?

Quote SpaceEngineer ()
the asteroid crashed on the Moon... lol

Yes hahahaha, I thinked the same. It reminded me that scene of transformers where some stupid thing collides with the lunar surface at similar speeds.

Quote SpaceEngineer ()
Explosions in a vacuum are looking very unlike explosions in an air.

Also agree
And I have to add that those "geysers" showing out of the cometary nucleus become turbulent like if the material where been spilled in a denser medium (air? ether?). Insted of been like a narrow shot of material it gets like liquid ink deforming rapidly.
Yes, like the great majority of documentaries it's going to fail in scientific accuracy (like in that moorgan freeman series about the universe xD). And yes, it's terrible to see that that comes from "official science". The fact is that Spain is having a terrible situation with the Scientific matter because the government is destroying all the possibilities por scientific research. With a financial cut of the 36% in investigation from 2009, closing dozens of laboratories, ect... now many scientist and organizations related to science in Spain are "prostituting the knowledge in some way" to be able to continue some researches (the other day a biologist appered at a tv show to win some price to finance their investigations, a scientist that has discovered the cure for a certain disease have been researching only with the money that he has gotten in a tombola made by his hood, and yes finally we arrive to the point that the scientific accuracy doesen't matter because the people are going to pay only for a SPACE SHOW!!!!).

Horrible situation for Science here (and we have very good science in the last decade). The next 26 of September there are going to be demonstrations and strikes of the scientific community in Greece, Italy, France and Spain perhaps we can make some presure to this medieval scumbags that we have at control.
 
Forum » SpaceEngine » Feedback and Suggestions » General suggestions (Post your suggestions here.)
Page 25 of 69«1223242526276869»
Search: