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 development

Postby Johan G » Wed Dec 14, 2016 9:34 pm

Flying toaster wrote in Mon Dec 12, 2016 6:53 pm:Actually I checked out that reference in doing the smokewinder.

Great! :D

Also, I have to say that as I saw the nice video from the 26/27 November Airshow 2016 I really got to appreciate how much smoke generators add to the impression of performing aircraft. I am definitively looking forward to see what performances some pilots may come up with.
Low-level flying — It's all fun and games till someone looses an engine. (Paraphrased from a YouTube video)
Improving the Dassault Mirage F1 (Wiki, Forum, GitLab. Work in slow progress)
Johan G
Moderator
 
Posts: 5545
Joined: Fri Aug 06, 2010 5:33 pm
Location: Sweden
Callsign: SE-JG
IRC name: Johan_G
Version: 3.0.0
OS: Windows 7, 32 bit

Re: F-20 development

Postby Flying toaster » Wed Dec 21, 2016 9:33 pm

Here is an updated version (hopefully it works).
It is still very much WIP but you will find among changes :
- Almost complete 3D models (textures to be improved)
- New liveries
- Functional smokewinders
- Integral cockpit lighting
- New vortex effects
- Some loadouts (don't ask me to make them active, I will not implement a weapons systems... like never)
- The begining of an autopilot for the C, addition of a radar altimeter, color DDI ...
- a few bug fixes in instrumentation
- Baro setting
- reversion to the standard menu system

Known bugs/limitations are :
- loadouts can take some time to appear (like minutes) due to what seems to be a bug with attached submodels. Most of the time the thing sorts itself out
- The AGL autopilot is a bit wild. Still needs a lot of tuning. Also don't engage below 1000ft if you don't want to run into terrain over mountains . And there is no instinctive disconnect or proper documentation (i.e. use at your own risk)
- Might need to hike drag a very little bit (range and max mach slightly optimistic)

TODO : waaaayyyyy too long to write down :(

The documentation is still sketchy.
I am releasing because holiday season is coming and I most probably won't be resuming work until 2017. But I thought it might have enough goodies for people to want to try (and give feedback) anyway

enjoy

Enrique

Google Drive download link
Flying toaster
 
Posts: 353
Joined: Wed Nov 29, 2006 6:25 am
Location: Toulouse France

Re: F-20 development

Postby Flying toaster » Fri Dec 23, 2016 4:16 pm

Has anybody experienced problems downloading or using ?
Flying toaster
 
Posts: 353
Joined: Wed Nov 29, 2006 6:25 am
Location: Toulouse France

Re: F-20 development

Postby Necolatis » Fri Dec 23, 2016 5:25 pm

Works fine for me. Get this though:

Failed to load model: Model file not found: '../../../../Aircraft/f16/Models/f16.ac'
at C:/Program Files/FlightGear 2016.4.3/data/AI/Aircraft/f16/Models/f16-Wingman.xml
from:C:/Program Files/FlightGear 2016.4.3/data/AI/Aircraft/f16/Models/f16-Wingman.xml
"Airplane travel is nature's way of making you look like your passport photo."
— Al Gore
User avatar
Necolatis
 
Posts: 2030
Joined: Mon Oct 29, 2012 12:40 am
Location: EKOD
Callsign: Leto
IRC name: Neco
Version: 2019.1.2
OS: Windows 10

Re: F-20 development

Postby Flying toaster » Fri Dec 23, 2016 6:34 pm

Hum yes, I do get it too. I thought it was a problem with my installation. I will investigate. There is also a nasty message about the throttle property not being bound... weird
Flying toaster
 
Posts: 353
Joined: Wed Nov 29, 2006 6:25 am
Location: Toulouse France

Re: F-20 development

Postby rominet » Fri Dec 23, 2016 9:34 pm

Necolatis wrote in Fri Dec 23, 2016 5:25 pm:Failed to load model: Model file not found: '../../../../Aircraft/f16/Models/f16.ac'
at C:/Program Files/FlightGear 2016.4.3/data/AI/Aircraft/f16/Models/f16-Wingman.xml
from:C:/Program Files/FlightGear 2016.4.3/data/AI/Aircraft/f16/Models/f16-Wingman.xml

This very much looks like a problem that should be fixed in FG 2017.1.0.
rominet
 
Posts: 557
Joined: Sat Nov 01, 2014 1:33 pm
Callsign: F-KATS
Version: Git next
OS: Debian GNU/Linux

Re: F-20 development

Postby abassign » Sun Dec 25, 2016 6:50 pm

Image

I my blog ( https://www.facebook.com/pg/flightgearitalia/posts/?ref=page_internal ) I posted some pictures of this beautiful F20, I hope the plane's development can proceed further because I think is it one of the best jet fighters of FGFS.
The flight is about 12,000 ft on the Po Valley in background the Mont Blanc group profile.
abassign
 
Posts: 823
Joined: Mon Feb 27, 2012 5:09 pm
Location: Italy (living 5 Km from airport LIME)
Callsign: I-BASSY
Version: 2018.3
OS: Linux Mint 19. x

Re: F-20 development

Postby Flying toaster » Wed Dec 28, 2016 1:07 pm

Thanks for the comment,
Nice shots on facebook.
As a minor help, you also have backlight on the keypad below the HUD (the knob for it is oddly placed below the armament panel).
As far as some low polygons on the model, it is due to the fact that once upon a time, low polygon count was a nice way to save CPU.
It truth be told, that no longer is the case (I mean, the thing runs at full frame rate with the high polycount cockpit on all the time). I may take some time to improve detailing. I will also look into the textures for the intakes to have them look less bright (normally it should be shadow obscuring them)
My attention is going to be turning to lighting, I miss some external lights, and yes the afterburner glow. That would be easy to implement in rembrandt and I will implement it, but I also need to give some consideration on performance. That being said, I recently changed the core hardware of my computer (i.e. not my GPU) and seen dramatic improvement on the performance of Rembrandt. Could it be that the render buffers are created in the central memory instead of the GPU (that would be a serious performance bottleneck indeed) ? I must admit that aircraft look better with Rembrandt (although I need to find a way to adjust brightness for textures not to look too "milky")
I am trying to replicate dynamic lighting in the direct rendering pipeline for those who are too "rembrandt challenged"
I am very interested in people reporting any issue with frame rate or other performance issues so that I can adjust to a level of detail that makes it playable for most people.
I also need to have the bleed/ECS working but that should be pretty straightforward...
Also radio communications and ILS are on my todo list (have to check how FG handles channels and UHF communications)
Then I will work on GUI to have cold/warm start, checklists ...
After that I think it will be ready for inclusion in the FGDATA (for which I will need help from official comitters).
Cheers
Enrique
Flying toaster
 
Posts: 353
Joined: Wed Nov 29, 2006 6:25 am
Location: Toulouse France

Re: F-20 development

Postby Thorsten » Wed Dec 28, 2016 2:47 pm

That being said, I recently changed the core hardware of my computer (i.e. not my GPU) and seen dramatic improvement on the performance of Rembrandt. Could it be that the render buffers are created in the central memory instead of the GPU (that would be a serious performance bottleneck indeed) ?


It's been discussed a number of times that Rembrandt more frequently runs into CPU bottlenecks than GPU bottlenecks for many people. Also, the performance scaling seems to be very odd and is not quite understood - for instance on a 3000 Euro gaming laptop I get a poor 15 fps in Rembrandt with decent shadows and an otherwise unassuming scene - while I can render ALS at 60 fps with highest shader quality in the same situation (and with 8 cores and 8 GB of main and 3 GB of graphics memory I'm not exactly lacking CPU power or memory...)

I must admit that aircraft look better with Rembrandt (although I need to find a way to adjust brightness for textures not to look too "milky")


I'm mildly surprised by this statement (unless you're specifically after dynamic lights) because the main effect code is practically identical but ALS comes with a number of tricks not implemented or possible for Rembrandt (grain texture, rain splashes, diffuse skylight reflection, glass effect, caustics, anti-aliasing, procedural light sprites, flames...).

So, apart from the lights, are you missing anything in particular?
Thorsten
 
Posts: 11191
Joined: Mon Nov 02, 2009 8:33 am

Re: F-20 development

Postby Josh_grtuxteam » Thu Dec 29, 2016 5:22 pm

Flying toaster wrote in Wed Dec 28, 2016 1:07 pm:Thanks for the comment,
..............................That being said, I recently changed the core hardware of my computer (i.e. not my GPU) and seen dramatic improvement on the performance of Rembrandt. Could it be that the render buffers are created in the central memory instead of the GPU (that would be a serious performance bottleneck indeed) ? I must admit that aircraft look better with Rembrandt (although I need to find a way to adjust brightness for textures not to look too "milky") ...........................................
Enrique


I can confirm what Enrique saiys about performance when using Rembrandt, it is is hardware architecture dependant , not GPU dependant but cpu , memory, motherboard dependant.
Before i had to move home i could use an equipment which was answering to the request for running Rembrandt and high detailed models ( the GRTUX ones) with performances never below 40 fps . Only the terrain introduced sometime some strange behaviour .
However that equipment was not that exceptional since built around AMD 8350 was 8 cores and memory at the max authorized freq ( i don't remember the value ).
The graphic Card was a NVIDIA 560 GTX Ti only 1280 MB Memory.

By the way i have the same graphics Card (NVIDIA 560 GTX) within the computer i am working with right now, but within an old Intel desktop , so at the moment no Rembrandt , no ALS :( (that last renderer is worse since i cannot get good colors ALS wants probably a higher graphics card, GPU and memory ).

@Enrique when using Rembrandt, give a try by tweaking the others parameters Rembrandt embedded you could get better effect with your textures.

I do it when i want some better effect inside the cockpit and cabin.

Cheers

Josh
User avatar
Josh_grtuxteam
 
Posts: 107
Joined: Mon Oct 17, 2016 2:58 pm
Location: France
Version: lastStable
OS: Linux OpenSUSE

Re: F-20 development

Postby Thorsten » Thu Dec 29, 2016 7:16 pm

@Enrique when using Rembrandt, give a try by tweaking the others parameters Rembrandt embedded you could get better effect with your textures.


Since you seem to have a hard time to understand this Josh: Please do not set any rendering settings aircraft-side (unless requested by the user) - if they're auto-saved, you generate very confusing behavior in other aircraft. Just leave the rendering dialog to the user. Thanks.
Thorsten
 
Posts: 11191
Joined: Mon Nov 02, 2009 8:33 am

Re: F-20 development

Postby Josh_grtuxteam » Thu Dec 29, 2016 8:45 pm

Since you seem to have a hard time to understand this Thorsten: Please don't force model developer to make their models at a lower quality , by ignoring the existing FG parameters which could improve the quality, won't a end user requesting the best ?

By the way were is the problem, is it only your way to demonstrate that Rembrandt is not usable ? , sorry i don't understand.

Josh
User avatar
Josh_grtuxteam
 
Posts: 107
Joined: Mon Oct 17, 2016 2:58 pm
Location: France
Version: lastStable
OS: Linux OpenSUSE

Re: F-20 development

Postby psadro_gm » Thu Dec 29, 2016 8:54 pm

hmm, absolutely not: If you are setting rendering parameters to make things look better aircraft side, then people with lower powered machines may then have unplayable framerates.
This is why we have the rendering dialogs. Let the user choose what is best.
8.50 airport parser, textured roads and streams...
psadro_gm
 
Posts: 751
Joined: Thu Aug 25, 2011 2:23 am
Location: Atlanta, GA USA
IRC name: psadro_*
Version: git
OS: Fedora 21

Re: F-20 development

Postby Josh_grtuxteam » Thu Dec 29, 2016 9:07 pm

Hi, psadro_gm

OOOPS, I didn't say that,
In order to avoid any misunderstanding, i don't force Rembrandt, which would be criminal.
BUT if a user is using Rembrandt he could discover some ugly side effect, under some conditions the developer can compensate it .
I do it.

In such case there is not any significant loss of performances


cheers

Josh

@Enrique, i apologize i should have guessed my suggestion would have been hijacked , polluting your topic.
User avatar
Josh_grtuxteam
 
Posts: 107
Joined: Mon Oct 17, 2016 2:58 pm
Location: France
Version: lastStable
OS: Linux OpenSUSE

Re: F-20 development

Postby abassign » Thu Dec 29, 2016 10:05 pm

As human resources are always a few, I prefer to work only on a rendering engine. ALS currently the most developed and therefore I believe that it configure the plane only on the engine type.
Thorsten has its own vision of things that often not coincide with mine, but it is working very well to the ALS, (pity very few people are able to follow him!) but the situation on Rembrandt engine is definitely the worst (at present no one maintainer!).
I wanted to do tests for the F20, to understand the difference between the three types of engine in use in Fgfs:

Basic engine: 50-40 fps
ALS engine: 40-25 fps
Rembrandt engine: 30-20 fps


The graphical configuration with the highest quality, the video 1024x104, configured trees with density 2.6, No clouds.

But the problem with all these configurations is that the CPU does not average more than 120-130% (I set the multi-threaded option). If Rembrandt does not use a separate thread, I get the impression that will always be presents this gap.
I would be curious that Rembrandt was run on a separate thread, but certainly this change is not easy as it requires a code change, that no one can do or will not do.
ALS does many things directly on the GPU, without going through the CPU, and this is the secret of its efficiency, although my tests point out that, compared with the basic rendering system, ALS is getting slower, because the new features they have a cost of GPU not trivial.

For some time I would like to merge the two projects because each of them has its advantages, but also disadvantages.
When I saw the shadows, the effect of rain, the position lights, the sky etc ... ALS is certainly great, but Rembrandt, looked very well the emission lights, especially the interior lights. The solution of external shadows adopted by ALS is still very poor, in that context Rembrandt is considerably higher. For the interior lights, Thorsten has demonstrated that it is possible to achieve very significant effects. The problem is that, the interior lights are very static, therefore can not have complex effects like with Rembrandt.

Therefore I would focus on the affects ALS, which are still not simple, hoping that in the future someone will maintain Rembrandt and integrate it with ALS.
abassign
 
Posts: 823
Joined: Mon Feb 27, 2012 5:09 pm
Location: Italy (living 5 Km from airport LIME)
Callsign: I-BASSY
Version: 2018.3
OS: Linux Mint 19. x

PreviousNext

Return to Aircraft

Who is online

Users browsing this forum: No registered users and 2 guests