Board index Other Hangar talk

It's No Moon

Talk about (almost) anything, as long as it is no serious FlightGear talk and does not fit in the other subforums.
Forum rules
Please refrain from discussing politics.

Re: It's No Moon

Postby D-ECHO » Sun May 10, 2020 10:30 am

The UdSSR had a reason to deny moon attempts, whereas I can not see any reason to deny something like this in FG, it just doesn't make sense to me.
User avatar
D-ECHO
 
Posts: 1995
Joined: Sat May 09, 2015 12:31 pm

Re: It's No Moon

Postby vitos » Sun May 10, 2020 10:36 am

The UdSSR had a reason to deny moon attempts, whereas I can not see any reason to deny something like this in FG, it just doesn't make sense to me.


You are omitting that reason, which USSR had - what Moon flight is tricky business, and something should fail for sure, while not to say and to make is much more expressive than to say and not make - and that expression means a lot at means of own position at hierarchy, etc.

If most motivating force is personal/national pride here - which Vostok vs. Shuttle story proved to me without a doubt, otherwise space projects with different historical background would be treated same - then same reason could be here.

I had heard "we do not need space" at that closed topic, then it turns out other way. Can't see why same could not be repeated - it just needs someone to make Moon flight first, to be ripped off. Definitely not me.

Let's return to discussion instead: people now at more or less same situation, as one USSR had fall into - space is tricky and risky still, and quite expensive, while Apollo was already. In fact, that situation going deeper, as Apollo was longer and longer ago, while space did not become cheaper much.
Waste of time: too unprofitable for work, too exhausting for hobby.
User avatar
vitos
 
Posts: 615
Joined: Sun Jan 25, 2009 8:10 pm
Location: Moscow, Russia
Callsign: vitos
IRC name: vitos
Version: 3.4
OS: Debian

Re: It's No Moon

Postby bugman » Sun May 10, 2020 11:26 am

As a moderator: We do not operate using an anarchist system. Like all other systems, we have rules. A spam post is 'off-topic' and it, and all subsequent discussion, will be removed (forum rule 7). If you mix this into your argument, because of emotion, your argument looses credibility.

As a user: There is interest in implementing the possibility of landing on the moon. But this requires a clear plan. How is it going to be implemented? What is the design? Who will write the code?

Previously you have made it clear - you do not like graphics tricks. But everyone else thinks differently - graphics tricks are needed. When on earth, the moon can be an OpenGL disk. You can switch to a spheroid, or sphere, once you are closer. Such graphics tricks are used all the time in games, 3D visualization software, and other 3D applications. Not accepting graphics tricks and arguing that everything must be simulated as in 'reality' creates an impasse. We cannot move past the design discussions.

No one can implement if there is no design.

Regards,
Edward
bugman
Moderator
 
Posts: 1774
Joined: Thu Mar 19, 2015 9:01 am
Version: next

Re: It's No Moon

Postby Hooray » Sun May 10, 2020 11:39 am

You cannot do a flight or space simulation withOUT playing tricks ...

Using the new compositor, it would be relatively straightforward to make it render a new/alternate scene graph - without using any of our existing terrain/scenery system at all.
You merely need a way to obtain the corresponding GIS data/DEM and a way to tell FlightGear to render that (e.g. a GeoTIFF)

In a way, this could work analogous to hooking up arbitrary 3D models to be shown by a Canvas, unrelated to any other subsystem:

http://wiki.flightgear.org/Howto:Extend ... _3D_models
Image

In other words, if you had a 3D model of a textured sphere, you could equally display that using the same approach. Or use the OpenGL disc that bugman mentioned when sufficiently far away.

http://wiki.flightgear.org/Canvas_Sandbox#CanvasShape
Image

Once you have the DEM, you can also render a height field using the same approach (via canvas and/or compositor, i.e. via the same OSG code):

http://wiki.flightgear.org/Canvas_Sandb ... eightField




PS: http://wiki.flightgear.org/Reaching_for_the_stars (yes, I do realize that the moon is not a planet)
Please don't send support requests by PM, instead post your questions on the forum so that all users can contribute and benefit
Thanks & all the best,
Hooray
Help write next month's newsletter !
pui2canvas | MapStructure | Canvas Development | Programming resources
Hooray
 
Posts: 11738
Joined: Tue Mar 25, 2008 8:40 am

Re: It's No Moon

Postby vitos » Sun May 10, 2020 12:03 pm

There is interest in implementing the possibility of landing on the moon. But this requires a clear plan. How is it going to be implemented? What is the design? Who will write the code?


I don't know, not me. I had became pretty unsatisfied with my previous projects, and not at part of projects itself. Even idea to make something better out of my already existed models making me angry.

Anyway, technically, virtual landing at Moon is possible at nasal level even, same way as virtual docking with ISS, but instead of ISS it's possible to draw some sphere, same sort as Earth at Earthview. Of course, math would be not exact as needed at closing, but it's possible to trick it, changing at some point of flight from absolute coordinates to relative ones, eliminating all mistakes with "correction burn", taking data from relative coordinates, which would be hidden from user. When Moon is shown at model level, then it's possible to draw it relatively to model smoothly enough... Looks alot like whole this thing, yeah.

In fact, I do wonder, why that "cheap Moon" approach was not implemented already. I did await something as that from well known and respected developers to Apollo-11 anniversary.

But, though, to make big plans and universal solutions is other way than to make something.

I was proposed to join someone with my MPClash at 2015 as I can remember, instead of just using it - my MPClash working as it worked then, and allowing adding as many types of missiles as needed, visible at flight to opponent who installed protocol at his model easily, while that Scripted AI Object did not provide any result for five years yet, but "at active development" still. To make some shell to someone else's solution, or some patch, and then to try to make it universal, which is impossible, of course - maybe with "cheap Moon" it turned something as that secretly.
Waste of time: too unprofitable for work, too exhausting for hobby.
User avatar
vitos
 
Posts: 615
Joined: Sun Jan 25, 2009 8:10 pm
Location: Moscow, Russia
Callsign: vitos
IRC name: vitos
Version: 3.4
OS: Debian

Re: It's No Moon

Postby Thorsten » Sun May 10, 2020 2:13 pm

If most motivating force is personal/national pride here - which Vostok vs. Shuttle story proved to me without a doubt, otherwise space projects with different historical background would be treated same - then same reason could be here.


ROFL!

I honestly had no idea that the Space Shuttle was the national pride of Finland :shock:

What you call 'prove without a doubt' is for the rest of the world known as an 'unplausible conjecture'. Just because you opt to work on craft from your country doesn't mean the rest of us does. :D

In your mind, the only possible reason for modeling one craft but not the other apparently is 'pride'. However, in reality, the fact of the matter is that I can conjure up thousands of pages of wind tunnel data, technical specifications and manuals for the Space Shuttle easily. They're public domain - like most of NASA material. Which makes realistic modeling possible.

See, here's my counter-proposal: We're not learning anything about a society here. We're not learning anything by the comparison between FG and the Soviet union either. All we learn is about the mindset of one person - you.

You're a really good aircraft creator - but that isn't enough for you. You want to be the best - so you talk down on the efforts of others, you make wild claims of having invented everything first (as if JSBSim hadn't been simulating orbital flight in a NASA benchmark test before) and you accuse them of stealing ideas from you (as if e.g. I with a PhD in theoretical physics would need any info about how to model spacecraft dynamics from you...).

And because you simply can't accept not to be the best, your solutions have to be the best. Your polygon-heavy models which load and render slow and crash weaker systems have to be the epitome of 3d modeling, you've never written any rendering code in your life, but you're the judge of how that is best done and that 'tricks' are bad.

And because you are convinced that your ideas are the best, you simply can't deal gracefully with the fact that people think otherwise and don't drop everything they're doing to work on 'what is best' in your mind. People don't collaborate on the basic principle that you tell them what to do - but rather than working from that simple insight, you come up with complicated reasons why that is - it has to be 'national pride' that people opt to contribute to the Shuttle, it couldn't possibly be that people simply find it interesting and have fun working on it - because they don't get told all the time what to do. It has to be a grand failure of society not to deliver your moon rocket, it's not enough that you're the only guy who wants it and no one else is interested enough.

See, I believe we're getting to read all these very involved social texts for one simple reason - to hide that you're 'only' a really good aircraft creator and not the best of the best FG has ever seen.
Last edited by Thorsten on Sun May 10, 2020 4:50 pm, edited 1 time in total.
Thorsten
 
Posts: 11580
Joined: Mon Nov 02, 2009 8:33 am

Re: It's No Moon

Postby Hooray » Sun May 10, 2020 3:07 pm

vitos wrote:In fact, I do wonder, why that "cheap Moon" approach was not implemented already. I did await something as that from well known and respected developers to Apollo-11 anniversary.


Well, I don't know, I cannot really relate to this kind of reasoning or the underlying attitude.
I think it's obvious that people in the FlightGear community, including contributors with a certain track record, are interested in extending FlightGear for space flight.
Obviously, there is a certain inertia involved, and you're dealing with something that the project never really designed/prioritized for, but nevertheless we do have a number of existing features that got implemented despite such shortcomings, often without any support from core developers at all.

In fact, you only need to look at the shuttle, earthview, ALS, Rembrandt, the Advanced Weather system or flug's bombable work, to see that highly motivated people with a corresponding background are basically unstoppable - regardless of whether the core project ("inner circle") is supportive of a certain feature/approach, it can still "happen". Basically, because FlightGear is a petri dish of a project - and opinions matter much less than concrete involvement - even if that involvement ends up using an approach or technology stack that is considered "inferior" by other contributors, who may have taken a backseat meanwhile.

The point is, your background does matter - which is exemplified by Thorsten's track record, but also flug's work. People with a strong background in maths, physics or a related CS/SE field, can be unstoppable, even without support from the inner circle, and even under adverse circumstances (c.f. AW being implemented in Nasal space, vs. many core developers wanting "core" functionality to be written in C++)

The "cheap moon approach" you're talking about might have been shot down immediately by ... people like you, wouldn't it ? :D
Seriously though, if there is some kind of anniversary and you have an idea in mind, just put up a challenge and offer your help to mentor others interested in your idea.

Telling the community that you are basically disappointed because something you envisioned didn't take place, seems rather pointless.

Then again, I also cannot relate to the contributor community's motivation being proportional or even just remotely related to national pride.
And to be honest, whenever history events were recreated in FlightGear, those were often military and many contributors didn't want to see FlightGear's image "tainted" by certain historic events.

So, I am not sure I can fully relate to your line of reasoning here - but if you'd like to encourage/motivate people to work on something you're interested in, there are certainly better ways - and you might even succeed, if you can find out a way how to contribute your own skill set and expertise to the whole effort. Either way, good luck to you !
Please don't send support requests by PM, instead post your questions on the forum so that all users can contribute and benefit
Thanks & all the best,
Hooray
Help write next month's newsletter !
pui2canvas | MapStructure | Canvas Development | Programming resources
Hooray
 
Posts: 11738
Joined: Tue Mar 25, 2008 8:40 am

Re: It's No Moon

Postby vitos » Sun May 10, 2020 3:14 pm

The "cheap moon approach" you're talking about might have been shot down immediately by ... people like you, wouldn't it ?


I personally was quite against switchable EarthView at nasal level from the beginning, idea as "Ok, now we don't see anything at this altitude, as simulator never was planned for that, so, instead of simulator improving, let's switch altitude patch on manually, it works by interpreter, at same level as plane model instrumentation" looks corrupted deeply to me. It did not stopped anyone; as my approval of really new terrain engine, which allowed smooth and nice view from any altitude, but demanded to compile simulator with it at least , did not helped its creator much.

Let me remind it:

Image

So, I suppose, matter is not "stoppable/unstoppable". At least matter of good working thing.

Of course I would make whole Moon project alone. But idea was not just to make it by any cost. Basically, I am not idiot to be unstoppable other way than defending people from idiots thinking they are unstoppable good.
Last edited by vitos on Sun May 10, 2020 3:52 pm, edited 1 time in total.
Waste of time: too unprofitable for work, too exhausting for hobby.
User avatar
vitos
 
Posts: 615
Joined: Sun Jan 25, 2009 8:10 pm
Location: Moscow, Russia
Callsign: vitos
IRC name: vitos
Version: 3.4
OS: Debian

Re: It's No Moon

Postby Hooray » Sun May 10, 2020 3:49 pm

I don't know, but Thorsten repeatedly described how Earthview works under the hood, and given your own interests and your contributions/background in Nasal space, I suppose that you could be walked through adapting/generalizing EarthView to make it render other spheres/texture sets, and then take it from there ?

Again, I've had my own share of disputes with Thorsten over the years, but he's always been responsive and helpful when I reached out to him afterwards, so I don't think there's any technical issue here, is there ?

The point being, you need to be able to walk before you run - and EarthView can help grow a community of people interested in space flight, exemplified by the earthview/shuttle developments and the monster thread detailing development/support of the space shuttle.

And like I said previously, with the compositor/canvas integration, it will become possible to circumvent much (all) of the legacy rendering/scenery engine and simply render a new scene graph that can be populated at will - including textured spheres or digital elevation. At that point, you'll end up with am empty "fgfs" window (osgviewer) that may or may not use CompositorViewer/PagedLOD, but other than it, it's an empty window that you can populate like you want, using a combination of 3D models, textures, XML, scripts, effects/shaders and Canvas/Nasal stuff.


Besides, that also means that you don't have to put up with TerraGear.

Something like Earthview could be implemented on top of this framework without having to "hack around" hard-coded limitations, without requiring core developer support/endorsement at all
Please don't send support requests by PM, instead post your questions on the forum so that all users can contribute and benefit
Thanks & all the best,
Hooray
Help write next month's newsletter !
pui2canvas | MapStructure | Canvas Development | Programming resources
Hooray
 
Posts: 11738
Joined: Tue Mar 25, 2008 8:40 am

Re: It's No Moon

Postby vitos » Sun May 10, 2020 3:58 pm

it will become possible to circumvent much


Something tells me it would become so slow, that You would need to buy whole new computer just to make it work at same speed as before with just same things.

Anyway, was new terrain engine implemented? Not, it uses same patch instead. Did simulator became multi threading? Not. It was grew more patches instead. You may just compile some circa-2015 version and get same picture but twice faster.

Matter, I suppose, exactly that some people are being unstoppable by others, while some people are not, while ones being stoppable are providing better solutions.
Waste of time: too unprofitable for work, too exhausting for hobby.
User avatar
vitos
 
Posts: 615
Joined: Sun Jan 25, 2009 8:10 pm
Location: Moscow, Russia
Callsign: vitos
IRC name: vitos
Version: 3.4
OS: Debian

Re: It's No Moon

Postby Hooray » Sun May 10, 2020 4:10 pm

Technically, fgviewer is just an osgviewer based subset of flightgear, but much more lightweight - which is why some people have been using it pre-view fgfs related 3D resources.
In other words, a lightweight fgfs mode without terrain/scenery subsystems running at all, is unlikely to be "slow" compared to the rest of fgfs.

You can already right now tinker with just osgviewer to load your 3D models/texture sets and open the osg stats to see for yourself, completely unrelated to fgfs.
Please don't send support requests by PM, instead post your questions on the forum so that all users can contribute and benefit
Thanks & all the best,
Hooray
Help write next month's newsletter !
pui2canvas | MapStructure | Canvas Development | Programming resources
Hooray
 
Posts: 11738
Joined: Tue Mar 25, 2008 8:40 am

Re: It's No Moon

Postby Thorsten » Sun May 10, 2020 4:39 pm

vitos wrote in Sun May 10, 2020 3:14 pm:Let me remind it:

Image

So, I suppose, matter is not "stoppable/unstoppable". At least matter of good working thing.



You're happy with that it seems, and here's the 'corrupted' approach:

Image

Image

To each his own I guess :D


vitos wrote in Sun May 10, 2020 3:14 pm:Of course I would make whole Moon project alone. But idea was not just to make it by any cost. Basically, I am not idiot to be unstoppable other way than defending people from idiots thinking they are unstoppable good.


Of course you could try to work with people rather than order them around, make use of their experience rather than declaring that you're expert of everything and know best - so you wouldn't have to do it alone.

The fact of the matter is that you couldn't do a moon project even if you wanted to, because you completely lack the background in numerics, coding,... about anything. Based on your codes and ideas so far, the solution you'd come up with would be so horribly inefficient (because you frown on 'tricks') that you couldn't get out usable framerate.

So naturally it's very comfortable to sit back and blame others - yet - we're not playing along, sorry. No moon for you that way :D
Thorsten
 
Posts: 11580
Joined: Mon Nov 02, 2009 8:33 am

Re: It's No Moon

Postby Hooray » Sun May 10, 2020 5:02 pm

To make the point, poweroftwo's osgEarth integration is another set of patches that demonstrate how a completely new scenery/terrain engine can be implemented next to FlightGear, without requiring any endorsement or support by the community of FlightGear core developers.

So, given that we do have both, an fgdata approach, and a core integration - you can make up your own mind how to proceed from here on. But it's not like it's impossible to turn FlightGear into a space simulator

PS: I suppose, Thorsten could be convinced to provide some pointers to make the EarthView 3D sphere/texture set configurable, so that people can tinker with other celestial bodies - and then take it from there, to generalize the whole idea/approach. All it takes, is a little dedication - and that includes people able to make concessions and find compromises. Which is the single most important reason why motivated folks tends to fail so often, just look at fgmembers or those TargetWare folks (Bomber) - they had all the right skills, expertise and motivation - they just didn't want to make concessions, i.e. they didn't realize that they needed to walk before they could run.

Earthview may not be what you want, but almost certainly Earthview will not look like it looks now in 5+ years time - depending on people like you, people wanting to use it for similar purposes.
Please don't send support requests by PM, instead post your questions on the forum so that all users can contribute and benefit
Thanks & all the best,
Hooray
Help write next month's newsletter !
pui2canvas | MapStructure | Canvas Development | Programming resources
Hooray
 
Posts: 11738
Joined: Tue Mar 25, 2008 8:40 am

Re: It's No Moon

Postby Andreas » Sun May 10, 2020 5:39 pm

vitos wrote in Sun May 10, 2020 8:08 am:If I was about hate, then I would rage quit immediately after my MiG-15 was not treated way I meant [...]

Hi Vitos! It's been a while!

I'm surprised: How was the MiG-15 not treated the way you meant? We stayed in contact while I made my improvements / corrections to your original model, I made sure to credit your achievement (which - I'll say it again - was really amazing) and provided detailed documentation about every change I made. Everyone is welcome to debate them here: viewtopic.php?f=4&t=28401

I even like to think that the MiG-15bis is a successful example of how open-source development should work, because we both covered different fields of expertise and that made us a good team, although I started my work after you finished yours.

You definitely weren't in rage back then and I'm not aware of any criticism about my edits.
How come your memory differs so much from your attitude back then?

Cheers Andreas
To an optimist, the glass is half full.
To a pessimist, the glass is half empty.
To an engineer, it's twice the size it needs to be.
User avatar
Andreas
 
Posts: 128
Joined: Sat Mar 14, 2009 6:11 pm
Callsign: D-AZ
Version: 3.4.1
OS: Debian Buster

Re: It's No Moon

Postby vitos » Sun May 10, 2020 5:47 pm

In other words, a lightweight fgfs mode without terrain/scenery subsystems running at all, is unlikely to be "slow" compared to the rest of fgfs.


Wait, as I could comprehend, it meant to be runned not instead of something, but in addition to something - how it could accelerate something then?

To make the point, poweroftwo's osgEarth integration is another set of patches that demonstrate how a completely new scenery/terrain engine can be implemented next to FlightGear, without requiring any endorsement or support by the community of FlightGear core developers.


I had tried it - it's just slow.

PS: I suppose, Thorsten could be convinced to provide some pointers to make the EarthView 3D sphere/texture set configurable, so that people can tinker with other celestial bodies - and then take it from there, to generalize the whole idea/approach.


That's exactly what I said previously - not just one additional nasal sphere, but two. Or even three, if someone would mind to implement Mars flight at coordinate and gravity systems with centre at Earth.

I just wonder why it was not implemented yet, to make it "first open flight to the Moon". Too complicated still, should be.

I'm surprised: How was the MiG-15 not treated the way you meant?


Ten years ago here was standard planes package, and a pile of any other models. MiG went not into package, but into pile, most models of which had't 3D cockpits at all. Of course, when all breakage system ideas was ripped off, and quality of greater amount of models had close to MiG, it was became safe to mention it just as common model.

While adding 10% max to already made model and own name into list of developers at same level as mine allowed someone to say it's example of how Open Source should work.

Just compare it with Su-15, which I made alone, and did't made it GPL just for that comparison.
Waste of time: too unprofitable for work, too exhausting for hobby.
User avatar
vitos
 
Posts: 615
Joined: Sun Jan 25, 2009 8:10 pm
Location: Moscow, Russia
Callsign: vitos
IRC name: vitos
Version: 3.4
OS: Debian

PreviousNext

Return to Hangar talk

Who is online

Users browsing this forum: No registered users and 1 guest