A New Simulator (June edition)

[Slater 0]

Here's a possibility I came up with.... Take Google Earth imagery, and set up an algorithm to allow for the photoscenery to be used as a template for applying default textures on to areas that look like that texture in google earth, for instance something that looks like a pasture may have a high resolution grass texture applied over the pasture in the orthoscenery

Something like an pretty much enhanced landclass. There's no need for users to download GE data. Only a precise and detailed landclass out of the box with beautiful grass, sand, swamp, pasture, tarmac, stones and other textures. And photoreal imagery available only for the main cities with the possibility to add more cities.

. Furthermore, place autogen based on regional and top down placement, allowing for the objects to be placed where the actual ones are in the photoscenery, and combine it with a regional autogen setup, where different building architechures will be used in the appropriate regions, along with different vegetation in the proper regions. On top of that, where grass texture is, put actual 3d waving grass shaders on top of the grass textures at low altitudes. This would allow for low level flight to show grass on the ground rather than just a texture. You could also possibly use the upcoming PhysX APEX for use for applying physics to vegetation among other things.

That would be so cool! Just imagine 3D grass geometrically instanced that can be seen fading in at certain LODs. Cool!

Use the GEBCO terrain data for the terrain mesh, and change the ocean to actually be water rather than just a texture. This includes caustics, allowing for realistic oceans and water over areas where for instance, you will actually see the depth of the ocean changing from above, allowing for the color of the water to change in the process of the depth changing. The GEBCO mesh is a combination of various bathymetry, SRTM, and Geotopo mesh data. This would allow for the most complete terrain mesh, and it would even be possible to merge greater detail terrain meshes than that using various tools into the mesh to make the mesh as accurate as possible.

I like these three suggestions better and it's possible, at reasonable performance, to implement them. For example if the ocean was actually volume with bouyancy and forces you could see your plane sinking in its waters instead of freezing with a "splash" message.

It's all about engineering. All Aerosoft needs is either a great engine and tweak it here and there to adaapt it for wonderful flight sim or hire some real good game programmers.

[Thaellar 23]

Please DO NOT have photorealistic scenery as a default. At low altitude the scenery becomes a flat, ugly, blurry mess.

[Aerosoft Team [Inactive Account] 51558]

Well...... Perhaps a really stupid idea, but have you thought of asking for donations. Perhaps there is some law against it, but it seems to me that Obama proved the power of soliciting the aid of like minded people.....

Well if you buy it when it is ready that will be enough donations for us. For the development we'll use our own money and will speak to the bank when we need more. A lack of money is not what will crash this project.

[Aerosoft Team [Inactive Account] 51558]

Going to comment on a few things because you sure got great comments here!

* Virtual cockpit technology that allows for high frame rate operation of sim AND gauges

I think by now it is well know how this should be done in FSX, you need to avoid gauges like the plague and build all functionality possible in the MDL file. See the Catalina, the VC is faster then a 2d panel could ever be.

* The hardcore simmers will hate this: A "smart speedup function". Right now the time acceleration in FS9 will result in deadly oscillations of the airplane. Switching to a simplified flight model during acceleration could be great.

* Hardcore simmers will hate this even more: A "Smart skip function". OK, so I am a working man. I often have only something like 15 minutes of free time every few days. Loading up the sim takes most of the time. Ever tried to make a decent flight with the PMDG747 in 10 minute segments? I propose a function that would allow the hobbyist simmers to skip the "less interesting" parts of the flight: Let's admit it... after switching the autopilot on, there is not soooooo much to do anymore. The function could be set up so that it the sim skips to any important event (Top of descent, a certain time [user defined], at crossing the ILS, in case of emergency, ...) and then pauses, waiting for user input. Of course this must be optional: Those who dont have time will skip some parts, those who want to do the whole flight won't have to use it.

Well I think they might actually like it. I think speed up will always have problems but an option to jump to 5, 3 or 1 mile to the next waypoint would most certainly make my flights a lot less boring!

* Hardcore simmers may like this even less: A "Quick interesting situations" collection, similar to the missions in FSX. User-Expandable, community (web) based ratings, ....

I like to hear more on this, how would this be different from a standard FLT file?.

* "Just in time" download of models & paints for online multiplayer scenarios: I propose an extension to the current airplane specification. In addition to the normal model&paints, every modeller should (or must?) include a "higher & lower polygon" model with "higher & lower resolution textures" to the package. The "standard model & textures" would be used for display as the actual aircraft in the sim (i.e. as it is right now). The other models would be uploaded to other players just before they enter visual range, so they would see the actual aircraft & texture instead of a standard one. Additionally I suggest the option of allowing players to select which model should be downloaded to their computer: None (display of an ugly standard model for very slow connections), low poly (for medium connections), high poly (very detailed model for fast connections).

I had something like this in mind but what if we force those simplified external models to be free, meaning that as you fly you slowly collect a set of low def models used as AI traffic?

Mathijs

[Dan2152 0]

Wow looking good!!!

I would love to see better and more accurate Flight dynamics for example thrust vectoring, side slipping, friction and air resistance.

I also agree with others and think that some form of damage would be cool, for example you damage an engine on takeoff (eg ground contact) and so that engine fails,(rather than you either getting a screen come up saying you have crashed or jumping 1000ft up like you do in FSX)

Advanced AI and ATC would be good, maybe with AI incidents/emergencies (for example overshooting and going around or landing gear failure and so that runway becomes unusable intill it is cleared.)

I know that many people have been calling for weapons but I would be happy with a proper dogfight training system like the versions that real airforces use. So with particular add-ons (eg aerosofts F-16) you can have special representational pods that, in real life, would send out signals to simulate missile lock-ons, estimate the trajectory of the 'missile' and work out whether it would have hit and 'killed'.(hope you understood what i meant there lol!!!)

[Dan2152 0]

...8. An easy to use editor for adding custom objects and airports to our world...

Hmm i definitly like that idea, SECONDED!!! LOL!!!

[Thaellar 23]

Probably already suggested but:

Saving cockpit layouts. I open some popup panels in my Twin Otter, move them around and resize them. Now I select "save layout" and save it a Twotter01 or Twin Otter Approach, etc. That way I can have saved layouts for all my different aircraft and call them up without having to recreate them all the time.

Thaellar

[newmanix 61]

May I please ask that Aerosoft Flight Simulator be able to run at MAXIMUM settings efficiency on the consumer technology available at the time of release? in 2011 or whatever...

As FSX still has it's users waiting for affordable "caught up" technology to fully see FSX's true potential... I still haven't bothered to buy a new desktop PC yet... And FSX has been around a few years now eh?

This one may be dumb, my girlfriend was watching me land the CS757 (i only fly in the VC) she asked: why is the cockpit empty??

How about a real virtual animated crew that really execute commands you assign prior to flight like FDC but as you can hear them you can see them taking action as well. If anyone owns the Wilco A400X, the VC copilot is an example, he is just useless...

[Hanimichal 77]

I need this acrobatic moviments

http://www.youtube.com/watch?v=5yDfIyNeAtE...feature=related

[John Gibson 1]

My wishes in a new FS would be:

1) Better Frame rates

2) ATC to be country/coninent specific (i.e. Asian accents in Asia, Germanic voices in applicable countries, British in the British Isles and Australian in Oz etc. Hearing American voices all over the flight sim world is unrealistic and spoils FSX. I know FDC fixes this problem but that is an add-on and is only useable in IFR flights.

3) Sids and Stars

4) Realistic and accurate representations of 3D cockpits in any plane that comes standard with the sim. Add-ons can handle the remainder.

Regards,

John Gibson

[Duco 5]

I think better wing flex and jetblast ( those heat waves behind the engine) would add some more realism...visually

I dont know if jetblast can be shown..and if it takes up too many frames, but im just throwing it outhere!

Duco

[=Hollywood= 0]

Hi Mathijs,

Have you looked into procedural rendering technologies at all? I've actually got a bunch of information that could be very helpful to you...

Regards,

Dean Mountford.

P.S. I emailed you as well yesterday so hope it came through ok...

[the nightfly 0]

Need I say that I'm thrilled to see Aerosoft considering (and that's the key word for the moment) such a project? However, if I dare risk the wrath of Mathijs for not contributing feature ideas myself, I think there's a more important question that needs be asked: Given that, if all (or even half) of the feature and performance suggestions here are implemented in a future Aerosoft Flight Simulator 1.0...how much are you willing to pay for it?

Most of us are used to shelling out for a new copy of MSFS every two or three years. Depending on which version, and where you purchased it, you probably paid anywhere from $30 to $60. Is that your idea for an acceptable range for the potential new product (considering that you'd likely then be spending the same amount for add-ons as that which you spent for each iteration of MSFS as well)? Or are you willing to pay more? What if it were to cost $100? $200? What if Aerosoft were to include all of the wish-list features, at a price of $500?

That's what, in the long run, Mathijs and his partners are going to have to ask themselves: is there a market for such a product at with a given feature-set and price-point? Is there a reasonable expectation of a profit? Face it, my guess is that it's going to cost Aerosoft a lot more to generate this than Microsoft would have needed for FS11, because so much of that would have been leveraged with code or design from earlier versions and paid for in part out of general operating expenses (such as already having office buildings, fleets of development machines, etc.) that Aerosoft would have to fund from scratch.

Personally, given our current economic situation (and my personal one), I'd be satisfied with a sim that equalled FSX with some bugs fixed and performance (frame-rate) improvement (and the possibility of greater future improvement through better multi-processor support) for around $100. If there were a lot of new, cool-sounding features, I might go as high as $149. But anything in the $200-$500 and above range would be too rich for my blood for the foreseeable future. Others might disagree. But what I don't want to see is everyone adding on wishlist feature after wishlist feature, only to complain loudly about Aerosoft being "greedy" or "trying to rip us off" if and when their "dream FS" comes out with a pricetag over $50.

So...how much are you willing to pay, and for what?

[massimiliano mancini 0]

Client server architecture, meaning environment (weather, airports, sceneries, vors freq, etc.) are given by a server while aircraft are given by client

[lonewulf47 83]

Hi Mathijs

I appreciate very much your (and your company's) guts to at least brainstom the idea of developing a future Flight Simulator. And with interest I read all postings for things that obvioulsy need to be imperatively included into a new Flight Simulator.

A huge amount of ideas. Some excellent and some going too much into details. I agree to all proposals that can focus on INITIAL requirements of a new simulator. One thing is for sure: it will be judged by the FSX standards and whatever can be achieved within (due to excellent add-on programmers like A2A, PMDG and others...).

So in my opinion it must feature everything that is already available (which is quite a lot, as many posters here might have overseen) PLUS:

Better aerodynamic model (be aware that even within x-plane with an admittently good aerodynamic engine some A/C fly like crap. A good aerodynamic model does not automatically imply perfect behaviour. It's what addon programmers make of it. but I agree that the THEORECTICAL CAPABILITES are better)

sloped runways (but please do not use x-plane as a reference. Those slopings are only mesh dependend and use the same "bending" algorithm as for the streets .. with sharp bends which are by no means realistic)

Updated NAV database (it is a pita to always use external programs for flight planning or to add/change navaids)

better multiplayer capabilities (the present quality of data transfer allows very accurate formation flying already, to set up a connection via GameSpy is the enervating part...)

Making use of today's PC standards (multicore processors, multi-processor boards) is something that should not be expressively asked for . This HAS to be included in any future program developments.

All the rest (and I mean ALL) can be achieved in future developments if the program architecture is reasonably selected. As for ATC my thoughts are a bit split. As I have spent over 30 years in real aviation I know that a realistic ATC will never be achieved. In this respect I fancy online flying as it is used on today's level (improvements are also still possible ... ) OTOH if AI traffic needs to be implemented (and it obviously needs AI traffic) there MUST be some kind of ATC .. at least in the background.

And .. PLEASE: maintain Multiscreen capabilities including detachable instruments/groups (I still love the unrestricted view outside on one screen and all the instruments on additional screens)

Oskar

[João Pinto 23]

For Me:

At last TRUE Fly-By-Wire flying and TRUE Auto-Trim ability for Airbuses and not some "close enough thing".

Helicopter dynamics as realistic as they can get and not a "rotary aiplane" feeling with some maneuvers like side slipping or rolls rendered impossible.

3D lightning effects that may actually interact with the scenery like real aircraft lightning

TRUE Mach speed aerodynamic effects

Aircraft wear (remembering the Aeroworx B200 engine hours accumulation, oil swap, etc)

Aircraft Stress causes lack of systems reliability like the Aerosoft DO26

So far, that's all I remember. The interest is to create a realistic flight environment based on my experience with FNPT's and other STD's

[bumprock 9]

....Personally, given our current economic situation (and my personal one), I'd be satisfied with a sim that equalled FSX with some bugs fixed and performance (frame-rate) improvement (and the possibility of greater future improvement through better multi-processor support) for around $100. If there were a lot of new, cool-sounding features, I might go as high as $149. But anything in the $200-$500 and above range would be too rich for my blood for the foreseeable future. Others might disagree. But what I don't want to see is everyone adding on wishlist feature after wishlist feature, only to complain loudly about Aerosoft being "greedy" or "trying to rip us off" if and when their "dream FS" comes out with a pricetag over $50.

So...how much are you willing to pay, and for what?

I suspect if we were to apply price sensitivity research to the flight simming market, such as Van Westendorp (too high/too low) or Gabor/Granger (I would buy X at $Y) we would find that the price curve of a Aerosoft's new simulator, with all the bells and whistles and even solving all of Microsofts problems, will actually be less, adjusted for inflation, than the price of the current product offering (FSX). This is how markets work, in fact, Aersoft's entry level offer will probably be a loss leader so that they can most rapicly acheive an installed base, their single most important key to success.

However, don't feel too bad for Mathijs, beccause I would also suspect that his offer and pricing will be as open and modular as his simulator. This way, what he give up to get this off the ground (so to speak) will have a big pay off with the add-ons and infrastructure.

SW

[panos95 0]

1) No screen freeze when I am right clicking -I've always hated this in FS2004 and fortunately it was fixed in FSX

2) As all the others have said, much better flight dynamics including ground friction

3) Take advantage of high-end cpu, gpu etc. For example I now have an Intel quad core QX6800 and a GTX295. However, FPS are horrible!

4) Better anti-aliasing

5) I'd like to see something like FSPassengers to be integrated and also work with VAs

6) Realistic weather and strong turbulence!!

7) I've heard that in FS2004 (maybe fsx too...) time messes up. I'd love to see this fixed

That's it for now....

[bandsma 11]

Given that a lot of the limitations of FSX right now are related mostly to hardware or lack of high quality ground photo or mesh data developing a new flight sim would not necessarily solve these issues. Third party add-on developers have done a phenomenal job tweaking the SDK to already produce a lot of the things the flight sim community have been asking for here. If you have already invested a lot of time and money upgrading FSX with add-ons it would require a significant step ahead in capabilities beyond what the existing add-ons can deliver to induce many to invest in a new flight sim.

I would be interested to hear what Aerosoft thinks it can deliver that can’t be achieved under the current FSX architecture. Best as I can tell the value added would come mainly from 2 areas: by setting a lower overall price level by integrating features in the basic software package that is currently achieved with FSX plus the currently available add-ons, or streamlining the code to achieve better FPS performance. Otherwise looking at the upward trajectory that the add-on industry is on, which Aerosoft is a big contributor of, every new release is more complex, more detailed, more feature rich and less resource intensive that the last so that it seems that we have yet to fully tap out the limitations of FSX.

[Versus 0]

I've fogot to add to the wishes list:

1) multimonitor function for all cockpit views and external views

2) easy-to-use camera with preinstalled views (availability to save views in airports)

[Bell417 0]

This one may be dumb, my girlfriend was watching me land the CS757 (i only fly in the VC) she asked: why is the cockpit empty??

How about a real virtual animated crew that really execute commands you assign prior to flight like FDC but as you can hear them you can see them taking action as well. If anyone owns the Wilco A400X, the VC copilot is an example, he is just useless...

I don't know about that, I don't like it when my viewpoint is perched on top of some head-less shoulders. Unless the others in the cockpit can be very realistically modeled, I think it would be better to leave them out. In my opinion, there is no point in having a superbly modeled VC when there is a bad mannequin sitting in the right seat.

[Arista 64]

Along with an updatable nav aid database, please don't forget to have an updatable magnetic variation databse as well. Sure, that sound like an obvious thing for a flight simulator, but for Microsoft, apparently, it hasn't.

2D cockpits: Don't even consider them, please. Instead, go for a 3D cockpit only design, but make them feel like the 2D cockpits some still love so dearly. For example, it should be possible to render (parts of) the 3D cockpit, without the ouside view, in more than one window, and on more than one screen. Also, each gauge should have its own 3D model. Ie. the knobs of a radio, for example, shouldn't be part of the underlying panel, like most of today's 3D cockpits are build; they should be an integral part of the radio itself. That way, gauges can easily be swapped, rearranged or removed as the end user sees fit, much like can be done with 2D panels today. That would probably need some kind of automated load-time (or creation-time) mesh consolidation and optimization for performance reasons.

This is more of a general thought: I would love to see some of the spirit and philosophy behind the Fly! series being revived in in the new sim. Fly! had a very thourough and in-depth system simulation of about everything in the cockpit. If it was there, it worked correctly, and was mostly full-featured. Sure, FSX has come a long way since, but almost exclusively because of a plethora of excellent add-ons, relying heavily on outside programming. But if the common systems simulation functionallity moved from individual gauge dlls to the sim itself, complex systems wouldn't be restricted to a few top-of-the-line payware aircraft whose authors can be bothered with advanced gauge programming - every aircraft would have that functionality. And, let's face it, which freeware creator is prepared to write a gauge controlling under which conditions the engine fires? Moreover, a bunch of redundant work could possibly be saved - how many different versions of the KX 155/165 NAV/COM receiver have been developed over time, for example? I envision a sim where aircraft and panels would become dumber again, or concentrate on the really advanced or unique stuff.

Which brings me to another idea: How about having engines as a self-contained entity, like panels or sound sets are today? Encapsulate everything that has to do with the engine - eg. under which condition it fires - in a separate unit, and just bolt it onto the aircraft like in real life. That unit would define some inputs it needs from other parts of the sim (eg. throttle position, OAT), as well as outputs it provides (eg. CHT, torque), and otherwise act as a black box an aircraft creator can hook up to controls and instruments. There aren't so many different engines, after all (in the GA world at least) - once created, many different aircraft could reuse the same engine over and over. Heck, why not do something similar with any complex system? If the modular concept would be taken to the extreme, every system, every instrument in an aircraft could be a separate module, with the master aircraft configuration file just hooking up as many systems as needed. Want a NAV 3, for example - no problem, just define three NAV radios. Sure, all this functionallity can't appear over night. But that's the beauty of such a massively modularized approach: not only can systems be added over time, but third parties could easily create them for the whole community to use, too. Of course, the question is whether we want an addon market where freeware plane A needs this engine by payware creator B along with that prop form payware product C, as well as D's radio stack and E's flight dynamics engine...

[Aerosoft Team [Inactive Account] 51558]

Need I say that I'm thrilled to see Aerosoft considering (and that's the key word for the moment) such a project? However, if I dare risk the wrath of Mathijs for not contributing feature ideas myself, I think there's a more important question that needs be asked: Given that, if all (or even half) of the feature and performance suggestions here are implemented in a future Aerosoft Flight Simulator 1.0...how much are you willing to pay for it?

Most of us are used to shelling out for a new copy of MSFS every two or three years. Depending on which version, and where you purchased it, you probably paid anywhere from $30 to $60. Is that your idea for an acceptable range for the potential new product (considering that you'd likely then be spending the same amount for add-ons as that which you spent for each iteration of MSFS as well)? Or are you willing to pay more? What if it were to cost $100? $200? What if Aerosoft were to include all of the wish-list features, at a price of $500?

That's what, in the long run, Mathijs and his partners are going to have to ask themselves: is there a market for such a product at with a given feature-set and price-point? Is there a reasonable expectation of a profit? Face it, my guess is that it's going to cost Aerosoft a lot more to generate this than Microsoft would have needed for FS11, because so much of that would have been leveraged with code or design from earlier versions and paid for in part out of general operating expenses (such as already having office buildings, fleets of development machines, etc.) that Aerosoft would have to fund from scratch.

Personally, given our current economic situation (and my personal one), I'd be satisfied with a sim that equalled FSX with some bugs fixed and performance (frame-rate) improvement (and the possibility of greater future improvement through better multi-processor support) for around $100. If there were a lot of new, cool-sounding features, I might go as high as $149. But anything in the $200-$500 and above range would be too rich for my blood for the foreseeable future. Others might disagree. But what I don't want to see is everyone adding on wishlist feature after wishlist feature, only to complain loudly about Aerosoft being "greedy" or "trying to rip us off" if and when their "dream FS" comes out with a pricetag over $50.

So...how much are you willing to pay, and for what?

Good post, thanks.

There is something you forget however. We sell you the sim cheap when needed because we want your credit card number to buy addons

But in the end this has to be a new platform for all to work on, that's our most important consideration. so the sim got to be affordable in the base version. You will then select the functionality you want (airports, aircraft etc, just as you do now) and buy those as addons. I hope we will able to offer existing customers low cost cross platform updates for products they got now. If you want Mega Airport Frankfurt for the new sim and you got it for FSX/FS2004 I think we should offer you the new version as a lower cost upgrade and not a complete new product. $49 is the sweet spot of course.

[ESzczesniak 25]

I admire Aerosoft's initiative and love their products, but to be honest I'm skeptical about the feasibility. Personally when I heard ACES was closing, I simply resolved that FSX was it and I'd make that the best platform. And truthfully, this isn't all bad. With the add-ons already out there like GEX and REX as well as some novel ones like FSAltitude, this simulator can be absolutely gorgeous. Add on to that the slow catch up of technology and I actually think I'd be quite happy just keeping FSX. I can't think of anything I'd want in a base simulator that FSX doesn't off--there's a lot I want, but I can't expect a PMDG level 777 or FSDT style airports to come with a basic simulator that I could ever afford. I expect that stuff to be added in aftermarket modifications.

May I please ask that Aerosoft Flight Simulator be able to run at MAXIMUM settings efficiency on the consumer technology available at the time of release? in 2011 or whatever...

As FSX still has it's users waiting for affordable "caught up" technology to fully see FSX's true potential... I still haven't bothered to buy a new desktop PC yet... And FSX has been around a few years now eh?

I've never understood this way of thinking. If you can run the simulator on maximum settings right when it comes out, that means it states becoming out-of-date the day after it comes out. Simply the simulator will never look better. True, it's frustrating watching slide shows of FSX, but I know as hardware catches up that it will continue to look a little better as I can bump up the sliders.

I think the key is to ask that the simulator is efficient so that we are not wasting computing resources and it looks as good as is possible on available computers. But if I can't get the sliders all the way to the right, it just means that it has continued growth as hardware catches up. I really think the problem with FSX is not that we can't have all the sliders to the right, but that it is an inefficient code that means the simulator either looks worse or runs worse than it should for any given computer setup.

[signmanbob 18]

Going to comment on this one as it is important.

I sure see the point you make and it is very valid at this moment. But how about in 7 years time? When have you last loaded FS2002?

The difference between FS2002 and what is possible now is mostly based on hardware. If we start a new game based fully on DX10 (or even DX11) things start to look very different. Just compared how other games increased in potential over the last few years. That's not because programmers are better now but because they got better stuff to work with.

I agree with this. As much as I like FSX, it will be outdated eventually. This concept is really needed badly.

Will there not be problems from Microsoft, since FSX addons will work with it? (hope not)

Mathijs, if anyone can do this, you and Aerosoft can.

The main thing that is important is that it is freely open to other payware developers. As much, at least, as is FSX now. If there is fees that payware developers would have to pay to Aerosoft to make their products work with it, it will discourage them.