Board index FlightGear Development Aircraft

Looking for aircraft with 'paper cut' bugs

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

Looking for aircraft with 'paper cut' bugs

Postby bugman » Sun Apr 26, 2015 4:26 pm

Split off from the topic Published recommendations, guidelines and rules.


Jabberwocky wrote in Sat Apr 18, 2015 4:03 pm:Well, since there is no defined quality standard for the planes and we can't even pull one out of the hat, refusal of a plane by a person who tried to get himself kind of an exclusive gatekeeper position can only be considered arbitrary, especially given, that many planes in the "official" repository are not even flyable.


Hi Jabberwocky,

You mention official aircraft which are not in a flyable state. I'm interested in going through the entire fleet to create a new base level that all aircraft will be flyable. I'm not talking about fixing clunky FDMs or incomplete parts, but really basic maintenance - aircraft which start with tonnes of errors, and especially aircraft that cannot be started. If you had a list or could point out a few, that'd be awesome! I know there is tonnes of information in old forum posts about this, but for me it's quicker just to start each up one by one and test. If you do have a good list, maybe it would be good to start a new topic.

Cheers,

Edward
Last edited by Johan G on Thu May 07, 2015 1:51 pm, edited 1 time in total.
Reason: Split off from another topic
bugman
Moderator
 
Posts: 1712
Joined: Thu Mar 19, 2015 9:01 am
Version: next

Re: Published recommendations, guidelines and rules

Postby Jabberwocky » Sun Apr 26, 2015 6:54 pm

Well, starting up with missing objects, nasal runtime errors of different severities:

787-8
all airbuses as far as I tried
most 737s
717
some versions of the MD80s, but I have no ist at hand

Balance problems that lead to plane crashes
some MD83s (not mine, took care of that)

Underpowered and almost unable to take off
almost all of Helijah's with the standard FDM

Autopilots with extreme crashibility
All 4mot bomber from Helijah (haven't looked at the smaller ones yet)

Most other planes not maintained since 3.0 and before

So, this is not a complete list, just the ones on top of my head. But it gives you an idea where to look
Jabberwocky
Retired
 
Posts: 1319
Joined: Sat Mar 22, 2014 7:36 pm
Callsign: JWOCKY
Version: 3.0.0
OS: Ubuntu 14.04

Re: Published recommendations, guidelines and rules

Postby KL-666 » Thu May 07, 2015 12:05 am

What kind of criterion is this?
I'm not talking about fixing clunky FDMs or incomplete parts, but really basic maintenance - aircraft which start with tonnes of errors, and especially aircraft that cannot be started.

So if you make a plane without even an attempt to an fdm, it can stay as long as it does not produce errors? Well, i guess if you do not attempt an fdm it will not produce errors for sure.

To me a *flight* simulator is still about flight, equals fdm. On that basis i can very quickly kick out 90-95% of the non flying aircraft on the main download page. Including all the planes that grossly exaggerate their fdm score.

Kind regards, Vincent
KL-666
 
Posts: 784
Joined: Sat Jan 19, 2013 1:32 pm

Re: Published recommendations, guidelines and rules

Postby Thorsten » Thu May 07, 2015 5:02 am

On that basis i can very quickly kick out 90-95% of the non flying aircraft on the main download page. Including all the planes that grossly exaggerate their fdm score.


I would basically agree with the sentiment that quality control is lacking. Personally, I'd like to see a fair application of the rating criteria to all aircraft - it's easy to come up with examples where this is drastically misused and ratings are exaggerated.

However... at this point, think about the enormous friction this would cause - people taking an application of the rating to their work personally, arguments about what is a realistic FDM and how to judge it (think the YaSim vs. JSBSim debate magnified by a factor ten)... if you have any idea how to arrive at a realistic quality control without these side effects, let me know.
Thorsten
 
Posts: 11118
Joined: Mon Nov 02, 2009 8:33 am

Re: Published recommendations, guidelines and rules

Postby bugman » Thu May 07, 2015 6:47 am

KL-666 wrote in Thu May 07, 2015 12:05 am:What kind of criterion is this?
I'm not talking about fixing clunky FDMs or incomplete parts, but really basic maintenance - aircraft which start with tonnes of errors, and especially aircraft that cannot be started.

So if you make a plane without even an attempt to an fdm, it can stay as long as it does not produce errors? Well, i guess if you do not attempt an fdm it will not produce errors for sure.

To me a *flight* simulator is still about flight, equals fdm. On that basis i can very quickly kick out 90-95% of the non flying aircraft on the main download page. Including all the planes that grossly exaggerate their fdm score.

Kind regards, Vincent


Hi Vincent,

To understand the minor changes I am making within the FlightGear project - both to the source code repositories and the official aircraft repository - have a look at the Paper cut bug concept. I hope this adequately describes my current clean up efforts (and that quote of mine).

Regards,

Edward
bugman
Moderator
 
Posts: 1712
Joined: Thu Mar 19, 2015 9:01 am
Version: next

Re: Published recommendations, guidelines and rules

Postby KL-666 » Thu May 07, 2015 12:39 pm

if you have any idea how to arrive at a realistic quality control without these side effects, let me know.


This is a very tough question on which i will not claim to have the answer. But let's say i have somewhat an idea of the direction to look for an answer.

I think it is not feasible to try to say: this 737 really flies like a 737. Then we need pilots for every model in fg. But it is feasible to judge if a plane behaves somewhat like a plane or like nothing or is totally nature defying. To test this in a reasonable objective manner a checklist could be used. If such list is explanatory enough, almost anyone could test a plane.

When i test a plane i already use such a list in the back of my mind.

- climb: Does the plane gradually get a harder time climbing when getting in thinner air? I have had planes that climed like mad and whould not stop climbing. That got me into "space" floating at fl 2000. Others climbed like mad and suddenly just stopped climbing at a certain fl.

- desc: Is the plane able to desc without speeding up? I have had planes that speeded up at the slightest angle of desc (without power of course).

- config change: Does anything reasonable happen with a config change? For flaps that would be in the first settings more lift than drag, and in the last few settings a lot drag for little extra lift. I have had planes where nothing happened. I could just flick through all flap settings without any effect. Another one put all the drag in the first 2 settings and after that nothing more.

- speed: Are take off and landing speeds reasonable? A jetliner that needs way over 200 kts to get off the ground is somewhat unreasonable. Same goes for landing. You must be able to get near 140 kts without stalling. Also cruise speed is interesting. Some subsonic jets happily fly faster than the speed of sound in level flight.

- balance: Check the trim setting in cruise. Is it near center, then it is fine. Is it fully at the end of the envelope, then it is rather bad.

- cross wind: Is the rudder effective enough to counter cross wind? Does the plane want to vane into the wind at high speeds on the take off run? Some developers have just glued the wheels to the ground so there is no such effect whatsoever.

If such a checklist could be agreed upon then there still can be disagreement, but i think it can be solved quite quickly. If a tester says there is no such-and-such effect and the developer says there is, but very little, then the tester or anyone else can verify that. There is no way to judge how big the amount of effect should be. Only if it is there or not.

About sensitivities of developers i can say: I am not thinking about throwing away a plane all together. Just removing it from being advertized on the main download page. It should still sit in it's version control repository waiting to be fixed so it can be advertized again.

Kind regards, Vincent
KL-666
 
Posts: 784
Joined: Sat Jan 19, 2013 1:32 pm

Re: Published recommendations, guidelines and rules

Postby Johan G » Thu May 07, 2015 1:59 pm

KL-666 wrote in Thu May 07, 2015 12:39 pm:It should still sit in it's version control repository waiting to be fixed so it can be advertized again.

As much as I would like aircraft to be in a working condition and presentable, I still have to ask this maybe provoking question:

Who would get an 'itch' to 'scratch' if almost no one are not aware of the itch? :wink:
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: 5530
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: Looking for aircraft with 'paper cut' bugs

Postby KL-666 » Thu May 07, 2015 2:30 pm

I think there is not an unawareness of where the planes are. First of all the plane developer has a stake in wanting to fix the fdm. If he can't himself he might be teaming up with someone that can. Or that someone has already started that work, being annoyed about the disappearance of the plane from the main download page.

In my mind the plane repositories are centralized, so findable for developers and advanced users that are on the hunt for new aircraft. The aircraft should at least be downloadable to be tested, and if someone thinks he can fix it, then he should get a chance to try.

All in all i am not worried about no one knowing where a plane has gone. It is at the place where it has always been (for development)

Kind regards, Vincent.
KL-666
 
Posts: 784
Joined: Sat Jan 19, 2013 1:32 pm

Re: Published recommendations, guidelines and rules

Postby Bomber » Thu May 07, 2015 2:49 pm

:|
KL-666 wrote in Thu May 07, 2015 12:05 am:
To me a *flight* simulator is still about flight, equals fdm. On that basis i can very quickly kick out 90-95% of the non flying aircraft on the main download page. Including all the planes that grossly exaggerate their fdm score.

Kind regards, Vincent


I remember saying something very similar about 4 years ago
"If anyone ever tells you anything about an aeroplane which is so bloody complicated you can't understand it, take it from me - it's all balls" - R J Mitchel
Bomber
 
Posts: 1934
Joined: Fri Dec 14, 2007 7:06 pm
OS: Windows XP and 10

Re: Looking for aircraft with 'paper cut' bugs

Postby Bomber » Thu May 07, 2015 2:53 pm

Oh... The solution is a flight models rating should only be assigned by the flight modellin community

And the quality of 3d cockpits by 3d modellers and skins by 2d artists...
"If anyone ever tells you anything about an aeroplane which is so bloody complicated you can't understand it, take it from me - it's all balls" - R J Mitchel
Bomber
 
Posts: 1934
Joined: Fri Dec 14, 2007 7:06 pm
OS: Windows XP and 10

Re: Looking for aircraft with 'paper cut' bugs

Postby Thorsten » Thu May 07, 2015 2:55 pm

If such a checklist could be agreed upon then there still can be disagreement, but i think it can be solved quite quickly.


We do have an agreed-upon checklist for rating FDM. A plane with 200 kt speed to takeoff is unlikely to comply with published Pilot Operating Handbook numbers, which would automatically make the rating two or lower for instance.

The problem is - what if someone doesn't care?

Take HerbyW's Space Shuttle rated with 4 stars in the FDM. This in theory should make it comply with 4: FDM matches PoH in 90% of configurations.

I happen to have the PoH, and in fact I also have 1800 (!) pages of wind tunnel data for the shuttle (NASA puts it all into public domain), so I know the aerodynamics of the real thing quite well by now. The real thing is yaw unstable above Mach 2 - HerbyW's is stable all the way. The real shuttle has pronounced cross-correlations between elevon deflection and roll/yaw commands. HerbyW's has none. The real thing is controlled by several stick-controls-rate digital autopilot modes firing thrusters in orbit - HerbyW's version has a modified wing-leveler of a cargo plane commanding the thrusters. I could go on for a few more pages.

A realistic rating for the FDM would be a 1 (it's just aeromatic output) or even 0 (the controls are really for another plane).

As far as the facts and the comparison of FDM to reality go, this is a very clear cut case in which I can cite (literally) hundred pages of evidence.

Yet, imagine what would happen if I would alter the rating on the wiki page. A simple note added to the effect that this is a third party work and distinct from the orbiter now residing in FGAddon caused a veritable outrage in the forum. You will find that unfortunately facts and clear evidence gets you only so far here, and are quickly discounted in the face of rumor, accusations, and such like. I invite you to make the test and just alter a few ratings in the wiki to their real values :-)

The problem is not having an agreed-upon checklist - we have that. The problem is also not establishing which aircraft should have what rating based on this list - your test criteria would all push a plane to a rating of two or lower, at which point it would be clearly marked as 'not ready'. The problem is really enforcing that the checklist is used fairly and that authors actually apply it rather than abuse it.
Thorsten
 
Posts: 11118
Joined: Mon Nov 02, 2009 8:33 am

Re: Looking for aircraft with 'paper cut' bugs

Postby Thorsten » Thu May 07, 2015 2:57 pm

Oh... The solution is a flight models rating should only be assigned by the flight modellin community


Be careful what you argue here Simon - you can now no longer claim I'm not part of the flight modeling community :-)
Thorsten
 
Posts: 11118
Joined: Mon Nov 02, 2009 8:33 am

Re: Looking for aircraft with 'paper cut' bugs

Postby IAHM-COL » Thu May 07, 2015 3:22 pm

@bugman

Dear Ed.
I have a paper cut bug you can pass yourself over the FGADDon

Several sound files are stereo. It seems that FGFS sound engine does not understand this, and disregards the sound file outputting a verbose error to the log. Easy fix is to change these files to mono (by replacing them). That's quite a paper cut thingy there.

In the other hand a more complex fix would be altering the FGFS sound rendering engine to use the stereo, and I don't know if this should be actually a more preferable fix, but "that's there"

Best, and thanks for all your paper cut cleaning.

IH-COL
If we gave everybody in the World free software today, but we failed to teach them about the four freedoms, five years from now, would they still have it? Probably not, because if they don’t recognise their freedoms, they’ll let their freedoms fall
User avatar
IAHM-COL
Retired
 
Posts: 4064
Joined: Wed Aug 08, 2012 5:40 pm
Location: Homey, NV (KXTA) - U.S.A
Callsign: HK-424D or ICAO4243
Version: 3.7-git
OS: Linux

Re: Looking for aircraft with 'paper cut' bugs

Postby Bomber » Thu May 07, 2015 3:23 pm

Thorsten wrote in Thu May 07, 2015 2:57 pm:
Oh... The solution is a flight models rating should only be assigned by the flight modellin community


Be careful what you argue here Simon - you can now no longer claim I'm not part of the flight modeling community :-)


A very very very junior members with sod all experience
"If anyone ever tells you anything about an aeroplane which is so bloody complicated you can't understand it, take it from me - it's all balls" - R J Mitchel
Bomber
 
Posts: 1934
Joined: Fri Dec 14, 2007 7:06 pm
OS: Windows XP and 10

Re: Looking for aircraft with 'paper cut' bugs

Postby KL-666 » Thu May 07, 2015 3:30 pm

Hi Thorsten,

Of course everything stands or falls with checking a checklist occasionally (or often). The problem with the list you mention is that it is based on putting in numbers in the fdm. Who on earth is going to check if all the numbers are put in right?

And there is another problem with that list. There are developers that claim (and probably did) put in all the numbers. But if i touch the plane for a few minutes i quickly find out that it does not behave anything near a flying craft.

On the other hand there are craft that probably can only be checked by numbers, like a space craft. So the numbers checklist is good. But i strongly believe there should be a practical test too for craft whose behaviour can be known.

Kind regards, Vincent
KL-666
 
Posts: 784
Joined: Sat Jan 19, 2013 1:32 pm

Next

Return to Aircraft

Who is online

Users browsing this forum: No registered users and 2 guests