Board index FlightGear Multiplayer events

EDDF-Triangle

Virtual fly-ins, fun flies, competitions, and other group events. Find out details of upcoming events, register for competitions, or organize your own tour of a favorite location.

Re: EDDF-Triangle

Postby Lydiot » Sat Dec 22, 2018 5:23 pm

jomo is never wrong
Lydiot
 
Posts: 1016
Joined: Tue Oct 22, 2013 11:50 pm

Re: EDDF-Triangle

Postby AAL545 » Mon Dec 31, 2018 2:11 am

I'm replying to my post from June 2018 regarding landing gear on B777 not showing properly over MP,
FG 2018.3.1 did not fix this issue and someone was supposed to look into it but I haven't seen anymore comments regarding this issue.
My issue is not only with the landing gear not showing properly, there are times when my thrust reversers don't work, could it be that the two are related because the wheels don't
know if they are on the ground (landing gear sunken into the runway).


AAL4955
AAL545
 
Posts: 227
Joined: Tue Aug 08, 2017 5:14 am

Re: EDDF-Triangle

Postby jomo » Mon Dec 31, 2018 11:44 am

AAL545 wrote in Mon Dec 31, 2018 2:11 am:I'm replying to my post from June 2018 regarding landing gear on B777 not showing properly over MP, ...

Have a look e.g. into movie http://www.emmerich-j.de/EDDF/Films/201 ... 59-101.mp4 at time 0:34:15.
There you see very clearly that the front-Wheel is located correctly on the ground.
Also the center-wheels are correctly centered on the GND - BUT: That are 3 wheels in a row that are tilted when in air and not straightened on GND. i.e. only the center-wheel is placed correctly on the GND, the front-wheel is over GND, and thus the Back-wheel is below GND.
--> Thus that is a design-problem of that model - obviously it was forgotten to level the landing-gear on GND. That is not a problem with FGFS but with the tricky gear-design of that model.
Also the thrust reverse is surely a design-problem of that model - not a FGFS general problem!

Thus the changes between FGFS + MP are working fine.
You must contact the model-designer to get changes - I wish you lots of luck for that (and anyhow: There are thousands of models that do not have that problem).
jomo / ATCjomo + EDDFjo + EDDFjo1 + EDDFjo2
ATC at EDDF Fr,Sa,Su,We from 20:00 to 24:00 CET/MEZ., see http://www.emmerich-j.de
User avatar
jomo
 
Posts: 1000
Joined: Thu Feb 12, 2009 7:46 pm
Location: Mainz, Germany
Callsign: jomo EDDFjo1+2
OS: UBUNTU 18.4

Re: EDDF-Triangle

Postby legoboyvdlp » Mon Dec 31, 2018 1:03 pm

The aircraft's gear looks fine on the user aircraft - however almost certainly the gear tilting on the ground is not transmitted over MP, so the animation breaks and always shows as titled over multiplayer. (That is, the developer was not as stupid as to not straighten them on the ground... ) So the gear tilting simply needs to be added to the list of properties that are transmitted over multiplayer. It may be that there is a limit to how many properties can be transmitted which is why the gear tilting was not added?
User avatar
legoboyvdlp
 
Posts: 7981
Joined: Sat Jul 26, 2014 2:28 am
Location: Northern Ireland
Callsign: G-LEGO
Version: next
OS: Windows 10 HP

Re: EDDF-Triangle

Postby jomo » Mon Dec 31, 2018 4:17 pm

legoboyvdlp wrote in Mon Dec 31, 2018 1:03 pm: That is, the developer was not as stupid as to not straighten them on the ground... ) S

Nobody (except You?) said the developer was stupid! But I am convinced a model-designer should make sure all his proposed functions do work inside the user-environment -- You cannot expect a user to finish up the design.
(If you are the designer, I suggest: Just remove the tilting - and nobody will ever notice!)
jomo / ATCjomo + EDDFjo + EDDFjo1 + EDDFjo2
ATC at EDDF Fr,Sa,Su,We from 20:00 to 24:00 CET/MEZ., see http://www.emmerich-j.de
User avatar
jomo
 
Posts: 1000
Joined: Thu Feb 12, 2009 7:46 pm
Location: Mainz, Germany
Callsign: jomo EDDFjo1+2
OS: UBUNTU 18.4

Re: EDDF-Triangle

Postby AAL545 » Mon Dec 31, 2018 4:43 pm

So the gear tilting simply needs to be added to the list of properties that are transmitted over multiplayer.


First off I am not complaining here, this is a wonderful and complex program.
Regarding adding it to the property in multiplayer, why was it removed?
The reason I say that (too bad I didn't save a film) is in late 2017 everything worked perfect, the landing gear flattened out and the speedbrakes came up.
Someone suggested to do some logging, if there would be a way to do a more pin point logging I'm willing to give it a try.
In the mean time, (for the developers) keep up the good work, the B777 is a wonderful plane to fly.


AAL4955
AAL545
 
Posts: 227
Joined: Tue Aug 08, 2017 5:14 am

Re: EDDF-Triangle

Postby tdammers » Thu Jan 03, 2019 9:04 am

jomo wrote in Mon Dec 31, 2018 4:17 pm:But I am convinced a model-designer should make sure all his proposed functions do work inside the user-environment


It's not that simple.

"The" user-environment and "the" model do not exist, the situation is more complex than that. We have two FG instances, two copies of (hopefully) the same model, and (up to) two MP servers in between them, plus (up to) 3 network connections. That's up to 9 (nine!) points where things can go wrong.

In a perfect world, the MP servers and network connections would be flawless, and the FG versions and configurations and models on both ends would match perfectly, and then everything would work out just fine. But we don't live in a perfect world.

Now clearly many aircraft authors *are* testing their stuff in an MP environment, especially with a model as popular as the 777 - but with 9 moving parts, it becomes impossible to test against all possible combinations, and it's a bit short-sighted to uncritically blame aircraft model authors.

Oh, also: this is open-source software, users are very much expected to do at least mild amounts of troubleshooting themselves.
tdammers
 
Posts: 391
Joined: Wed Dec 13, 2017 11:35 am
Callsign: NL256
IRC name: nl256

Re: EDDF-Triangle

Postby AAL545 » Sun Jan 06, 2019 3:25 pm

It's not that simple.


I realize that, but as I mentioned in late 2017 everything worked perfectly until after an update of one or more systems,
either the aircraft, FG, multiplayer which gets updated alongside FG, and also Orcam.
Are you telling me there is no way a developer can compare the two?
Besides what we see in the EDDF films there needs to be a debug system (logging) where one can see what gets
sent over MP and what is not.
Again, if this is available then I need someone to lead me in the right direction and show me what to do on my end.


AAL4955
AAL545
 
Posts: 227
Joined: Tue Aug 08, 2017 5:14 am

Re: EDDF-Triangle

Postby wkitty42 » Sun Jan 06, 2019 3:37 pm

AAL545 wrote in Sun Jan 06, 2019 3:25 pm:Again, if this is available then I need someone to lead me in the right direction and show me what to do on my end.

--loglevel=debug

that'll certainly give it to you but you may want to limit the logging to only MP or AI or whatever logging class(es) MP uses... you can specify multiple classes, comma separated, if you need to... http://wiki.flightgear.org/SG_LOG
"You get more air close to the ground," said Angalo. "I read that in a book. You get lots of air low down, and not much when you go up."
"Why not?" said Gurder.
"Dunno. It's frightened of heights, I guess."
User avatar
wkitty42
 
Posts: 9148
Joined: Fri Feb 20, 2015 4:46 pm
Location: central NC, USA
Callsign: wk42
Version: git next
OS: Kubuntu 20.04

Re: EDDF-Triangle

Postby tdammers » Sun Jan 06, 2019 4:24 pm

FWIW, I can replicate the 777 landing gear tilt thing locally. I'm running the 2018.3 release compiled from source, and I have a recent git version of the Seattle 777 installed. Landing gear tilt on any 777, including pilots running protocol version 2, seems to stay in the default (tilted) position throughout, even on the ground. So there does seem to be some sort of issue there. I don't know what it is though - could be different versions of the aircraft model being incompatible, could be that the 777 model simply doesn't transmit the property at all, could be that the 777 uses too many MP props and thus the gear tilt one gets dropped. I don't know. In any case, I'm watching a 777 land at EDDM as we speak, and the gear bogies aren't flattening out, so *something* is clearly wrong, but figuring out what isn't exactly trivial.

On a completely different note; I'm curious what happened yesterday. I tried to fly in, but since I wasn't hailed at 60 nm out as usual, I tried to contact ATC multiple times via MP chat. I never got any response though, so I assumed some sort of bug dropping my chat messages, and after about 30 minutes of repeatedly trying to make contact and switching MP servers, I gave up. Watching the films today, however, I noticed to my astonishment that my messages came through just fine, my callsign appeared in the flight strips section, I was visible on the radar, I just got ignored completely - meanwhile lots of other pilots were serviced over MP chat without issues. A simple "standby", followed by advice at a later, less busy time would have been fine. "Unable MP chat, please use mumble" would have been fine too, I can accept when it gets too busy, and I could have flown elsewhere. But I got absolutely nothing.

There is also some rather unprofessional behavior, which I think is completely unacceptable - regardless whether you want to simulate full realism, or whether you approach this as a fun game, screaming at people is just not OK, especially not when the other person stays perfectly calm and friendly. People make mistakes, and dealing with that in a civilized manner is what makes things fun on the game side of things, and vitally important on the realism side of things. If someone gives a useless flight plan, there's no need to get aggressive, just say "negative, please say ICAO and altitude"; "I don't have time for this, I can't do your routing for you, ..." is just useless clutter on the frequency. If someone doesn't follow advice, just calmly neglect them, there's no need to yell at or insult them, or type a dozen exclamation marks. In hindsight, I'm glad I wasn't on mumble this time, 'cause it was absolutely ghastly. I also saw a bunch of other mistakes on both sides, that weren't dealt with in a professional manner - incorrect (or no) use of proper phraseology, loss of discipline, that kind of thing.

Now, I'm not here to blame anyone, or make any demands; I'd just love to see this wonderful event get back on the rails. Some discipline on both sides, a more forgiving attitude, and maybe some workload-reduction strategies on the ATC side, would be tremendously helpful here, I believe. In the meantime, I'll go scratch myself behind the ears real hard and wonder whether I should try again or just give up on EDDF sessions.
tdammers
 
Posts: 391
Joined: Wed Dec 13, 2017 11:35 am
Callsign: NL256
IRC name: nl256

Re: EDDF-Triangle

Postby Isaak » Sun Jan 06, 2019 6:38 pm

I gave up on visiting EDDF. Never got shouted at myself, and I was a very regular visitor about three to zeven years ago, must have participated over a 100 times (callsign was OO-ISA at that time), but have been around quite a few times where Jomo would out of nowhere start shouting at someone who made a slight mistake and that's just not the type of event I want to be part of. If you promote yourself as an event where Newbies can learn to fly under ATC, be happy to learn them something then, instead of shouting that they are wrong. Now I'm happily flying on VATSIM, I've never met a controller that came in the neighboorhood of jomo's attitude, and I 'm talking of controllers who sometimes control over 50 planes when they e.g. staff Eurocontrol CTR. No more FG EDDF for me, sorry.

On the 777's landing gear: I' m going through a very stressful time now, but if I find some spare time I'll take a look at it and try to fix it in FGAddon.
Want to support medical research with your pc? Start Folding at Home and join team FlightGear!
Isaak
 
Posts: 768
Joined: Sat Jun 04, 2011 3:52 pm
Location: Hamme, Belgium
Pronouns: he, him
Callsign: OO-ISA
Version: next
OS: Windows 10

Re: EDDF-Triangle

Postby tdammers » Sun Jan 06, 2019 9:56 pm

BEL ISAAK wrote in Sun Jan 06, 2019 6:38 pm:On the 777's landing gear: I' m going through a very stressful time now, but if I find some spare time I'll take a look at it and try to fix it in FGAddon.


No pressure, I mainly mentioned this to support the "it's not necessarily 100% the aircraft developer's fault, and you should be more careful blaming people" thing. Not that it wouldn't be appreciated, mind you.
tdammers
 
Posts: 391
Joined: Wed Dec 13, 2017 11:35 am
Callsign: NL256
IRC name: nl256

Re: EDDF-Triangle

Postby AAL545 » Mon Jan 07, 2019 12:23 am

On a completely different note;


Not sure why you guys have to do this in the middle of a topic.
Listen, I appreciate your help here I really do, and you know what, I won't even take it personally if you have an issue with EDDF/ Jomo
even though I fly there almost every weekend, I guess because he's always there but that's my business and your business is your business.
Anyway speaking of business,
could be that the 777 model simply doesn't transmit the property at all

thanks for bringing that up tdammers, now I remember someone mentioning that in the forum here,
the newer MP protocol was changed in how many messages are being sent
why I don't recall but I'll look for that comment and will post it.


AAL4955
AAL545
 
Posts: 227
Joined: Tue Aug 08, 2017 5:14 am

Re: EDDF-Triangle

Postby AAL545 » Mon Jan 07, 2019 12:45 am

Okay this is what I have found.

remember, the MP protocol has fixed sized packets it sends... if a craft tries to send to much, the packets will be truncated... that could easily lead to problems like this...

Posted by kwitty42.
2017.3 Changes
New property /sim/multiplay/transmit-only-generics, when this is set only the position header, all of the sim/multiplay/generic/ and instrumentation/transponder/ are transmitted over MP. This is intended to allow models to use an alternative encoding (such as the Emesary PropertyNotification) to transmit a well known (to the model) list of properties packed into the generic properties.

The maximum size of a string was also increased to 768 bytes (from 128).

Not sure who posted this one!

Using the Emesary Multiplayer Bridge[1][2] is an effective way to transmit more properties using less space[3]; however if the transmit filter property base[4] is also set to say 10000 or higher then only the really important properties will be available to other craft.


Posted by Richard.



AAL4955
AAL545
 
Posts: 227
Joined: Tue Aug 08, 2017 5:14 am

Re: EDDF-Triangle

Postby tdammers » Mon Jan 07, 2019 7:44 am

AAL545 wrote in Mon Jan 07, 2019 12:45 am:Okay this is what I have found.

remember, the MP protocol has fixed sized packets it sends... if a craft tries to send to much, the packets will be truncated... that could easily lead to problems like this...

Posted by kwitty42.
2017.3 Changes
New property /sim/multiplay/transmit-only-generics, when this is set only the position header, all of the sim/multiplay/generic/ and instrumentation/transponder/ are transmitted over MP. This is intended to allow models to use an alternative encoding (such as the Emesary PropertyNotification) to transmit a well known (to the model) list of properties packed into the generic properties.

The maximum size of a string was also increased to 768 bytes (from 128).

Not sure who posted this one!

Using the Emesary Multiplayer Bridge[1][2] is an effective way to transmit more properties using less space[3]; however if the transmit filter property base[4] is also set to say 10000 or higher then only the really important properties will be available to other craft.


Posted by Richard.



AAL4955


I think the most important takeaways here are these:

- Transmission of properties is not guaranteed, the MP can drop properties at either end; however, how many of them are dropped depends on the FG versions on both ends and is thus difficult, if not impossible, to predict
- Property assignment is specific to an aircraft model and may even change between versions (of that model)
- There is no way to unambiguously identify a model and version over MP, so we routinely get mismatches on the MP system

So we really get two problems interacting: dropped properties, and incompatible aircraft models. Proper solutions to both would, I believe, require breaking changes, and I don't think those will fly given the current state of the FG MP ecosystem.
tdammers
 
Posts: 391
Joined: Wed Dec 13, 2017 11:35 am
Callsign: NL256
IRC name: nl256

PreviousNext

Return to Multiplayer events

Who is online

Users browsing this forum: No registered users and 6 guests