Board index FlightGear Development Aircraft

F-20 development

Questions and discussion about creating aircraft. Flight dynamics, 3d models, cockpits, systems, animation, textures.

Re: F-20 teaser

Postby Thorsten » Thu Dec 20, 2012 7:25 am

It seems that adding that many polygons to the model does not have a hit on frame rate (weird)


That'd depend on the framework. Rembrandt has a very simple geometry stage and does all the work per pixel, i.e. it can crunch models with high vertex counts incredibly efficiently but loses performance as compared to other strategies on scenery where polygons are large. Dependent on the precise set of effects running for the model, a vertex-shader dominated strategy might execute rather slowly.

Also, it depends on how many 'that many' are - my old GPU could crunch about 500.000 vertices in terrain and models without large performance deterioration, then it weakened significantly (with the new one, I haven't ever been able to measure any performance hit due to terrain, it largely goes with clouds only).
Thorsten
 
Posts: 11191
Joined: Mon Nov 02, 2009 8:33 am

Re: F-20 teaser

Postby Flying toaster » Thu Dec 20, 2012 6:26 pm

Thanks for the insight ...
So I guess it is going to be a high polygon count model fest :) !
BTW, I put a very very very (did I say VERY ?) alpha version of the tigershark download link in the "Canvas" forum. I guess I license it as GPL just in case
Feel free to give me feedback in this thread on the performance you get on you PC.

Cheers

Enrique
Flying toaster
 
Posts: 353
Joined: Wed Nov 29, 2006 6:25 am
Location: Toulouse France

Re: F-20 development

Postby Flying toaster » Mon Jan 07, 2013 9:29 pm

Comes complete with start cart :) :
Image
not functional (yet) though...

and an animated drag chute

Image

To download follow that link : F-20A alpha version
The link may be broken if my ISP imposes a bandwith cap though
The HUD is fairly broken, as I am in the process of migrating to Canvas/Nasal

Once more feel free to provide comments (I am interested in particular if somebody experiences performance issues in order to check the minimum requirements)

Cheers

Enrique
Flying toaster
 
Posts: 353
Joined: Wed Nov 29, 2006 6:25 am
Location: Toulouse France

Re: F-20 development

Postby Mercury » Mon Jan 07, 2013 11:15 pm

Performance issues are not found, in KSFO almost 60 fps (MiG-29 - 30-40 fps). But FGFS "swears" on impossibility to find part of objects, and also on an undefined name in F-20.nas.
Code: Select all
Could not find at least one of the following objects for animation: 'Slat1'
Could not find at least one of the following objects for animation: 'Slat1.001'
Could not find at least one of the following objects for animation: 'Slat2'
Could not find at least one of the following objects for animation: 'Slat2'
Could not find at least one of the following objects for animation: 'LoSpdBrk', 'LoSpdBrkJack1', 'LoSpdBrkJack2', 'LoSpdBrkJack11', 'LoSpdBrkJack22'
Could not find at least one of the following objects for animation: 'CanopyInside'
Could not find at least one of the following objects for animation: 'ReheatL', 'InNozzleL', 'InNozzleLOpen'
Could not find at least one of the following objects for animation: 'ReheatR', 'InNozzleR', 'InNozzleROpen'
Could not find at least one of the following objects for animation: 'HeatBlurL'
Could not find at least one of the following objects for animation: 'HeatBlurR'
Could not find at least one of the following objects for animation: 'ShockWave'
Nasal runtime error: undefined symbol: canvas
  at /home/sergey1/cprogs/fgfs28/flightgear/lib/FlightGear/Aircraft/F-20/Nasal/enginstr.nas, line 9
Nasal runtime error: undefined symbol: canvas
  at /home/sergey1/cprogs/fgfs28/flightgear/lib/FlightGear/Aircraft/F-20/Nasal/clock.nas, line 9
Nasal runtime error: undefined symbol: canvas
  at /home/sergey1/cprogs/fgfs28/flightgear/lib/FlightGear/Aircraft/F-20/Nasal/HUD/HUD_main.nas, line 14
Initializing Liveries
loading scenario 'refueling_demo'
Nasal runtime error: undefined symbol: updateEngineInstrument
  at /home/sergey1/cprogs/fgfs28/flightgear/lib/FlightGear/Aircraft/F-20/Nasal/f-20.nas, line 70
User avatar
Mercury
 
Posts: 39
Joined: Wed Jan 18, 2012 8:58 am
Callsign: Merc317

Re: F-20 development

Postby Flying toaster » Tue Jan 08, 2013 5:23 am

Oops...

The missing objects are due to the fact that I started the animations from my F-14 model (the bits missing come from it).
I am using a recent git version to test the F-20, so basically all the trouble you get from Nasal is because you are using a version where Canvas was not yet standard feature.

So, definitely, the version needs to be strictly superior to 2.8

Thanks for the feedback :)
Flying toaster
 
Posts: 353
Joined: Wed Nov 29, 2006 6:25 am
Location: Toulouse France

Re: F-20 development

Postby Flying toaster » Mon Jan 14, 2013 9:17 pm

At long last I managed to have a FDM do a flat spin :) (watch alpha, it was spining something like 90deg/s, don't know how to make a youtube capture :oops: )
Image

Two limitations though... Need to have the drop tanks heavy, and it recovers all too easily (the F-5/F-20 are supposed to be hard to depart, but once departed hard to recover... according to a NASA report)
The "bare" FDM is coming along nicely, it is quite lively (rearward Cg) and yet tractable.

The modifications are NOT in the alpha package yet (no need to download again for the moment)

I'll post an update later if people want to test the improved FDM (without Control Augmentation System)
Flying toaster
 
Posts: 353
Joined: Wed Nov 29, 2006 6:25 am
Location: Toulouse France

Re: F-20 development

Postby Flying toaster » Tue Jan 22, 2013 7:24 pm

Uploaded the updated model (follow the link)

Change log (so that you are not obliged to download if you don't want yet another broken package ;) )
- Improved pitch model (more responsive, will only be forced to depart with wing tanks loaded)
- Customized GUI for loadout (giving a try at canvas)
- External fuel tanks (even though they are not used for the time being, so you are pretty much stuck to short range flying, only much heavier)
- Hidden the Canvas HUD so that the F-20 is more usable until the aforementionned HUD is fixed
- Added 3D model for smoke generators (smoke generation not yet available)
- Relocated the start cart model to avoid interference with drop tanks (external power not modelled yet )

As you see the list of TODO things is still a whole lot larger than that of things already done, in particular no Control Augmentation yet...
I need to have a proper "nasty" bare aircraft before I put all the nice features in, such as a yaw damper, auto-trim or g-demand pitch law (all of which were available in the real deal).
For the time being, be advised it can make you sea sick in yaw, and it is a pain in the neck to trim correctly (especially with autoflaps kicking in) :)

As a side note, the loadout menu uses a dirty hack (transparent procedure drawn objects for the "click sensitive"areas instead of SVG objects).
Hopefully this will be fixed but it is still workable if you want to give it a try ...

Cheers
Flying toaster
 
Posts: 353
Joined: Wed Nov 29, 2006 6:25 am
Location: Toulouse France

Re: F-20 development

Postby Flying toaster » Sat Jan 26, 2013 9:05 pm

Note : I just uploaded the latest git codebase. The Dialog class of the gui.nas in Canvas has been renamed Window, so I would expect most of the people who tried out the F-20 with the new menu should get a nasal error and no menu.
Will be fixed in next issue ;)
Flying toaster
 
Posts: 353
Joined: Wed Nov 29, 2006 6:25 am
Location: Toulouse France

Re: F-20 development

Postby daveculp » Thu Feb 28, 2013 5:09 am

What a pretty airplane. I bet it was a thrill to fly. :D
User avatar
daveculp
 
Posts: 503
Joined: Sun Feb 24, 2013 1:50 am
Location: Las Vegas, USA
Callsign: DCulp
Version: 2017.3.1
OS: Ubuntu 17.10

Re: F-20 development

Postby Flying toaster » Sun Mar 03, 2013 4:30 pm

daveculp wrote in Thu Feb 28, 2013 5:09 am: I bet it was a thrill to fly. :D


I guess it was ...
On the down side, two of the three flying prototypes crashed due to g-induced loss of consciousness.
After the "lively" demo routine (involving high g turns) both pilots passed out (on separate occasions) with the consequence you can imagine.
Also true, a world tour to demonstrate a fighter that has a hard time selling does also put some stress on the bodies of the flight crews.
I think the same kind of mishaps occurred early on in the F-16 fleet. Even though centrifuge tests have been common practice since the early 50's (at least for astronauts),
we had to wait for the early 80's to see aircraft capable of sustaining high g's routinely, and being able to do so without fear of loss of control.
So yes those accidents can be blamed on new aircraft that were breakthroughs in terms of maneuverability. But I guess that didn't bring any relief to the families of those pilots. RIP.

On a lighter note, I was quite surprised when a test pilot, who had hands on experience on the T-38, told me that it was unpleasant to fly (actually I think the word was "nasty").
I must say that I have always found the Northrop T-38/F-5 family to be extremely elegant (much more than the F-104 for instance), and as the saying goes, "if the aircraft is pretty..."
I was all the more surprised that this is not the kind of handling qualities you would expect from a trainer.
Too bad he did not elaborate on the subject. I guess those LERXes added to the F-5 of all sorts did have their usefulness after all.
Kind of strange they were never retrofitted to the Talon fleet though (even though the wings were re-manufactured).

Well enough babbling and back to work...

Cheers
Flying toaster
 
Posts: 353
Joined: Wed Nov 29, 2006 6:25 am
Location: Toulouse France

Re: F-20 development

Postby daveculp » Sun Mar 03, 2013 4:59 pm

I found the T-38 a joy to fly, once you got used to the speeds. It was stable, yet sensitive to the touch. Above 400 knots it was very sensitive and easily over-g'd (I over-g'd one on a check-ride once. Oops). One thing you had to get used to, as with any tiny-winged airplane, to get a useful turn rate you had to bank and pull. No pull, no turn. Learning to land was tricky at first. Speeds were high, buffet was normal, and you land in a crab.
User avatar
daveculp
 
Posts: 503
Joined: Sun Feb 24, 2013 1:50 am
Location: Las Vegas, USA
Callsign: DCulp
Version: 2017.3.1
OS: Ubuntu 17.10

Re: F-20 development

Postby Flying toaster » Sun Mar 03, 2013 5:42 pm

Well I guess I will come back to you for flight testing my FDM then :)

Now that you mention it, I seem to remember he felt quite uncomfortable with the landing and particularly with the landing speed ...
I don't remember him mentioning turn behavior though.
Actually now I am thinking about modeling a few things such as tail blanking at high alpha (spin, spin ...), and loss of roll effectiveness in the same regime.
While there is not much wing sweep to create early tip stall, I would still expect the taper ratio to degrade roll response at high alpha (even though with little pitch up effect).

Your inputs/thoughts are always welcome

Cheers
Flying toaster
 
Posts: 353
Joined: Wed Nov 29, 2006 6:25 am
Location: Toulouse France

Re: F-20 development

Postby Flying toaster » Mon Sep 09, 2013 8:00 pm

I am sort of late ... But the F-20 is updated somehow so I thought I might put a fresh tarball online.
The canvas HUD is working (if you have a 2.12rc on your computer).
The fuel system is working
External lighting is mostly working (rembrandt)
Altimeter has been added but that really is no biggy.
Bad news : the loadout menu is broken. Using the del() method on the window destroys the off-screen canvas (since it is attached to no texture) so I tried to work around that ... to no avail.
Any insight on the issue is welcome :D
A screeny of an AI interception
Image
And the link (if you are reluctant to go up)
F-20A alpha version
Flying toaster
 
Posts: 353
Joined: Wed Nov 29, 2006 6:25 am
Location: Toulouse France

Re: F-20 development

Postby TheTom » Tue Sep 10, 2013 3:10 pm

Flying toaster wrote in Mon Sep 09, 2013 8:00 pm:Bad news : the loadout menu is broken. Using the del() method on the window destroys the off-screen canvas (since it is attached to no texture) so I tried to work around that ... to no avail.

I will have a look at this problem.

If there is a problem with the Canvas please put a message it into the Canvas forum or send me a message - possibly with a description how it worked/is expected to work and what it actually does. Also a link to a full example/aircraft would be nice :-)
TheTom
 
Posts: 321
Joined: Sun Oct 09, 2011 10:20 am

Re: F-20 development

Postby Flying toaster » Tue Sep 10, 2013 3:28 pm

TheTom wrote in Tue Sep 10, 2013 3:10 pm: Also a link to a full example/aircraft would be nice :-)

Check just the post above ... right below the screenshot ... :D
Flying toaster
 
Posts: 353
Joined: Wed Nov 29, 2006 6:25 am
Location: Toulouse France

PreviousNext

Return to Aircraft

Who is online

Users browsing this forum: MSN [Bot] and 1 guest