Feedback regarding newstars and fading orbits...

[Chuft-Captain]

1. The newstars feature is very very nice, however one thing I noticed is that when increasing the "magnitude limit", the brightest stars are also increased in brightness. If this behaviour is intended then the "magnitude limit" needs to be renamed, because it appears to be affecting absolute brightness (rather than / as well as) magnitude limit.
I personally feel that this would look better if it progressively allowed dimmer and dimmer stars to come into view without increasing the absolute brightness of the already bright stars.
(ie. As a true cutoff filter)

2. The fading orbits is a good idea, but I'm not so much a fan of the implementation.
I think this idea would work better if the amount of fading was a function of the distance of each part of the orbit from the viewer.

- Most distant parts of the orbit would have the most fading.
- An orbit viewed from directly above (at 90 degrees) would have uniform fading across the entire orbit line.

This would be very nice IMO, as the orientation of the orbit to the viewer would be visually represented. (ie. No more ambiguity as to whether Apparent Eccentricity was due to Actual Eccentricity of the orbit or merely the result of a tilted viewpoint).
(This would probably be more computationally restrictive, as it would have to happen within the render cycle, but would certainly be more useful IMHO.)

Just my 2CW
CC

[t00fri]

1. The newstars feature is very very nice, however one thing I noticed is that when increasing the "magnitude limit", the brightest stars are also increased in brightness. If this behaviour is intended then the "magnitude limit" needs to be renamed, because it appears to be affecting absolute brightness (rather than / as well as) magnitude limit.
I personally feel that this would look better if it progressively allowed dimmer and dimmer stars to come into view without increasing the absolute brightness of the already bright stars.
(ie. As a true cutoff filter)

...

Just my 2CW
CC

CC,

had you made this proposition over at CelestialMatters, I'd have shown you my way of reaching app.mags of 25 or so in celestia.Sci. Here I obviously don't want to talk too much about celestia.Sci ...

Fridger

[Chuft-Captain]

1. The newstars feature is very very nice, however one thing I noticed is that when increasing the "magnitude limit", the brightest stars are also increased in brightness. If this behaviour is intended then the "magnitude limit" needs to be renamed, because it appears to be affecting absolute brightness (rather than / as well as) magnitude limit.
I personally feel that this would look better if it progressively allowed dimmer and dimmer stars to come into view without increasing the absolute brightness of the already bright stars.
(ie. As a true cutoff filter)

...

Just my 2CW
CC

CC,

had you made this proposition over at CelestialMatters, I'd have shown you my way of reaching app.mags of 25 or so in celestia.Sci. Here I obviously don't want to talk too much about celestia.Sci ...

Fridger

Glad to hear that you are also making improvements.
I haven't been around here much over the last few years, so I don't know the history or motivations for your fork of the code, but I really don't see why you can't discuss it here as well as at CM. It's still surely a version of Celestia and derived from work that's gone into the original code-base over the years.
You've been very keen in the past to improve the quality of discussion on this board. This would be an opportunity to do so, surely?

Anyway,
In addition to the cutoff issue raised above, I've also noticed that the newstars functionality is applied to spacecraft as well as to stars (they appear brighter in newstars mode than in fuzzy mode). I don't know if this is intended, or if the newstars code should be restricted to stars only.
Personally, I'm on the fence as to whether this change in the appearance of spacecraft is desirable or not. With my models, under some circumstances this has an enhancing effect, in other circumstances, it appears to be detrimental. (Purely from an asthetics point of view.)

What did you think of my proposition regarding the fading orbits?

Cheers
CC

[t00fri]

Glad to hear that you are also making improvements.
I haven't been around here much over the last few years, so I don't know the history or motivations for your fork of the code, but I really don't see why you can't discuss it here as well as at CM. It's still surely a version of Celestia and derived from work that's gone into the original code-base over the years.
You've been very keen in the past to improve the quality of discussion on this board. This would be an opportunity to do so, surely?

...
Cheers
CC

CC,

things are unfortunately much more complex. Chris and I have peacefully agreed that I will not be using his site too much to tell people about all the nice features that celestia.Sci offers, given the situation that Celestia development is in deep coma since a long time. I think this is fair.

celestia.Sci will look less and less like Celestia 1.6.1 with time.

You've been very keen in the past to improve the quality of discussion on this board. This would be an opportunity to do so, surely?

After most of the forum posts have been taken over by a bunch of fanatic StarWar/StarTrek fans, there is not much that remains for me to say or to improve over here.

That's why people can find me increasingly over at my CelestialMatters site. Be assured that I am not alone at CM .

Cheers,
Fridger

[Chuft-Captain]

In that case, as NEWSTARS is clearly Chris' initiative, it's appropriate for me to address my comments to him, so this is clearly the appropriate forum for that.
So I'll continue with that here, and keep a lazy eye out for any CM developments as usual.

Orion is much more well defined and looking very good at close to the default magnitude limit ...

ORION maglimit 7.9 - sidebyside.jpg

.. although, on my display I noticed a subtle washing out of the colors in NEWSTARS. Betelgeuse is not as red, the belt stars are less blue. Perhaps this is an unavoidable consequence of the extra contrast. (I'll have to check out the real thing next time there's a clear sky to see how it compares.)

CC

[t00fri]

In that case, as NEWSTARS is clearly Chris' initiative, it's appropriate for me to address my comments to him, so this is clearly the appropriate forum for that.

I am afraid we have not seen Chris here for a long long time...

So I'll continue with that here, and keep a lazy eye out for any CM developments as usual.
...
CC

Apparently you missed that celestia.Sci also has "newstars" (<=> shader stars) BOTH for normal star rendering and for the stars of all globular clusters. The corresponding code is largely different from that of Chris or more precisely that of VESTA@ASTOS. Both Chris and I use a combination of a varying disc size and a gaussian halo in our shader codes, in order to be able to account for the huge required difference in star brightness.

I am quite content so far, since the faintest globular stars (appmag <= 25m!) match well with really big sized stars. See e.g. this image of the dim globular NGC 2419 that I showed already previously at shatters.net:



Note that the two "super bright" stars on the right are only 7.2m and 7.9m, respectively! The not-so-bright one near the bottom is 11.16m. Most visible stars in NGC 2419 are within an app.mag range of 20m - 25m. i.e extremely faint!

The display does pretty well in comparison with the SDSS DR7 image of NGC 2419


Fridger

[Chuft-Captain]

Maybe Chris lurks from time to time, but he's certainly not posted or responded to PM's for some time.

Apparently you missed that celestia.Sci also has "newstars"

No, I didn't miss that ... because you told me about it a couple of posts ago:

had you made this proposition over at CelestialMatters, I'd have shown you my way of reaching app.mags of 25 or so

What I mean is that, because I am only using Chris' pre-release NEWSTARS, my comments in this thread are limited to that.

I can't really make any comparisons with, or give any useful feedback wrt. Celestia Sci, as it hasn't yet been released in any form that I can use, AFAIK.
( and I wouldn't want to endanger your gentleman's agreement with Chris by leading you to discuss Celestia Sci in this forum. )

Regards
CC

[t00fri]

Maybe Chris lurks from time to time, but he's certainly not posted or responded to PM's for some time.

Lurk mode cannot be excluded, but I am afraid that Chris has largely cut the line to Celestia since the beginning of this year. We can also tell from all the important unanswered mails in the "developers list".

Apparently you missed that celestia.Sci also has "newstars"

No, I didn't miss that ... because you told me about it a couple of posts ago:

had you made this proposition over at CelestialMatters, I'd have shown you my way of reaching app.mags of 25 or so

What I mean is that, because I am only using Chris' pre-release NEWSTARS, my comments in this thread are limited to that.

OK, well . But there is more to my NEWSTARS method than that single sentence of mine.

I can't really make any comparisons with, or give any useful feedback wrt. Celestia Sci, as it hasn't yet been released in any form that I can use, AFAIK.

True, but also the above post of yours to the Celestia Devs will be futile, since Celestia development
has ceased to exist. (If not, I would be happy, of course...)

At CelestialMatters, a number of experienced Celestians are currently discussing various rendering features of celestia.Sci with me, despite a public release being still unavailable. This is certainly most helpful for me.

Regards,
Fridger

[Chuft-Captain]

After spending a little more time examining the fading orbits, I now better understand the rationale behind the implementation. The idea is to draw the orbit slowly fading (eventually to nothing) behind the object as it moves in it's orbit. A bit like the fading trail of a comets tail. There is a logic to this which makes sense as it gives a nice visual indicator of the object's direction of movement in the orbit.

While I think there's still some merit in my suggestion at the beginning of this thread, Chris' approach is growing on me.

In any case, I think we should have the option to select between the standard old style orbit lines and any new options such as this (if space permits on the View->Render menu), as the original style still has advantages in some display circumstances (eg. in visualizing the plane of an orbit).

CC