Posts by Jet-Pack (IPACS)

    In regard to the "addons add complexity and that slows down the sim":


    Hmm, not that much of an issue from my point of view. I think the add ons aren't that optimized maybe or the actual connection to the sim slows everything down. As far as I know a lot of aircraft add ons do a lot of computation outside of FSX in their dlls and in X-Plane you have lua scripts I think. That could be the reason why the performance takes a hit but I'm not too familiar with their architectures. Not my job to build aircraft for the other sims :)

    All I can speak of is my years of experience with Aerofly aircraft development. At the moment the engines swallows everything I throw at it. There isn't actually that much computation needed in our architecture, we have our simple building blocks that run relatively short code, only through the connection of them we build up the complexity. So the code is still simple and very optimized because it doesn't have to check a lot of things, each thing does their intended purpose and we're good. Even talking about other complex systems like fuel, electrics, etc, we have already added that in some aircraft and it's still running just fine. I think the engine will be able to calculate even double or three times the dimension of freedoms that we have right now, no problem. As long as it stays in the engine and we don't do crazy stuff to waste computing power, e.g. having a weather or terrain radar scan every cubic centimeter, there shouldn't be an issue. We can just implement things like weather and terrain radars the way they work in the real world: scan from left to right and back and then there won't actually be a performance hit. I think it's stuff like that where the add-ons for other sims may not have enough access to the core and thus need more computing power, e.g. to run a weather radar.

    For the future we will probably just keep adding building blocks and with that you can create your aircraft in the tmd files. We/You can create insanely complex systems with failures and what not and you don't have to code anything, just connect the objects that exist. This way performance stays very good and is not an issue. We may have to add dll support in the future, but then performance could decrease if the other end doesn't code it properly. We can't control what code they run outside of the Aerofly and if they don't optimize it and do crazy things like wait for a download or a file to read or something, sure that could slow things down.

    But just us adding complexity the way we do it now, will that slow us down? Na, we're good!

    For now:


    The auto throttle in the A320 only works correctly if you set your thrust levers to the climb detent. Look at the thrust levers in the cockpit and let them snap to the first setting from idle to full. You can turn the auto thrust on with your levers at idle, then you have to move your levers up to CLB so that the auto thrust system has full authority.

    If you advance your thrust levers above the climb detent the auto thrust will disengage and you command manual thrust. They activate again when you pull the lever back to CLB.

    So key is: Move your thrust levers to climb (somewhere at around 60% ish on your axis), then the auto thrust system should work just fine.

    Hey guys,

    Yes, there will be increased system depth and we will improve the other aspects step by step in a natural way. We aim to create a simulation that provides the easiest way first and then provides options for those users that want to go deeper and care about all the complexity, I think automated engine start will be a thing of the future but you will also be able to start engines manually, we are working on an improved version of the copilot overlay that will make it easy to fly on some sort of simple autopilot with the press of a few buttons and then provide easy access for more advanced things like LNAV/VNAV. We aim for a simulator that is both accessible for beginners and provides enough customization for experienced flyers.

    Another example: You will always be able to create a new flight plan with very few clicks. Select the aircraft, click the start and destination airports, than magic happens and you get a full route with everything planned for you. We will probably select SIDs and STARs automatically for you in the future and calculate the fuel and possible payload etc. so that you can definitely reach the destination. You then will also see options to change the flight plan around, change the runways, SIDs, STARs, add waypoints in between, add more fuel or take some away, make the plane empty, fly as heavy as possible. We also plan to enable full flight plan programming via CDUs for those that want to do this sort of stuff. And when you then click on the runway you will be ready for takeoff, as always. If you don't want that, you can click on the parked aircraft on the map and specify if you want the full "cold and dark" or just the quick turn around so that you don't have to click that much but can still do push back and engine start.

    I'm not sure when we will get there but we're certainly moving towards a user friendly, yet hopefully very detailed and very realistic simulation. So there is nothing to be afraid of, the future of the Aerofly probably has both to offer: deep systems and easy of use for those that don't yet know how to handle all the system depth. Maybe an intelligent copilot that deals with all the stuff you don't want to do, automatic engine start is just one of the things I'm thinking of right now.

    In the mean time the scenery guys will probably add more airports and expand the regions and there may be auto gen to fill the empty gaps, this may also include improved night lighting, street lights along the roads, etc. we'll see. We're not sleeping, all of us are working as hard as they can to improve this sim.

    You're right, we had this discussion before. So you want stormy winds at the mountain tops when you set the wind speed to roughly a light breeze, is that what you mean? I've been in the mountains IRL and was never pushed across the ledge like in the old wind system, so I actually find the current wind field more realistic in Aerofly FS 2 and don't want the old one back. What do you prefer in the old system that the new one doesn't offer? From my experience slope soaring still works in Aerofly FS 2 so what is it exactly that you want back? Turbulence simulation? I'll ask about that, yes.

    General overhaul of weather system is on the wish list, that will probably include changes to the wind system towards more detailed wind systems, I hope. But right now we are busy with other things, like ATC for example.

    I'm also a glider pilot and would love to see the winch back in the game and also a towing plane. I'm sort of waiting for vehicles that can interact with the aircraft and those are very likely to come since so many people want push back trucks. When that happens I'm also going to urge adding a towing plane and which back in or even program them myself. I do have a real world license to be towed by an aircraft and a winch so I hope I can get everything right.

    Do you guys tow in the "normal" position behind the aircraft, above the propeller wash or do you tow in the low position underneath the propeller wash? What aircraft types do you have for towing in the real world?

    Also for planning purposes: how long are your winch cables and how much horse power do you got? Steal or fiber?

    And Luis, can you please specify what parameters of the thermals would like to see fixed?
    We had a similar thread in the past where I said we could maybe tweek the values for the diameter, spacing and strength.

    The lines aren't that many polygons, that should be fine. I mean look at JFK, that airport has all the lines as well. And buildings are typically square-ish, again not that many polys. I don't know if there is an absolute limit, I doubt it. More a recommended limit - I guess you could easily go up to the 200,000 mark and beyond without running into performance issues. I'd reconsider at half a million or more if it's just the airport, depends if there is going to be a detailed city nearby I guess.

    Do the scenery builders have any other info to add here?

    Jeff has probably more info to add.

    Hi guys, we sort of had the discussion about V speed call outs in the Learjet in the past. To my knowledge the Learjet does not have automatic call outs, so that is that... We may add call outs but I think automatic ones have priority in development. The "crew immersion package" with manual callouts of v1 or "v go" and "gear down" etc would be an entirely different topic and are a lot further down the line I think. Don't know if that is a highly demanded thing, let me know, if there is public demand for this the priority is higher.

    For a combined V1, VR callout we'd have to write a bit of code. Right now we can't just que the sound but that may come in the future, especially with ATC. We'll see, right now it's not super easy but possible to do a combined V1, VR or separate V1, VR, depending what you selected prior to takeoff.

    I agree, actual gear extension or retraction sounds and sound for the extended gear would be awesome! The issue is: how do you get just that exact sound file without anyone talking in the background or any airspeed sounds additionally?? A320 has quite the distinctive nose gear extension sound actually, if you have any sound sources for this that we can use, please contact us :) Also, anyone out there who has lots of experience with sounds, e.g. for other simulators and wants to contribute to the Aerofly? Also contact us :)

    Confirmed. That's why I mentioned that a lightweight 747 performs/accelerates much better than a heavy one.

    I thought that FS2 already does these calculations automatically and simply putting the fuel (weight) into the correct locations is the only thing needed.

    It works that easy in FSX, x-plane etc.

    Well the user interface will allow you to do that but the fuel tanks in the aircraft will be defined by their positions and dimensions in the tmd file. We're not just arbitrary going to add "some generic value" to the aircraft's mass and some inertia, for us it is actually easier to just specify the tank positions and then work from there.

    The mass and therefore inertia change are very important, roll inertia is just one example, there is might not be as noticeable. But you can't tell me an empty 747 and a fueled up 747 accelerate in the same time on the runway. Weight is carried by the wheels then and the friction isn't that high to play that much of a difference. The difference in acceleration comes from the mass not just weight. Weight is so boring it just points towards earth and doesn't actually do much only in terms of lift and downhill force component, etc. That is just the gravity force of the aircraft, doesn't say anything in terms of physics. Mass is what comes up in all the equations, centripetal force, acceleration (newtons laws), even the gravity forces are described in terms of masses.

    I'm not sure if you quite understand where I'm going with this, weight is a gravitational force, a by product of mass. Which is why if we simulate mass, its gravity and weight will be a by-product but only tell a small portion of the story. So no just adding weight to the aircraft is not going to cut it, that would just break the laws of physics we programmed :)

    Fake fuel flow or real fuel flow with the associated weight change?

    Real fuel flow without mass changes. I'm saying mass changes cause we won't just change the gravity force "a.k.a. weight" but add inertia due to the position of the mass and the dimensions of the fuel tank. So with full fuel tanks the roll inertia will be higher for example. Not just the aircraft needs more lift for level flight...

    I can implement the entire fuel system, with pumps, cross feed and what not. It's just that the tanks don't yet add any inertia or mass to the wings. The rest of the system is working, I can create a full fuel line networks with multiple fuel tanks and connect them together to my liking.

    In the C172 this is already implemented fully, and with my cold and dark mod of this aircraft the engine is also shutting down if you fly one tank empty or select the cut off. Now, I can't yet kill the LJ45 engines but I can make it so that you can draw more fuel from one tank if you idle one side and add full thrust to the other... stuff like that is doable.

    But I do not understand one thing: why in Cessna you can refuel with fuel at the required distance of flight, and the fuel is properly consumed, but there is no fuel in Learjet at all? Fantastic engines - perpetuum mobile!

    Because the fuel system is not complete yet and we have other topics to work on.

    It's not difficult to just make the fuel level go down on the display, I could add that to the Learjet in a few hours. The tricky part is reducing the mass of the aircraft dynamically thus changing it's inertia dynamically. So at the moment it is not yet implemented because it doesn't have any effect yet when the level decreases or hits zero. The engines are not shutting down and the aircraft is not yet getting any lighter. So there is no reason to implement it other than: the fuel number went down :O

    But it is nice to hear that you guys want something like this, has been a while since anyone talked about fuel. And just for that I'm actually going to look into the fuel system of the Learjet and may be able to quickly add this in.

    That is not possible. "2" is the key to step through all cockpit cameras, all cameras with the flag "InCockpit = true".

    You could delete all other cameras but then you have only your radio stack left.

    If you do only the pilot camera and your radio stack you could switch back and forth with the "next camera" command, for that delete all cameras with "InCockpit" true from the tmd and only leave the pilot cam and your radio stack.

    R0 is the camera position, x = forward, y = left, z = up, all in m, relative to the center of the model, which is roughly the center of gravity.

    For a radio stack camera copy an existing camera like the "CameraPilot" object and past it underneath, then rename your new cam, adjust the x,y,z and just reload, try again, reload, until you got it.

    I'd guess something like this would be the result:

    Code
            <[string8][object][camera_head]
                <[string8][Name][CameraRadioStack]>
                <[string8][Body][Fuselage]>
                <[tmvector3d][R0][ 0.55 -0.05 0.35 ]>
                <[tmvector3d][Direction][ 1.0 0.0 0.0 ]>
                <[bool][InCockpit][true]>
                <[string8][Tags][cockpit]>
                <[string8][Description][Radio stack]>
            >

    I don't have a great mental model of how FS2 interacts with the Oculus, but I used to use it just fine, starting from Steam and telling it on the 3-part pop up menu to use the Oculus (other) VR, but now it just starts up (steam says it's running) and then silently crashes (steam says it's not running). I've tried this with and w/o starting the Oculus software first. No diff, and starting it from the Oculus library appears to do exactly the same thing (although from inside the oculus) -- starts, then just immediately crashes. All my updates are up to date, I think. Is there a crash log someplace I can read. (I'm on W10, all latest updates to everything, as far as I know.) Thanks!

    Aerofly log file is located at "C:\Users\ ... \Documents\Aerofly FS 2\tm.log" check the last couple of lines for any hints.

    My best guess is: outdated graphics card drivers. Those are usually not updated automatically with windows 10 updates.