New Celestia-1.4.0pre-FT1.1 Version for Download

[PlutonianEmpire]

I must side with Cham and Malefant on this one, Fridger.

[BlindedByTheLight]

Yes, I do agree the dots are large and smugdy... but I can also imagine the reason for this... only so many dots can be handled for a reasonable CPU...

A few thoughts for solutions... is there a "spread" adjustment on each of the dots? So that they are more diffuse? Or perhaps a diffusion gas-like cloud could permeate the galaxies... to give them more of a sense of being FILLED with stars.

Are such things possible? Do they contribute to a greater overhead?

[BlindedByTheLight]

I must side with Cham and Malefant on this one, Fridger.

I will side with all three - Cham, Malegant AND Fridger...

AKA

...could be better (C&M), don't have the resources at this time (F).

Yatta!

[PlutonianEmpire]

Wow. You're right.

[hank]

Fridger,

the template we had some weeks (months?) ago was better, IMO. It was a better representation of most barred spirals, and I'm pretty sure it was satisfying your constraints.

The problem with this new template is its apparent asymetry. It just needs some corrections on the arms.

Again, you forget that the present converter does NOT allow to manipulate individual points! All you can do is a most tedious procedure by pre-treating the input .bmp model in contrast and brightness in certain areas. Then after each modification and re-conversion to .pts, you got to copy the new model into Celestia, start it and inspect it...Moreover there is randomness involved. So the thickening on the rhs arm of the template was such a random effect. I just can hardly do anything about it at this time.

The present SBc template you dislike cost me fare above 100 such cycles. You can easily estimate the large required time/template...

Bye Fridger

Clearly what is needed is a Celestia-based interactive galaxy template editor. Until we have that, there's not much point in complaining about the template quality because there's nothing that can really be done about it. Of cousre, if someone wants to try their hand at the laborious process described by Fridger, I'm sure that Toti's conversion program could be made available.

- Hank

[ElChristou]

Guys, I think the look of the templates are not a big problem, this can always be upgrated later...
For this we need to wait Toti tool in next big release and see what can be done or not.

Fridger, concerning my pict, I'm conscient of the wrong comparaison, but you must admit the m100 in Celestia is very similar to the heart of the photo...

To my point of view, there is no problem with your extraction from the pro catalogue; brightness and color must be right.
Cham has a point concerning the fact to use this or that galaxy as a template. We should create from scratch our own generic representation.
Talking about those famous template... can you show us one or two? Are they similar to something like this?

[ogg]

The criticism here has veered away from the constructive (Cham's in particular). This is a work in progress and Fridger has made clear what the limitations and priorities are at the moment. Getting things right under the hood regarding the code and the catalog should be the priority now, the paint job can wait until later.

Anyway, in terms of design philosophy this dispute is suggestive. There've always been two schools of thought about the goals of development: scientific visualisation vs shock-n-awe visuals. Proponents of the first will favour simple, schematic visuals and models, to cut the representation at the joints which science discovers. The second want rich content for models and textures so it 'looks right'. But whatever else it is, Fridger's SBc template is a classical barred spiral in the sense that it's schematically classic. It's structure as SBc is obvious, and you can coast past a galaxy with that template and instantly know there's an SBc there - which is what visualising this catalog data should be all about. The more detail you bundle into these templates, the less informative they will be.

And seeing that these templates are just files, and not hard coded, there'll probably be plenty of scope for people with differing design philosophies to eventually make and release their own galaxy packs. Just like add-on makers now. But it's CelestiaFT not CelestiaAnyoneelse we're talking about, so it's the FT combo that get to decide what they initially look like. Fridger's being entirely too defensive: if we don't like it, we can lump it.

All hail the makers of free stuff.

Ciao

[hank]

There've always been two schools of thought about the goals of development: scientific visualisation vs shock-n-awe visuals. Proponents of the first will favour simple, schematic visuals and models, to cut the representation at the joints which science discovers. The second wants rich content for models and textures so it 'looks right'.

Whatever that means. We've all been spoiled by long exposure observatory photography, which integrates the light onto a plate/chip over a much longer time than the retina is capable of. If you could look at those galaxies through the eyepiece, what you'd see wouldn't look like the images. If by "looks right" you mean "looks real", visually, then the images you're used to seeing aren't a good standard.

- Hank

[BlindedByTheLight]

Wow. You're right.

Do I detect a note of sarcasm...?

[ogg]

There've always been two schools of thought about the goals of development: scientific visualisation vs shock-n-awe visuals. Proponents of the first will favour simple, schematic visuals and models, to cut the representation at the joints which science discovers. The second wants rich content for models and textures so it 'looks right'.

Whatever that means. We've all been spoiled by long exposure observatory photography, which integrates the light onto a plate/chip over a much longer time than the retina is capable of. If you could look at those galaxies through the eyepiece, what you'd see wouldn't look like the images. If by "looks right" you mean "looks real", visually, then the images you're used to seeing aren't a good standard.

- Hank

Yeah I agree. That's why I put 'looks real' in inverted commas: I'm not really sure what it's supposed to mean either (except perhaps, 'like I see on APotD') and I'm on the side of the schematic approach. In this case though it seemed more of a 'structural' thing though rather than about pure photorealism. Perhaps I'm wrong about what Cham dislikes, but calling the templates 'crude' and when saying things like the following...

Fridger,

then Toti's program isn't good enough. It isn't adequate (or not "smart" enough). You need something more sophisticated to model a "typical" galaxy.

... it sounds like the point is being missed (and rudely). The stated point of this part of the development wasn't "the capacity to pick any galaxy you like and say, 'gee, that's pretty'", and doesn't sound like the main aim was to model a 'typical' galaxy at all. Fridger said time and again that the primary goal is to model the distribution of thousands of galaxies in celestia; to do for galaxy data what's already long been done for stars. I'm far more interested in that then any aesthetic judgements shouted from the sidelines by onlookers.

Anyway, enough ranting from me

[PlutonianEmpire]

Wow. You're right.

Do I detect a note of sarcasm...?

Concession/admission/whatever of defeat, actually.

[Malenfant]

... it sounds like the point is being missed (and rudely). The stated point of this part of the development wasn't "the capacity to pick any galaxy you like and say, 'gee, that's pretty'", and doesn't sound like the main aim was to model a 'typical' galaxy at all. Fridger said time and again that the primary goal is to model the distribution of thousands of galaxies in celestia; to do for galaxy data what's already long been done for stars. I'm far more interested in that then any aesthetic judgements shouted from the sidelines by onlookers.

Nobody has any issues with the distribution - that's working out well for everyone I think. That 'primary goal' has been achieved, and everyone's happy with that. It seems however that you're making a false argument here by turning that into the issue when it isn't at all.

The issue is that the appearance some of the templates is lacking in the opinions of several people here. I have to admit that in the example Cham showed above, the new template is considerably uglier than the old ones IMO - it could certainly use some refining. Now obviously, as Fridger points out, we can't get the images looking exactly right for each and every galaxy - the point is to get it "mostly right" for 10,000 galaxies. But that's still no reason why we can't try to take a little time to attempt to improve the appearance of the templates, surely? Frankly, I'd prefer something that looked more like the little picture in ElChristou's last post.

As for realism... well, yes we're spoiled by the long exposure photos. Though really, not everything in Celestia appears as the human eye would see it anyway. Certainly the stars don't look like what the human eye would see, because to do that would involve making glares and so on that monitors can't do very well. So there's already a precedent to tweaking the brightnesses of objects in the program there anyway. And besides which, if the point is to make an 'archetypal' galaxy template to stand in for the real ones, then surely we can have a little latitude in its appearance? The structure is what matters more really.

Certainly, there's no reason to stifle comment/criticism about anything here.

[BlindedByTheLight]

Concession/admission/whatever of defeat, actually.

lol...

and perhaps we can agree on the term "enlightenment" instead of defeat?

[BlindedByTheLight]

But that's still no reason why we can't try to take a little time to attempt to improve the appearance of the templates, surely?

If I may be so bold as to intrude again... everyone who posts here about "improving" the templates seems to be continuously missing the MAIN objection to such improvement from Fridger. And the quote about a "little time" I think highlights this...

Fridger's answers have REPEATEDLY explained the exact reasons why a "little time" cannot be taken for this endeavor.

1) It won't take a little time...

Let me just state that I spent weeks experimenting with the templates and I am certainly NOT a beginner in image manipulation. It's just NOT easy to do better given the constraints we have.

...and 2) He has used up all time has does have already on template-improvement at this time...

I just can hardly do anything about it at this time.

...implying, perhaps, later more will be done. If I am not mistaken, Fridger is well aware of the problems with the galaxies. My guess is he is also well aware of the time commitment required to deal with that problem. My further guess is that he is simply saying...

"Yes... could be better. No, I do not have the time to fix that issue right now."

I must point out that this all, of course, is not my area of expertise. But moderating between opposing parties who keep missing the point each other is making IS, in many ways, my area of expertise... hence, again, my "contribution" to the world of Celestia.

There is no stifling of opinions here. There has been a debate, then an explanation that more cannot be done at this time. To that, I would say, "Great! Now let's talk about the things that CAN be done at this time! Like incorporating doctorjoe's GOTO Spacecraft Start Time patch into CVS!"

[ElChristou]

... it sounds like the point is being missed (and rudely). The stated point of this part of the development wasn't "the capacity to pick any galaxy you like and say, 'gee, that's pretty'", and doesn't sound like the main aim was to model a 'typical' galaxy at all. Fridger said time and again that the primary goal is to model the distribution of thousands of galaxies in celestia; to do for galaxy data what's already long been done for stars. I'm far more interested in that then any aesthetic judgements shouted from the sidelines by onlookers...

I also suppose nobody missed that... this is an accepted point, nothing to say about this anymore apart bravo... everybody has been more than thankful I think. The dev of FT will go on, so I suppose we can make a pause and "try" to talk a bit about the actual state before begining new important tasks (clusters etc...). Seems Fridger lost some precious time with those templates and I'm a bit worried about this...

Steven, for sure there is several others important points, but IMO, one thing after the other...

[Malenfant]

I fully understand the limitations that Fridger mentions and that people keep repeating, but last time I looked this was an open source project. if Fridger is unable to tweak the templaters, couldn't others try to do so?

[BlindedByTheLight]

I fully understand the limitations that Fridger mentions and that people keep repeating, but last time I looked this was an open source project. if Fridger is unable to tweak the templaters, couldn't others try to do so?

Fair enough... any takers? I would... but I'm still not sure exactly what we're even talking about...

[t00fri]

I fully understand the limitations that Fridger mentions and that people keep repeating, but last time I looked this was an open source project. if Fridger is unable to tweak the templaters, couldn't others try to do so?

Fair enough... any takers? I would... but I'm still not sure exactly what we're even talking about...

Steven,

sorry for the next 2 weeks I will have very little time. I will also be in Rio de Janeiro for a week. Also Toti is right now very busy.

You analysed the situation above quite well. ElChristou is trying right now, ...i.e. learning more explicitly why the problem is so hard . Here are some quick remarks that I also wrote ElChristou by email:

The /input/ files for Toti's vectorizer are grayscale 128x128 .bmp images that can be modified with Gimp, PS etc. The output (.pts file) of bmp2pts is a set of 3d points (x,y,z) each with an intensity value i (0<i<1). These .pts files can be displayed in Celestia or other software like Maple, but NOT edited anymore! Typically a .pts file has 5000 such points or more.

+++++++++++++++++++
The tedious part is that a good-looking .bmp input may well give a much worse final .pts point distribution than a BAD-looking one. So starting from a nice .bmp template image is NO guarantee whatsoever that the .pts result is usable. Usually the nicer the .bmp looks the worse the resulting .pts!
+++++++++++++++++++

In particular, it depends on the /absolute/ brightness and contrast levels at each point in the .bmp image whether a corresponding dot is generated by Totis program or NOT. In addition one must set a level of randomness that is necessary, but hard to control as to the way it looks at the end.

So thinking about .bmp template images without trying them out right away through Toti's 'bmp2pts' is COMPLETELY useless! If I send you such templates you will be completely mislead, I guarantee. I was also mislead initially.

Toti's code is standard C++ code that you may compile in any OS you like. Toti is just (correctly) reluctant to distribute it, since then we get so many questions that we have to stop working or ignore the questions. It only makes sense to distribute it to people who can read directly from the code what the program is doing and HOW it is doing it.

Cheers,
Fridger

[Malenfant]

Hm. Well that goes some way to explaining how the templates end up looking the way they do at least...

what happens if you remove the randomness when generating the pts file?

[hank]

Typically a .pts file has 5000 such points or more.

I believe the current template for Sb galaxies has 9,793 points.

- Hank