• IPACS is aware of the issues with magnetos etc. in the Switch Panel. They confirmed they are working on it, let's hope the next update will resolve this. I am quite dependent on the Saitek stuff.

    The next logical question would concern the FIPs (Flight Instrument Panels) but I don't dare to ask before they got the Switch Panel issues resolved.

    Initially, this would have been a task for Saitek to provide proper drivers, but given they are half-dead this would never materialize.

    Kind regards, Michael

    Intel i7-6700K 4.0 GHz / Asus MAXIMUS VIII RANGER / Kingston 32 GB DDR4 / Samsung SSD M.2 500 GB + Samsung SSD 1 TB + Intel SSD 500 GB (AeroflyFS2) + WD HD 6 TB / EVGA GTX 1080Ti 11 GB / LG 34UM95 3440 x 1440 / HP Reverb / Win 10/64

    • Official Post

    [...] Do you think there is a chance in a next release to have these switches linked to those defined in Aerofly FS Saitek panels menu ?[...]

    Hello Antoine,

    well since there isn't a way for me to add new control assignment via the settings dialog and the gc-map.mcf file doesn't use the same strings for identifying a message, I'd probably have to create the FSUIPC for the Aerofly FS 2. Which I could because with the SDK (that IPACS already send out to couple of people for evaluation) also contains an interface dll project that makes it possible to send data to the simulator. This way it could very likely be possible to read in other joystick or panel inputs. But I am really no expert in that area, would probably be faster to let the developers from IPACS implement it in the Aerofly FS 2.


  • Thank you Michael and Jan for your replies. Roger that. IPACS is doing a great job and are apparently actively gathering feedback from third parties editors to develop their SDK. I'm looking forward to seeing what's coming out of this, especially how AFS2 will deal with high loads of 3D buildings and vegetation like current standards in FSX/P3D. Looking at San Francisco it seems very promising.
    Regarding aircraft, as long as the SDK allows it, I'm pretty confident that editors like A2A will be able to program sophisticated models of systems behavior as they currently do for FSX/P3D.

    I tried integrating AFS1's nice Robin in AFS2, but remained unsuccessful... Looking at the files and structures, it might be the case that only a few things are actually missing...

    Best regards

    Best regards

    Config : i7 6900K - 20MB currently set at 3.20GHz, Cooling Noctua NH-U14S, Motherboard ASUS Rampage V Extreme U3.1, RAM HyperX Savage Black Edition 16GB DDR4 3000 MHz, Graphic Card Gigabyte GeForce GTX 1080 8GB, Power supply Corsair RM Series 850W, Windows 10 64 bit.

    • Official Post

    The major issue is the ping between the two devices. Some people are pleased with 10ms ping but others have 300ms. So worst case: each person has a ping of 300ms, then you lose 0.6 seconds just by sending the data back and forth. If you have to calculate something based on that data, you easily lose 0.7 - 0.8 seconds total. In that time the dr400 moved almost 30 meters, so you really would have to extrapolate the position of the aircraft well in order to get a smoth experience. You can't just have the towing plane jump left and right, up and down for more than a few centimeters in an instant, otherwise you would get in trouble with the rope physics calculation. Or you leave away the physics, so that the towing plane is never affected by any rope forces...
    And for a formation flight of F18s near Mach 1.0 it could quickly be 250+ meters. So when one player sees the other player at its side the other player sees him 250meters behind, unless you find a way to synchronize the times, maybe give every player itself a delay of 100ms that isn't to noticeable but still send its actual position to over the network... receieved data could be interpolated and then both pilots would see each other at identical locations.

  • Wow, that might just get complicated. Hadn't thought of ping/netlag kind of things. A standard 100ms delay for all wouldn't be possible I think, that would mean you'd need massive servers if you want to calculate 10 or 15 aircraft and all their positions/physics etc.

    So, impossible? Might be. Sad :D But thanks for the good reply :)

  • Is it possible in any way to add to this mod that when you click on the sockets for the pilot's headphones that the engine sound could be reduced and muffled somewhat. This is a feature on an FSX aircraft I have and nicely simulates wearing headphones and cutting out aircraft engine noise.

  • Dear Jan (Cessna 172 mod dev),
    AeroFly FS 2 is by far the best eperience in VR according to flight sim products - it really is. I enjoy the immersion in my Oculus very much. It looks simply amazing. The crisp details of all panels (needles/gauges) coupled with detailed scenery and spot on situational awareness makes this sim a truly VR masterpiece. I enjoy the C-172 mod very much and would like to see more details in its physics and avionics interaction.

    The aplication of right rudder during takeoffs roll is a bit exaggerated on the C172, I highly recommend to reduce this effect at least by 40 percent. I also tried lower the sensitivity of my rudder pedals, even special setting of the the dead zone on the yaw axis, but it doesn´t help. I fly real C172 aircraft pretty often and I´m sure that this minor bug should be easily improved in next C172 mod build.

    Next to it: the next fault is the RPM setting during straight and level vs climb or descent. In real C-172, when you climb, you just set the maximum RPM which should be at least 2.300 (depends on the airplane model) and you just hold the AOA on 75-85 kts (airspeed indicator). You cannot control the RPM with AOA - that´s wrong RPM behaviour. The RPM setting should remain constant during all maneuvers, only the airspeed is changing according to AOA setting. I highly recommend to correct the RPM behaviour.

    Finally: I would like to see a better simulation regarding radio navigation equipment: Per example: When you set the FRQ on the DME, the distance in NM instantly appears on the display. In real world: there is a slight delay from the signal (0-3 seconds - depends on the station distance). You cannot read the distance from VOR/DME FRQ when the station is not in range. Also: It would be great to have fully functional audio panel (incl. morse code, markers beeps, etc...).

    Now the positive side of your C-172 mod: in VR it is by far the best most natural Cessna 172 physics tested on a desktop simulator, very good and smooth during straight and level, turns and stalls. Just lower the YAW sensitivity and take off roll RIGHT rudder aplication. :)

    Thank you very much for your hard work!

    BTW: I wrote the same on STEAM AFS2 discussion, but I feel, this is better place. Thanks again!

    • Official Post

    Thank you very much for your feedback, maybe I should look into the flight model alterations again, maybe tweek the engine to more realistic power values. How do you like the changes in flap behavior compared to the original version and do you think that the overall handling is more realistic?
    I'll look into the rudder effectiveness as well.


  • Hi Jan,
    thanks for fast comments :-). The flap behaviour is almost spot on, the Vso (stall speed, flaps full down, power off) should be 40 kts, the Vsi (stall speed, flaps full up, power off) should be 48 kts on the C-172SP. Almost perfect. :) According to Vso and Vsi intentional stalls and spins technique: the C-172 is still very kind and a bit dozy airplane. When ailerons and rudder are in neutral position and the aircraft is in trimmed config, the plane will almost always (no turbulence, no wind gusts) fall down over the canopy without the instant spin effect. It will not start to turn left or right (usually pretty often in low-wing aircrafts). In case, you will lower the yaw sensitivity and takeoff roll, you can also make this behaviour a bit more forgiving. I flew almost any Cessna aircraft, and C-172, C-182, C-206, C-210 are very stable indeed. Just the C-150/152 is a bit more sensitive.

    I also like the fact, that you can still fly straight and level on stall speed in the Vso config - which is correct!! As I said, you done amazing job here, and the super natural physics is maybe a bit better than the A2A C-172 - which is a benchmark in FS world. That´s because AFS2 is running so smooth even in VR (full details on my machine). I still cannot believe how the devs managed such a smooth flying in the VR :-). I also flew several ILS approaches and it´s simply brilliant, how good and stable (in trimmed config) the airplane is.

    Of course, I can understand that you cannot correct everything, because the AFS2 is still in development, so many many basic features are still missing. Per example: we cannot set non-linear profiles on ailerons and rudder - which should be a standard feature in AFS2 and this could help us to diminish the oversensitive controls. The CH Products Flight Sim Yoke USB is the best cheapest yoke on the market, because you can set the non-linear controls of all axis outside the simulator, which works very good indeed and that´s the difference between crappy Saitek and superb CH. Because non-linearity is still missing in AFS2, DX controllers are not very useful here and hence your work will be appreciated very much by many real pilots - I´m sure. :-). Summary as follows:

    - lower the application of right rudder during takeoff rolls
    - correct the RPM behaviour
    - weaken the intentional stalls - too sensitive turning tendency under the wing (Vso / Vsi config)
    - when brakes are on, the aircraft is still moving forward TOO MUCH, please a bit more effective brakes
    - maybe more checklist features (magnetos check, proper amp and vac function, etc...)

    In case you will update your mod, just let me know and I will be more than happy to support your work and provide next feedback.
    Again: Thank you very much!

  • Please remember that indicated airspeed stall speeds seriously under-indicate. I remember a PA-28 180 doing impressive 40 knot, 40 degrees of flap stalls but I eventually remembered the angle of attack vs asi error table or graph in the flight manual (haven't looked at it for almost 20 years). The position error with the nose way up is really big. If the zero wind, low altitude, std atmosphere ground speed indicated in the top display or GPS/DME is accurate, the ASI should reflect the somewhat fanciful low speeds we got used to.
    If the Cherokee is ever reintroduced could it have the lovely wings level soft mushy stall that is so reassuring? The Cessna stalls are great fun at altitude though.

  • Please remember that indicated airspeed stall speeds seriously under-indicate. I remember a PA-28 180 doing impressive 40 knot, 40 degrees of flap stalls but I eventually remembered the angle of attack vs asi error table or graph in the flight manual (haven't looked at it for almost 20 years). The position error with the nose way up is really big. If the zero wind, low altitude, std atmosphere ground speed indicated in the top display or GPS/DME is accurate, the ASI should reflect the somewhat fanciful low speeds we got used to.
    If the Cherokee is ever reintroduced could it have the lovely wings level soft mushy stall that is so reassuring? The Cessna stalls are great fun at altitude though.

    Understood, the problem with AeroFly FS 2.0 is that proper atmosphere is not modelled yet, even the weight and balance is not on the list now. There are many basic features they need to be done properly in order to talk about real aviation here, that´s the main problem in this amazing piece of software. It´s still early access stage, but AeroFly FS 2 devs achieved a very good basis for further development and the "real feel" and 1:1 VR perspective (FOV) is something that cannot be compared to FSX, X-Plane or P3D. For my part: It´s miles ahead (and yes, I´m talking about the VR technology). We really need to support this simulator and I´m sure that regular updates will attract more virtual flyers even real flyers. We need world wide Jeppesen Navaids, world wide terrain with airports (even with generic textures). I don´t need to fly by numbers in AeroFly FS 2.0 the same way I fly in real world, but.... :) :-).

    I really appreciate very much the natural feel of VFR flying experience, it´s just a good start and I hope that AeroFly FS 2.0 will attract also third party devs because of SDK (very smart idea). The core elements should be done properly and key areas for AeroFly FS 2.0 devs are following: world wide Jeppesen NavData / airports, atmosphere simulation / weather, weight and balance, fully functional radionavigation/radiocommunication equipment, the ATC may be a good addition in later stage. Cheers! :)

    • Official Post

    No, the ISA atmosphere is already properly modelled from what I can tell from my experience as an aircraft developer for the Aerofly FS 2. Only very high up I found that the density is a bit too high yet - so before we do reentry simulation of spacecraft we would maybe need to change few things there. For the normal airliner and fighter altitudes though, its pretty realistic. I have not written the atmosphere code myself and haven't seen it but I have written the code of the airdata computer that gets the total air temperature and total pressure from the atmosphere model and then uses mathematical formulas to calculate the mach number and true airspeed from it. And the result of these calculations are quite accurate and plausible because when comparing them to the ground speed values there is no observalble difference. And the ground speed comes from the actual rigidbody simulation (ground speed with zero wind)... So I highly doubt that there is anything wrong with the implementation of the atmosphere.

    Of course I am just talking theoretical standard atmosphere here, I am not talking about weather or climate which is high up on my personal wishlist and rather important to a great number of customers.

    For the radio navigaiton: What things would you like to see added there?

    Weight and balance is certainly an important part of aviation, how detailed would you like it to get. I am happy if I get a note that sais: your airplane is fully loaded today, recommended flaps are ..., trim should be set to.... . But I know that there are useres out there that would like to set the payload and fuel exactly by the pound/kg.

  • I can understand that ISA is there, but the ISA without the weather and climate doesn´t make much sense in simulated environment, but again. I can understand that we cannot have everything right now. Anyway: glad that it is on the devs list. With regards the radio navigation: basic features are there, but not properly modelled, the good source (how each part of avionics works in real life in basic aircraft such as the C-172) is available here . If there is a majority of users which prefer the payload model as a gal/kg/pound - it will be very useful - even in basic concept.