Open letter to developers who wish to or currently simulate GA type aircraft for X-Plane

[pryoski 84]

If we are to believe the stats from X-Plane, most people fly with the mouse anyway.

 

That is amazing! 

By the way, in Carenado's early X-Plane releases, I think it was the Cessna C185F, the circuit breakers worked. Okay, it is not SO difficult in X-Plane to have basic breakers functionality, but still -- Carenado with circuit breakers! Back then I asked Dan Klaue about that, and why he did not include that feature in later releases. Well, users simply did not care about that.

 

Yes .. and even un REP-ed the Carenado C185 is one of my favourite airplanes which I regularly fly

 

 

The main reason I prefer GA over tubeliners is the simplicity and pure "flying" as opposed to operating complex systems, so this may have something to do with it. I get enough of managing complex systems at work. At home, I just want to relax

 

I understand Jimmi! That is why I love bush flying GA.

 

I also like to drive my car. When I drive my car, I know all the switches work and yet I can still jump in and take it for a spin. I do the same for the AirFoilLabs C172 because in reality it's not a complicated plane; but I choose the AFL 172 over the Carenado 172 (which I own) and the X-Plane default 172 because it *feels* better and it's more complete ... maybe just in my mind lol!

 

BTW, I can live without circuit breakers being modeled either especially in larger planes ... but nice to have in the smaller airplanes I think.

[uwespeed 497]

One more aspect - or example respectively - one should not forget: the 80/20 principle (or be it 70/30, …). It’s not that hard to achieve 80% of a goal, namely by investing 20% of effort. For the remaining 20% of the goal, you need 80% effort.

 

Even more extreme in this regard is e.g the Formula 1. The car of a midfield team needs, say, 90 seconds per lap. The team invests e.g. 100 million $ per season. The top teams are driving 2 seconds (=2.2%) faster. But they have to invest 500 million $ (=400% more) per season for that. Actually totally mad and unreasonable. Of course, F1 is a special case, but you see the point.

[LAdamson 240]

I don't see a point in replicating all circuit breakers, with full functioning electrical systems.  More than not, a lot of breakers, as well as fuses, won't be reset in flight. I think that too many simmers get into the "what if", and want to play with items to the point of expecting "break downs" within the systems, and are disappointed  when simulation doesn't portray it accurately. Kind of like the "what if", if I intentionally  crash, or make a very hard landing.  It a developer has the capacity to model all electrical systems, then fine. I just don't expect it to be a requirement, or a "bar" that must be met, since single pilot operations,  won't be chasing down most electrical failures in flight, and flight engineers took care of most it, on those larger vintage machines.

 

P.S.  --- even these single engine planes take an incredible amount of wiring.  Unless you're just part of a group, who builds wiring bundles on a board, for commercially built airplanes, it can take weeks or months, running it all ( as in more sophisticated experimentals).  I don't think a GA developer needs to learn the whole process, just to replicate it for the flight simmer. If a simmer wants to learn the whole process, then by all means, start studying. 

[pryoski 84]

I don't see a point in replicating all circuit breakers, with full functioning electrical systems.  More than not, a lot of breakers, as well as fuses, won't be reset in flight. I think that too many simmers get into the "what if", and want to play with items to the point of expecting "break downs" within the systems, and are disappointed  when simulation doesn't portray it accurately.

 

You're right! Scrap the breaker modelling.

I guess what I'm really looking for is complex flight management a'la P-51 in DCS ... actually nearly all the modules in DCS (I own most) but, and it's a big but ... I'm not at all interested in the combat flight sim element and DCS doesn't have a civvie element at all.

 

I think some GA planes have that in X-Plane already or a just about there (AFL, Red Eyes, vFlyteAir) and SimCoders REP applied to Carenado planes I think is a good idea heading in the right direction.

[jcomm 5,925]

TBH, A2A's modeling of systems goes beyond DCS's IMH.

 

OTOH, the flight dynamics and overall physics modeling falls way behind what DCS fine tunned, custom developed models offer. A2A's P51d, or a recently P51d released for X-plane 10, can't really be compared to DCS's p51d in this area... Strangely, rotary wing perform a bit better in X-Plane than fixed wing, in comparison to that reference sim, but still not to the level of DCS's uh-1h, mi-8 or ka-50.

 

The simplest test I do to a new flightsim, when possible, is flying a good glider model. With 35 yrs flying gliders, I think i am credited enough to express my opinion, and for sure MSFS and X-plane have really lausy performance, even for the best models I tried on both. Even Aerofly 1 did a much better job, and none compares to the quality of Condorsoaring or SilentWings... Why can't a "simple", unpowered ( less effects to simulate ) be made to fly plausibly on this sims ? I really can't understand....

 

As a simmer, I was "forced" to use combat sims like DCS and il2 BoS because I can't find anywhere near quality in flight dynamics - maybe Aerofly 2 can offer some good alternative, but I do not own that one.... So, as much as I would like to use aircraft in X-Plane, I decided to completely give up and use the sim simply as a visuals generator, a task it does pretty much convincingly... and fly a Boeing 747 which is as detailled as it could be in a PC-based platform.... flight dynamics, systems and weather modeling wise ( I also benefit from the visual aspects of it's weather modeling in X-Plane ).

 

If a GA could be designed for X-Plane performing at least as good as the props in DCS, even if only flight dynamics wise, I would feel really glad about it....

 

Last time I spent some time and enthusiasm with the flight dynamics fine tuning in X-plane was with an Me 109 G-2 commercially available for X-Plane. I made some suggestions to the author which were incorporated in the first patch, and indeed the level detail, flight dynamics and systems wise, is very good, way above the best offer of a similar ww2 warbird for the MSFS and derived platforms, but still way behind DCS or IL-2 ....

 

I don't like to see the need to do things out of the sim core in order to enhance flight dynamics, just as A2A and PMDG do with MSFS and derivates... I accept it when it comes to complex systems modeling of course, but heck! A complex core like the one X-plane offers should, IMHO, do much better....

 

I'll wait and see what XP11 will bring of new in this area...

[Paraffin 2,197]

I think it's great that there are at least a few GA models with deep systems modeling, but if that were a baseline requirement, I wouldn't be able to fly some of my favorite planes in X-Plane because they're not as popular as the typical GA planes.

 

If the bar was raised that high for payware, we'd probably have a stable of only the most popular Cessnas, Pipers, and Beech planes that would sell in large enough numbers. We probably wouldn't have the planes I enjoy flying like the MU-2, the Turbo Goose, the LES DC-3, or the Twin Otter. These are what a systems nut would probably call medium-level sims at best, not actual study sims. But I love flying them anyway. I'm glad someone cared enough to model them even at this level, because they would probably never be done at all if the requirement was 100% systems modeling. 

[Jimmy RFR 406]

I can easily say I've spent much time enjoying releases of minimal complexity, just as I have spent months at a time enjoying 'ownership' simulations such as Marcel's DA20.

I personally enjoy having working breakers, even if the underlying systems that might cause them to pop aren't modeled. The fact that they aren't yet a guaranteed thing in every developers release means that I still have a little bit of a 'oh wow' moment when I explore the cabin of a new release and find them present. However, it's very true that the majority of the time they're simply eye-candy (mouse-candy?) and once discovered, they're not really used much.

A few things in my memory stand out though: On the Dodosim 206, there were only a few breakers modeled, but they were absolutely a key part of the startup routine. The fact that they were present AND intended to be used every time you started or shut down the 206 was wonderful.

Also, on the enjoyable Globe Swift that was released by AH for FSX/P3D awhile back, it had a single animated breaker on the panel that had an exposed LVAR but no functionality. It was for the electrical operation for the hydraulically activated gear and flaps. I coded a custom gauge for the Swift that allowed for wear and tear caused by flap over-speed to be persistent, and cause a partial flap failure where only one side would fail, which in turn caused an interesting amount of roll / yaw that had to be counteracted while attempting to raise the remaining good side. The catch was that I also caused that 3D modeled and animated breaker to pop, meaning that the pilot had to reset the breaker before trying to raise the flaps.. or even lower the gear for landing.

It was quirk, likely not a terribly accurate simulation, something that I couldn't really share since I heavily borrowed flap failure sounds from the RealAir Legacy, but it was enjoyable. And of course I would of never have been able to tie it into a working breaker, had it not been modeled in the first place.

I guess this really brings me into something I've been thinking about.

Where I really appreciate the work done by developers isn't necessarily into things that are simply for looks, but for things that teach us that there are consequences for not doing things correctly. The idea of persistence, even if it's very minor, is something that I think is a great area to be explored. One of the major strengths of Marcel's DA20, or the A2A fleet, is the enjoyment of keeping an aircraft in good running condition by treating it properly. One of my favorite memories of the first version of RealAir's Legacy was the first time I came in on descent far too fast, lowered the flaps, and was treated with the loud 'bang' and resulting flutter when they failed.

I think this is an area that GA developers could carefully explore. There's likely a fine line here - you don't want to alienate customers that don't care about such things or just want things to work. For all of the enjoyment I have gotten out of my A2A aircraft, I've also faced annoyance when the real world weather I was flying in gave me a very chilly morning, and I just couldn't get my 182 to start without killing the battery, forcing a trip to the hangar. There's a point where it likely becomes too real. But it's an interesting thought process, and there's likely numerous small ways for developers to introduce subtle (and optional) methods to reinforce proper operation.

On that topic, I was also recently thinking that perhaps a lot of this weight doesn't need to be carried by the developers. Wouldn't it be great if the platform itself did the heavy lifting? The portion of the XP11 features presentation about PBR got me thinking about this, when Ben was talking about developers not needing to create different reflective materials, but simply defining a set of criteria for how the material should reflect light and letting the rendering engine do the rest. The same theory could apply to systems - the idea being that a developer could provide a set of criteria for an aircraft (temperature limits, oil use, rate of accrued wear, anything really) and it could be up to the platform to provide the ability to track these things, in the same way that it might track fuel usage, for example, and all the developer might have to do is provide feedback (if desired) to the user.

Just thinking out loud there, and I know there's 'bigger fish to fry' but (in my mind at least) the idea sure sounds neat. :wink:

[Paraffin 2,197]

Where I really appreciate the work done by developers isn't necessarily into things that are simply for looks, but for things that teach us that there are consequences for not doing things correctly. The idea of persistence, even if it's very minor, is something that I think is a great area to be explored.

 

Persistence is an interesting idea, although if it's modeled realistically it would also include things like small reductions in maximum piston engine performance over time due to component wear. Also slightly lower maximum airspeeds as the rivets loosen up and the airframe goes a little bit out of alignment, etc. No older plane performs exactly like a brand new one fresh from the factory, which is basically what we get every time we load a model in our flight sims. I'm not sure how many people would want something like this.

 

There is a small nod to accumulated wear in the FSEconomy game, where the virtual dollar value of planes is higher when they're brand new (i.e. first appear in the database), and the value declines a bit as they accumulate hours logged in the game. You also have to pay for 100 hr. service, and even engine replacement after a certain number of hours. So you may still be flying a factory-fresh plane every time, but at least you're paying fees related to accumulated wear. Both the accumulated airframe hours and how close it is to the nearest overhaul or engine replacement affect the price you pay when you buy a "used" aircraft in the game. On the other hand there are no penalties for mishandling the aircraft in a way that would cause damage, so it's just a minor attempt to model real world impacts of plane ownership. 

[Slayer 190]

Comparing to DCS is tough as most of the flightmodel has to come from the aircraft developer not the sim as in FSX/P3D/X-Plane. DCS  have the same level of access to the real thing as A2A in most respects, that's why they are so good.