Posts by perestain

    I'll add it in the next update, no eta though since I'm still pretty busy. I think I have to rework the way planes are organized and chosen by rng. Right now it's not really expansion-friendly. I didn't really think a lot about upcoming aircraft back then.


    As a quick and dirty workaround until then you could just allow yourself to fly a R22 whenever you rent a pitts (same amount of seats) and then edit the journal entry manually. Fuel cost and range will be a little off, but they're not that accurate to begin with.

    Yes, that's exactly what was happening back then, a slow rudder increase to the left for one second and then a sudden return to the middle. It hasn't occured since to me but it's nice to hear you found and resolved it. It made some approaches a little more exciting than necessary, hehe :)

    Hey perestain I have another idea for an improvement. I've used Pilot Journey for every flight i've made since you launched it. I love the variety of time and weather it generates so i've been thinking of other aspects of my flights which i don't tend to vary but i could.


    How about if PJ did some ATC style things like specify what runway you should land on and whether you had to join the traffic pattern or should fly straight in?

    That sounds like a cool idea.


    I have tried pilot2ATC for Xplane once and all this flightplanning and ATC stuff is very fascinating, but I'm afraid it's also well beyond the scope of my aviation knowledge or programming abilities. For example it would probably be wrong to set the active runway randomly, since it is has to depend on the conditions and whether it has a published IFR approach etc etc. Also I would probably have to manually research all that info which I barely understand for all the runways on all airports which would take a lot of time.


    To add more realism, right now its probably best to do the flight planning manually, i.e. use the navigation link to http://www.airportnavfinder.com to get published depature and approach procedures, decide the runway based on wind and conditions yourself and then head to https://skyvector.com/ and create an actual flightplan along the published VFR or IFR routes (whatever applies) and edit that flightplan into FS2 as closely as possible. http://www.navfltsm.addr.com/ is an excellent ressource for learning how the navigation works. But I am really just a beginner at following all that stuff myself and far from being able to create automated results that are convincing here.


    I could of course add some random occurences, things happening during flight, like a sudden change of destination. flying an extra round around the airport, some dramatic stuff, engine failure which you had to manually simulate or things like that. Its on the idea list, but I can't promise very much right now, I'm currently too busy with my actual work to heavily delve into a big PJ update.

    x, y, z is the karthesian position in space relative to earth's center, rotated with earth reference frame. X is towards 90° East meridian, Y towards 0° and Z up I think. And it's an ellipsoid model of the earth...

    Wow, thats that's pretty cool. Thanks a lot for the insight. For some reason I expected spherical coordinates for that shape but the values just didn't make sense for that.

    PJ already switches planes.

    The problem with positioning is that main.mcf does not use lat/long values but 3 parameters which are presumably x/y/z coordinates.

    This would be a hint towards the earths surface possibly being modelled as a flat projection and not as an actual sphere or ellipsoid.

    i'll make a note of where i see them too. We can take a stab at it empirically.


    As for fuel burn i see that vary according to throttle/mixture but suspect the total doesn't reduce.. i don't see fuel mentioned in main.mcf.

    I would expect height in ft. to be just a linear function of the value. I'd suggest just checking min and max height, and then maybe one or two to other values in between just to confirm. Should only take a few minutes, but I cannot do it atm because I have other work to do.


    Edit: And yea theres no fuelburn or weight. Would be cool to have it as it would make it more of a gamble to fly a mission with very little fuel and connect PJ even more to the sim. I'd have to add a "I have run out of fuel and performed an emergency landing"-button haha. The problem though is that the fuelburn calculation in PJ is dead simple and doesn't even account for weight. It probably would need an overhaul to be adjusted to a fuelburn simulation in FS2. And for that we'd need to first figure out how exactly the fuelburn in FS2 works. Otherwise the PJ calculation might drastically differ, so that in theory it could happen that you could never reach your destination in FS even when PJ estimated that it is within range for example. Unless someone from the developers would share some information this would include some heavy testing to get it right for all the planes.

    The cloud height and density slider im the main.mcf have decimal values between 0 and 1. No idea what the numbers translate to. The height could simply be checked by trying it out in FS2 though.

    Thanks for all the detail. Maybe I will be looking for something in the middle. I have zero interest in competition or high score but, I do like a little spice added to my flying. I will load it up and see how it goes. Thanks.


    Regards,

    Ray

    Yea to be honest the highscore is more a playful gimmick and a hommage to oldschool games from the past. It's not like you could actually brag with a score that is generated and stored in plain text xml fomat.

    Sorry if I was being naive here. If it really is that much of a legal problem then of course its best to completely stay away from sharing over the internet, even if all the images themselves are available to the public. I guess there's a reason after all why people need to study for years to understand law :)

    I have always found it questionable that it is considered perfectly legal to download and modify bing map data, while it suddenly becomes illegal to pass those modifications on to a friend. I'm no legal expert, but if you ask me, this sounds like a rule that people who make a business out of creating scenery must have made up.


    I mean, the positive side is that it turns flightsimming into this slightly anachronistic elderly man's hobby from the 80ies, where everyone spends hours sitting at home preparing exactly the same stuff for his hobby and it becomes the most time consuming activity of the whole thing. Just like building railroad models tracks, or painting tabletop strategy game miniatures, but in the digital domain. I can sympathise with that, but I believe this is a choice and not a necessity. The technology for sharing is definitely there.


    The bandwidth argument is even less convincing. People speak of a few giga or terabytes as if it was this huge amount of data. It's 2018, a a time where millions of people stream hd video over the internet all day long and there are filesharing possibilities where you can upload and download tons and tons of terabytes of data almost for free.


    My hypothesis is that people in the flightsim scene actually do not want to just download scenery from somewhere, but rather enjoy spending a lot of time creating it for themselves because it is an enjoyable part of the whole hobby experience. The copyright and bandwith arguments seem more like quick little excuses t justify that. But what do I know...

    Can I specify that I only wish to fly the Learjet 45?


    Regards,

    Ray

    Lets see...

    if you play a regular game, you would need to improve your pilot license to level 3 (jets). To get there it takes $61,000 for the licenses and at least 30 missions after which you could afford the licenses comfortably. Unless maybe you wanted to make more cash and got caught smuggling drugs or something. But even after all that you could only rent a learjet if one is available. To fly one everytime you would have to buy one, the cheaper used ones are around $800,000 which takes a significant time to save up to. And while you could probably fly it almost every time, it won't necessarily be the most profitable choice to do so. So flying learjet all the time is more a scenario for when money is not an issue anymore (just like in real life I guess lol). That said, I had one testing game, where there was a phase during which I made a good profit flying several missions with the learjet. But usually once you are allowed to fly jets, heavy charter missions over the pond net the best profits. During my tests I wasn't simming though so I didn't have to care too much about how long the flights would take.


    But if you really just want to fly learjet missions and don't care too much about the game progression you could just enable cheats, switch to outlaw mode, (which would allow you to fly anything without license), give yourself a million dollar and buy a learjet on day one.

    And then just play as if it was a normal game. The only difference would be that your scores will not officially count towards the highscore. And your stats panel will display that you cheated. Thats it. And if that's actually bothering you, you could modify your savegame with an xml editor to turn it into a normal game again.


    Spit40

    Wow, that sounds like a really cool experience!


    I agree that some randomness goes a long way to making things feel realistic, interesting and even meaningful with a little imagination. The computer is a clearly defined system where everything that happens follows pre-designed algorithms and can be perfectly controlled. (Perceived) reality absolutely doesn't work that way, no matter what you do there might always be something catching you off-guard. If you want a computer simulation to feel realistic, you need to trick yourself and leave room for unexpected situations to happen to you, which is just more likely when you randomize a lot of variables.


    The chances for bad conditions in PJ are not equal btw. There is a bias for higher visibility and lower wind and turbulence, to make really bad conditions more rare. Visibility is internally a value between 0 and 100 and it is calculated by 20*sqrt(x), where x is a random decimal number between 0 and 25.


    So for example the chance to have a visibility lower than 20 (out of 100) is less than 4% in PJ.

    I do have another suggestion too. I don't like flights of more than around 30 minutes. Perhaps you could have a setting option of maximum flight time/distance to feed into your random mission generator. Sometimes i load it up and there's no flight i'm interested in - just a couple the right duration and they're in a direction I've just come from or a quick joyride that makes no money. So i close the app and reopen. Recently i did that over 10 times to get something suitable.


    Closing and reopening to refresh the mission board is basically the same thing as waiting, except you circumvent the 1 week game progress atm. I think I might find a middleground here and make the wait button only progress one day, but make loading a save progress a day too. So you don't get an advantage from closing and reopening.


    The mission generation is highly dependant on the size of the airport. Smaller airports spawn more short missions.


    When a mission is to be generated, the generator looks at all destinations in the game and puts them into several brackets, depending on the distance.

    He then decides by chance from what distance bracket a destination is chosen. Here are the chances, in relation to the current airport size:

    // Airport size 0: 80% local, 20% short

    // Airport size 1: 60% local, 36% short, 4% mid

    // Airport size 2: 35% local, 55% short, 10% mid

    // Airport size 3: 10% local, 30% short, 40% mid, 20% long

    // Airport size 4: 20% short, 20% mid, 25% long, 30% overseas, 5% far overseas


    local: 0-50 nm

    short: 50-275 nm

    mid: 275-1000 nm

    long: 1000-2500 nm

    overseas: 2500-3500 nm

    far overseas: 3500-7800 nm

    out of bounds: >8000 nm


    If you look for flights that are half an hour long, your best bet is probably hanging around on airports size 1 and 2.


    Another option to save time is to skip some of the autopilot cruise, and just fly the take-off and the landing. The mission timer is rather forgiving and not proportional, i.e. for a longer mission it will not make you wait as long before you can confirm the landing.


    the formula is atm: missiontimer = 4.3 * (expected time to fly the mission) ^ 0.75 (in seconds)


    For example,

    the missiontimer lets you wait for 30 minutes if it expects the flight to actually take around 52 minutes.

    and it lets you wait a 60 minutes if it expects the flight to take 2 hours and 12 minutes.


    The expected time calculation is not very precise though, it just uses the distance and divides it by a fixed cruise speed for that plane, which may not even be 100% correct and does not account for any circumstances whatsoever.



    Anyways, rather than forcing the mission generator to produce flights which take a certain amount of time, I'd like to adjust the above formulas and chances, so that it's more playable. That way the mission generation is more robust when more airports are added later.


    Maybe I'll just add an option to cap the missiontimer at 45 minutes or an hour or something. That way you could fly all the missions and skip a part of the cruise which is not that exciting sometimes. I simulated a full 900nm flight from Miami to NY in a baron once and I admit it got pretty boring during cruise :)

    Video with Pilot Journey, hope you don't mind perestain.

    Not at all, it's cool to see people having fun with the game. Nice video, I especially sympathised with the landing, because thats how I tend to land oftentimes, even with usb peripherals :)


    I've just finished a few updates, the main improvement is that you don't need to tweak all the sliders for the weather anymore. If you launch FS2 after starting the mission, the plane, time and weather should be set up alright (thanks to a tip from Spit40). You only have to check your starting location and navigation. The newest version also has buttons that link to the LiveATC website (only works for larger airports, but I love it) and to navigational help websites.


    Updates are on the first page.

    Updated to v1.01, downloadlink on first page.


    ------------------------------

    Regarding autosave, how about you treat auto save like the other slots 1,2,3 ? Call it slot 0 allow it to be loaded and saved manually but also it autosaves everything on closedown. Another thought is to put all those autosave files in a subfolder. My desktop is getting busy.

    I think I like having more and potentially older autosaves just in case. The autosaves have a number that corresponds to the ingame-day, so it shouldn't be more than 7 at any given time. It's probably best to keep everything in a folder and make a shortcut for the desktop.



    Your flight plan solution sounds really cool, I'll try that out myself. I didn't even know oculus home had such a feature. I was using a small notebook and a pen so far...


    Which pdf viewer do you use?

    One small addition would be to pick a random time along with weather conditions as opposed to UTC all the time.


    Of course I could just put in any time I like, but the whole point is the choice being made for me lol.

    Yea, I agree, I felt the same. I put it on the list.

    This looks pretty interesting. TBH I have no idea how hard it is to do it in java. But I'll give it a try for sure when I make a new upgrade, this would be a cool QoL addition. I hope there is no coypright or other problem to alter original FS2 files. So far I found it nice to not interfere at all. But preparing the weather manually gets boring rather quick.


    Regarding the autosave, I made it only after flights and not all the time to have a better safety net of old files incase the current game gets somehow corrupted. But autosaving after writing in the journal is probably not a problem, because it doesn't happen automatically all the time. I added it to my list.


    You're right, Wangen-Lachen airport is not in FS2. Ooops, my bad.

    I initially took all the Switzerland airports from the list in this thread, thats also when the whole project started:

    Switzerland DLC - Airport ID - Sorted by City Name


    I found out a lot later that it isn't 100% the same airports as the Switzerland DLC and corrected a few, but I guess I have overlooked Wangen-Lachen, probably because by then it already sounded really familiar after all the testing. I'll take it out of the game in the next update, so this doesn't happen to other people. An easy workaround would be to just take a bus to another airport until then.

    I wasn't really aware about the existence of fscloudport when I started. Maybe there should be a way for people to freely add their own airports at some point. The mission generator should be robust enough to handle it. The scorekeeping would need some overhaul though.


    I can't explain why your location suddenly changed to KSEZ though. It must be a coincidence that of all airports it happened with the one that is not in FS2. Nevertheless a location change like this shouldn't happen at all, except when the airport database changes. It might be a bug though. Are you sure you have loaded the right savegame?


    BTW the online web or multiplayer port is not gonna happen, I'm afraid. I have no clue about that stuff and rather little motivation to learn it from scratch. I have had enough with java for now, I guess :). The other problem is that it is extremely easy to cheat scores. In terms of validity, a forum thread where people just post their score would probably archieve the same thing. To make a actual meaningful ranking of it it would require lots of work and probably a completely different approach. And i must admit I even like it a little bit that it is offline and single player. It somewhat feels like playing in the 80s. When computer games were for lonely nerds and maybe their friends who came over, and nobody else knew about it. :)

    No worries, thanks a lot for taking your time to openly comment on this.

    It's completely understandable that this is a very niche problem and that there are other things to add that will have a much bigger impact.


    I'm not up to date on all the development rumours but if something like ATC really arrives eventually I would happily deactivate my virtual yoke for that :)

    Yeah I did think about that. Disabling the virtual yoke seems like an option, but not ideal

    I'm also wondering whether it would be possible to edit the aircraft config so that the yoke responds to aircraft movement (via yoke hardware), but doesn't respond to "vr hands" input.

    Couldn't we just ask for a toggle added in the FS2 options to disable touch input to the yoke? It would seem more like the proper way to do it and at least in my imagination quite an easy fix.

    This looks interesting for sure.


    The problem I see is that using the gloves with an USB yoke you'll likely give unwanted commands to the virtual yoke as well. Maybe just removing the virtual yoke will do, although I like having it so far. Not having to pick up and lay down the touch controller after every start and landing is probably well worth disabling it though.