My opinion about AF 2

  • Hello,


    The answer to this post from IPACS will probably be, that the forum is not the place to discuss the topics brought up by me in the necessary depth and that we need to arrange a user meeting someday. I did not get an invitation so far, and since this is your support forum, get ready for some support.

    When I first started to play AF 2 after it was released, it got me into my rant against the flight info bar in another thread, but that was only a fraction of the post I wrote back then. I decided only to publish this part to give IPACS time to sort the things that were obviously bugs out, and because I had a slight hope that they would start addressing the major issues (see my post above) by themselves, before I proceed crashing the party around a freshly launched product.

    So I waited. And played a competing sim that may have working ILS, but can only be used with 500m visibility to avoid eye cancer.

    In the app store, costumers are used to a certain timeframe where the initial bugfixes after the release of a new version or major updates need to show up. As a developer, you can stretch this timeframe by acknowledging that the bugs found by users are being worked on. This would actually be no deviation from IPACS company policy of giving no dates and little info on what to expect next. A bug, unlike a new feature, is something which I as a costumer payed for not to happen. If it happens, it does not put me in a state of cheerful expectation when I do not know if someone is working on the problem. It makes me angry.

    You did not publish an update, nor did you give any info on when to expect it. And slowly but surely we are getting to the point were probably not only in my user experience the awesome (working) new stuff brought by AF 2 starts to be getting outweight by the stuff that is also present, but not working properly.

    So what needs to be fixed?


    Radio Navigation

    Albeit the great work that obviously went in making flight planning and radio navigation possible in AF 2, from a user perspective it is much closer to a bug then to a feature. There is not one ILS approach I tried in this sim, that worked properly: the automatic choice of frequencies is erratic to the point of dysfunctionality, the glideslopes and localizers are often inconsistent with the corresponding runways, which is often the result of VORs being displayed as ILSs with a stuck glideslope (at least thats my theory) and most often they show up around 3 miles from touchdown at best, on perfectly established visual approaches from 15 miles out. The only airport where low visibility approaches are a little fun is the eastbound approach to Monterey Regional.

    This is not something were I as a user can sit down and write down for you what works and what does not. It is to far from working for this. I know that there are not yet a whole lot of people out there who use this extensively in a mobile sim, but you came so far bringing a perfect representation of this functionality to iOS, given the platforms constraints, that it hurts me seeing it fail due to a few bugs.


    Flight planning

    In its current state, the flight planning mode gets something done, that the competition actually can not do: plan a complete route with climbout and turns to a straight landing in under 5 seconds. Which is alright for most users who have never played a sim before. For everything and everyone else, it sucks. The flight planning feature as it is would be awesome for aerofly FS (1): it gives me a line I can fly with my Cessna and finally arrive at my destination

    It does nothing for me when I want to know the course to my destination, the altitude profile I need to fly, which VOR or NDB I can use on the way, which runways have ILS etc. Of course I can look all this up, but someone talked about making an accessible sim on this forum, and I remember it was IPACS.

    So what we need is:

    - Map with all NAVAIDS
    - Planning mode that makes it possible to integrate them in a route
    - During the flight, only the preselected NAVAIDS from the flight plan are used (solves the problem with erratic selection for most users)
    - ...and shown, optionally, together with the route, on the MFDs of the EFIS equipped aircraft and not only in the (pointless) info bar. (optional: swipe across the MFD to zoom, would be a great start to try invisible touchzones)
    - an ETE (NOT in the info bar) or something else to make rudimental descent calculations (optional: let the flight planning calculate this automatically. If you do this, you also have everything you need to offer a basic LNAV and VNAV functionality for the autopilot)

    These things are not "nice to have". They are essential to bring purpose to the amount of work you put in radio navigation.


    Last point includes everything from the flight info bar to limited rudder authority:

    Stop making a sim for people who do not play sims


    Right now, it is a fact that the (overall inferior) competing product outsells aerofly by a huge margin. It does not only sell better, but after buying, people keep using it more often then aerofly. Look how many videos are posted on Youtube playing either sim. Look at the activity in the corresponding Facebook pages. It is the continuation of the old rule in the simulation market, that consistent functionality regarding the representation of real aviation on a given platform will always prevail over eye candy.

    That is why add ons to MSFS by PMDG sell for more then the original sim. That is why Falcon 4.0 is still played today. That is why nobody here has ever heard about Flight Unlimited.

    You can lament that with App Store economy everything has changed, that your game is only downloaded by 12 year olds who can't use an ILS and that you can not charge them more then 5 €. We had this discussion before. Meanwhile, your competition sells annual memberships for almost 50 €.

    Fifty.

    Euros.

    Annually.

    Which I payed. Happily. Together with thousands of others. Like in the old times, when Sims where 50€. In an App Store were the average revenue per customer is something about 0, x €. Because I get from them what I am asking for in a good sim, which has nothing to do with the central feature for this price being multiplayer.

    If you want a feature for your sim that is a massive selling point, like multiplayer for your competition, make it system depth. Your concept of reducing complexity, until the Extra does not do anymore what an Extra is build for, did not work. People are buying your sim, but you destroy long term motivation and lose in in-app sales. And neither you nor your product deserve this.

    I am still looking forward to a user meeting someday.


    Jonas

  • I would pay for system depth over multiplayer any day!

    Sorry to disagree, but I want both and will pay for both. I am no 12-yr old, and have flown DME arc approaches in the real thing. So having the system depth to do that sort of thing in AFS is important to me. But for purely recreational flying -- and that is a very big part of why we all fly, whether in a sim or the real thing -- multiplayer provides the opportunity to enjoy the experience with other pilots. For now, the only way to get this is to "simulate" multiplayer using an audio link:

    http://www.ipacs.de/forum/showthread...-in-Aerofly-FS

    And multiplayer mode does have training value as well. One of the skills a private pilot has to learn is how to enter the traffic pattern safely when other aircraft are present. Practicing this in multiplayer with "real" pilots flying other aircraft provides excellent practice. Also, group fly-ins, spot-landing competitions, aerobatic demonstrations, and so on could be conducted with fellow AFS pilots. This would create a level of community involvement that goes well beyond what is possible in this forum, great as it is.

    I am not asking for multiplayer as an arcade-game gimmick, but rather as a serious enhancement to the AFS flight experience. Multiplayer isn't "unrealistic" as some have said; rather it is the ultimate in realism. The only reason it gets the bad rep is because right now most multiplayer apps are arcade games appealing to 12-year olds who just want to blow things up. Accuracy in flight dynamics is not important to that demographic. But the unfortunate association between multiplayer and poor simulation fidelity is not inevitable. If anyone can prove that, it is IPACS.

    Regards,

    Adak47

  • Hi,

    I agree with you, Adak, for the most part. We need a combination of both, system depth and multiplayer. I know several people who don't want to buy the aerofly yet because its missing a multiplayer.

    My suggestion is splitting the multiplayer:
    Have a derp around multiplayer (easy mp), where you can join any time and just have fun
    and secondly a more professional multiplayer (pro mp) where you get banned for not following certain rules.
    You'd maybe have to do certain missions in the simulator before even be able to join to that pro mp. And redo some of them if you get banned? (just ideas)


    [OT] How cool would it be to challenge each other with missions? Say you just saw your friend do a horrible landing. If you then dare him to do the "spot landing challenge" 10 times he can get some credit points or extra archievents if he does it. [/OT]


    Do you know VATSIM or even Pilot Edge?

    Vatsim is an online network/multiplayer where you file a flightplan and then connect to it with FSX, P3D or X-Plane. You then get to enjoy real people as controllers, fly-ins with a lot of serious simulation pilots from those simulators and usually a more professional environment. Trolls are very rare and banned really quickly. -> Amateur to professionals

    Pilot Edge is basicly the same thing except that you have to pay for the service and get real life controllers! And you can do virtual exams that are as hard as real life ones -> professionals

    Vatsim would be an amazing addition to the aerofly FS when we have a decent navigation system with real navigation data (including world wide airways, SIDs and STARs).

    Cheers,
    Jan

  • My suggestion is splitting the multiplayer:
    Have a derp around multiplayer (easy mp), where you can join any time and just have fun
    and secondly a more professional multiplayer (pro mp) where you get banned for not following certain rules.
    You'd maybe have to do certain missions in the simulator before even be able to join to that pro mp. And redo some of them if you get banned? (just ideas)

    ...

    Vatsim is an online network/multiplayer where you file a flightplan and then connect to it with FSX, P3D or X-Plane. You then get to enjoy real people as controllers, fly-ins with a lot of serious simulation pilots from those simulators and usually a more professional environment. Trolls are very rare and banned really quickly. -> Amateur to professionals

    Jan,

    Yes, I like your idea for an easy and pro mp flying environment. And Vatsim sounds like it would really be interesting.

    A big issue in multiplayer is how many players per room, and how many rooms? This depends on connection speeds, device capability, and the complexity of the flying environment, among other things (such as server size, number, and speed). And who defines the characteristics of the room, and how is that implemented? Some flexibility here would be a good thing. For example, a room might allow only two pilots, but provide excellent response speed whereas a larger room size might be accompanied by lower response speed (producing some lag in the updates of aircraft position). There is also the issue of local vs global objects in the room. What, if any, objects and conditions (wind, clouds, etc.) in the room are provided by the room itself, and what are local to each user's AFS install? For example, if we fly in Switzerland and someone has installed wind turbines or huts as add-ons, they will presumably see those when they fly near them in the public room. But other pilots in the room won't, unless they have installed the same add-ons. Call these local objects, because they are specific to each user's AFS install. Now, suppose I set up a room for multiplayer AFS, and want it to have an object (say a bridge over a river, or a blimp moored over Bern) accessible to all who fly in the room. This would be a "global" object because it is available to everyone who enters the room regardless of their specific installation. Now it can't be simple to blend both the local and global objects into the flying environment, and I am not sure it is even possible or feasible. But if it is allowed, then the question is how does the room creator install them on the server so that all pilots visiting the room get to experience them? Of course, the easy way out is to not allow global objects at all, and what you see is whatever you have installed in your own scenery.

    The same issue applies to conditions, such as wind and clouds. Do room settings over-ride current user settings? Or does each pilot fly with their own settings in effect? The more I think about it, the more I realize why multiplayer is so difficult to implement! But if I think about all the things I can set in the AFS menus, and imagine those settings being stored in a configuration file, then in principle there is no reason such a file couldn't be used to temporarily set each pilot's settings while they are in the room to the room standard. Upon exiting multiplayer, the pilot's original settings are restored. Something like this might work with scenery objects, as I discussed in the previous paragraph. First, it is assumed the default scenery will be used in the room as the starting point (i.e., user-installed add-ons are not loaded in multiplayer), and variations from the default are provided by the room creator in a configuration or a scenery add-on file that "customizes" the room experience. This way everyone in this multiplayer room will have the exact same out-the-window experience. And likewise with wind, clouds, and other conditions (as discussed previously).

    Well, if IPACS can do something this ambitious, I will be amazed and even more impressed than I already am (if that is possible) with the devs programming skill!

    Adak47

  • Hello Adak47,

    I'd prefer to only have "local objects". If you want to have a group event where everyone is flying limbo under a bridge, well then everyone needs to install that add on first.

    As far as I understand the big issue is the latency of the connection and the graphics cards memory. The actual amount of data that is streamed is not really an issue I think.

    Ideally we would have a system that adopts to your graphics card and connection. If your PC can handle multiple highly detailed aircraft then you should be able to get those. But if your device's hardware is limited then you'd only get a rough graphical experience or less aircraft visible.
    How about a global session that can handle thousands of aircraft at once? You'd only see those aircraft on your navigation display, get a TCAS warning and maybe a rough 3d model of it when you get in an 8km range to it. (has to be good just enough to represent an aircraft from that distance, no detail needed at all, its going to be just some pixels anyway, maybe synchronized every .25 seconds with interpolated position? - does not matter if they are delayed by one or two seconds, really).

    Additionally you could have a short list, based on either your direct surrounding or a user defined list of friends. Each client on that list would send/receive more detailed information to/from you and is therefore rendered in more detail. The possible length of that list could vary for your systems performance. So on an iPad you might only get one aircraft rendered in detail while on a desktop PC you could have a small aerobatic team even.

    The weather should be real world weather downloaded from a different server or just a global weather preset as a user overwrite.
    It does not really matter where you see the clouds on your device unless you are flying a glider competition or share one cockpit.
    The time should also match the real world time or be a user overwrite.

    Cheers,
    Jan

  • Jan,

    These are all excellent points. Thanks for the detailed response. I like your suggested solution of variable numbers of other aircraft based on device capability. And yes, aircraft far away won't need to be displayed in great detail or updated as often as those flying alongside you. All of this helps to reduce load.

    If we only allow local objects, then it would be helpful if the room creator could post the necessary files for downloading by the room participants. Better yet, Aerofly FS could provide for installing these "global" objects in a special folder in the user's Aerofly directory, and then optionally deleting them after the participant leaves the room. I say optionally, because some users may want to delete the objects and some may want to keep them. But having AFS handle the overhead of installing and/or removing the necessary files will remove the need for each and every user having to do his or her own install/uninstall. Of course, the current members of this forum are probably willing to do their own installs and uninstalls -- but I am thinking ahead to the larger user-base which will contain many people who are not adept at doing that soft of thing, but who want to participate in multiplayer. (Also, my philosophy is that any manual process that can be handled by software, should be.)

    The weather and time need to be set by the room creator, or default to the room creator's local region (assuming it is located in an AFS scenery area). Within any given multiplayer room everybody needs to be flying in the same conditions and at the same time of day. But of course, clouds and the like need not be identical for everyone. In Storm Raiders, I believe clouds far away are generated independently for each player, but clouds close in are not so that pilots near each other will be flying in and out of the same clouds.

    Your idea about a user defined list of friends is outstanding because it allows the user to control whose aircraft are given priority in rendering, while at the same time allowing large numbers of participants in the room. (Storm Raiders has a "private" room option which accomplishes the same thing, but you are still limited to a maximum of 8 pilots.)

    This raises the issue of pilot identification. How can you know that a particular aircraft is one flown by a friend? In SGSR you have the option to display the pilot's handle over the aircraft. But in the interest of realism we might want to use aircraft type, audio communication, and/or a distinctive flight maneuver (wing waggling, barrel roll, etc.) for identification.

    Finally, there is the issue of the appearance of customized aircraft. Suppose someone alters a default aircraft such as the Corsair by adding a distinctive color scheme or insignia. That is a "local" property, so when that Corsair appears in multiplayer other pilots will presumably see only the default Corsair. And what about Krzysk's DHC-1, Stearman, or any of the other user-created aircraft or soaring plane? Hopefully they will show up in multiplayer. What I am unsure about is if the user makes mods, like Krzysk's adding a color scheme and tail-number to the DHC-1 for me, will it be preserved in multiplayer?

    These are just some ideas and questions to bounce around.

    Best,

    Adak47

  • The aerofly can only display what is installed in its folders. A repaint is only displayed if you have the repaint installed as well. Otherwise you will always see the default textures.

    Same counts for aircraft. The aerofly can only display the 3D models that are installed. A custom aircraft will not show up unless its installed on the client's device. The aerofly RC 7 and aerofly 5 show a warning in the chat window "unknown aircraft" and won't show anything in replacement. FSX does replace unknown aircraft by a similar default aircraft. That causes graphical issues because the aircraft may then float above ground etc. Other RC simulators send the whole aircraft through the internet just so everyone gets to see everything. I don't like either of those tactics. I rather see nothing instead of a very bad replacement or risk my work in progress aircraft causing any problems on other people's devices because I am loading it in the multiplayer. Due to the high quality of the models such a synchronization of 3D models and textures would need a lot of time, the file size is quite big (up to 200mb!)... Not everyone has a connection that is fast enough to get the download in reasonable time.

    How about this:
    If you see that someone is loading an aircraft that you don't have installed you can look up some public information about it inside the simulator ("aircraft profile"). These information could include a download link for example that you could copy to your clipboard or add to some sort of "which list". Or open a private chat with the person who loaded that particular aircraft and ask them directly for a link...

    Regards,
    Jan

  • The aerofly can only display what is installed in its folders. A repaint is only displayed if you have the repaint installed as well. Otherwise you will always see the default textures.

    If you see that someone is loading an aircraft that you don't have installed you can look up some public information about it inside the simulator ("aircraft profile"). These information could include a download link for example that you could copy to your clipboard or add to some sort of "which list". Or open a private chat with the person who loaded that particular aircraft and ask them directly for a link...

    Thanks Jan for the very clear explanation. Your suggested solution is acceptable, but less than ideal. Someday bandwidth will be adequate for the transfer of 200 MB (or more) customization files. The way VR programs are headed, we will need that. But progress is coming. I remember when uploading a 1 MB file was a big deal (before your time, I suspect). Nowadays I think nothing of sending a 10 MB high res photo of a Mars Rover Lander pic to my daughter! Soon 1 GB will be no big deal either.

    Actually, for most purposes of multiplayer it is not terribly important to display customized mods of aircraft or even scenery. The defaults are good enough, and the main fun with multiplayer is the interaction with other enthusiastic pilots.

    So, yeah, that would work. If it is important to me to see my friends' custom mods, I will install them on my system. Otherwise I will just fly as is. It will still be good!

    Thanks,

    Adak47