FSX, Crysis and framerates

[GHarrall 494]

If there is anybody on this thread that should know about what Flight Simulator should be capable of, it's Jean Luc.Reality XP has pushed the FS boundaries in so many respects already.Glenn

[Guest]

Well it says "over 8km", so that's not "up to" 8km. It's 8km and more! I really can't find much information on the limits either. Maybe there is no limit here?The other report says a view distance "up to" 60km, thats a lot. But let face it the engine is scalable and if some developer wants to license and make a flight sim, the technology will be there to do it. In fact a military training sim (called Vigilance) will be based on Cryengine so its possible that will include flight simulation.http://www.incrysis.com/index.php?option=c...ask=view&id=522

[Tabs 2,790]

>Last I checked>human bodies don't have cockpit panels with 20-30 gauge>systems interacting with the game. ;)And Flight Simulator doesn't have tons of weapons trajectory modeling, physics down to the way individual plants move when hit with something, AI enemies that learn and adapt to your tactics in order to hunt you down, a full blown real time lighting engine etc...

[WarpD 824]

Boy... some of you really haven't got an inkling. Honest.Multi-million dollar sims have issues with fidelity and they have C64 level graphics... FSX has pretty good fidelity AND fairly realistic graphics... for under $100.There is absolutely nothing about the CryEngine that implies it can support a flight simulation at the fidelity level of FS. Now, if you're looking for the flight sim capabilities of BF1942... it's probably a great start.Flight physics is radically different than ballistics physics.

[Guest]

I have sent a request to Crytek for information on this. If and when they reply, I will post a new thread on it.

[jcmissionary 6]

I've been playing Ace Combat quite a bit lately, and I was thinking the exact same thing, in regards to it's graphics engine. Maybe for FSXI......

[Holger 0]

Hi Jean-Luc,you bring up good points for outsourcing but the primary question, IMO, is whether there actually is an engine, existing or in development, that can be adjusted to the specific requirements of a global flightsim.With engines like those for Crysis or AceCombat, gameplay is initiated by loading a map and then the player moves through this static map. With FS, gameplay is initiated by loading data around the user aircraft but then the map itself starts moving as soon as the user aircraft moves and it moves across the continuous database of information that approximates a round Earth.So, yes, both approaches require only a limited amount of information to be hold/displayed at any given point in time but the basic difference is this static map vs. moving map approach (there's probably an appropriate technical term for "moving map" I'm not aware of). I'd be interested in seeing information that shows that FPS or single-theatre flightsim engines can indeed be extended to provide a moving map approach. Maybe RPG engines would work better? Or, alternatively, perhaps the current moving map approach of FS could be somehow changed to fit a static map engine?Cheers, Holger

[Guest weeniemcween]

Yeah, flight physics is a lot simpler, actually. And it's all just a chosen range of equations anyway, with some being more sophisticated than others, sometimes because what is simulated is inherently simpler, sometimes not. Just download the Brian Beckman (physicist at Microsoft) interview and you will see how simple aircraft physics are in comparison to tire physics.Multi milion dollar sims are designed to have realistic inputs and hopefully sensations but the demand for and prestige of graphics isn't there as it is for 3d games, obviously.There is absolutely nothing about CryEngine or Unreal 3 that implies they CANNOT support flight simulation. Human thought at its best is supposed to be about ingenuity and not saying it can't be done. Of course it can be done! Just look at what reality/airliner xp and airsimmer are doing for complex glass gauges by not using GDI+ (I know that airsimmer uses antigrain, for instance.)In flight simulator we have just barely begun to see HDR, but it's been around for years. The level of lighting detail in crysis is absolutely ridiculous and fsx doesn't even come remotely close, although it still certainly is pretty and the vc modeling is great. The reason why we dont have realistic looking lighting, autogen buildings, grass and other vegetation in fsx is because it's had fundamentally the same engine for years. It's time for a change and it makes sense to at least consider implementing/drawing on the top engines right now which are crytek and unreal 3, the latter of which has proved very popular with third parties and is very performant, albeit not as detailed as the former. For ACES not to entertain such things would be utter hubris or a perverse in house pressure from Microsoft.

[JSkorna 137]

Wow, you mean it acts the same way as FSX does? New games need new hardware with sliders full right? What a concept!Jimhttp://www.hifisim.com/banners/hifi-community-sigbanner.jpghttp://www.hifisim.com/

[LAdamson 240]

>Yeah, flight physics is a lot simpler, actually. And it's all>just a chosen range of equations anyway, with some being more>sophisticated than others, sometimes because what is simulated>is inherently simpler, sometimes not. Just download the Brian>Beckman (physicist at Microsoft) interview and you will see>how simple aircraft physics are in comparison to tire>physics.>IMO, way over simplified...Developers have been wrestling with sophisticated flight physics for many years now. It's almost a black art, where only those with years upon years of experience, seem to get it close to being truly believable!To pretend that a convincing flight model is just a compilation of equations ready to be thrown into a mixture of flight simmers brew, is an understatement! I've been around flight model beta testing for quite a long time, and have noticed that it's one of the harder things to get right. L.Adamson

[Guest weeniemcween]

I wasn't talking about the nuances of the flight model but saying that flight physics, contrary to what a lot of people believe, do not require a lot of computational power or a super special engine. I'm a huge fan of realair for the reasons you cite, but in terms of raw data it's all contained in a number of simple equations within the sim itself plus ingenious but still relatively simple-quantity of data not the underlying thoughts-realair definitions for table look ups (the air and aircraft.cfg files), tables being sufficient for plane physics but not tire physics. Again, what I mean is that bullet and car physics are more complex to simulate than plane physics, and that the engines being discussed could certainly accomodate plane physics, which would ideally be associated with specific flight models created by talented developers.

[WarpD 824]

>I wasn't talking about the nuances of the flight model but>saying that flight physics, contrary to what a lot of people>believe, do not require a lot of computational power or a>super special engine. I'm a huge fan of realair for the>reasons you cite, but in terms of raw data it's all contained>in a number of simple equations within the sim itself plus>ingenious but still relatively simple-quantity of data not the>underlying thoughts-realair definitions for table look ups>(the air and aircraft.cfg files), tables being sufficient for>plane physics but not tire physics. >>Again, what I mean is that bullet and car physics are more>complex to simulate than plane physics, and that the engines>being discussed could certainly accomodate plane physics,>which would ideally be associated with specific flight models>created by talented developers. Actually... ya know that new physics card you can purchase and install in your PC? It can't handle flight. ;) Guess flight's too simple an equation for it. ;)Aerodynamics with good fidelity is far more complex than the ballistics of a bullet. Far, far... did I mention far? more complex.

[Guest]

err, Brian Beckman is the physics expert. If he says its easy then I'll take his word for it.Now, the other thing, if you try the Crysis demo, you will notice the CPU's staying at a low usage. Kind of makes sense for the graphics card to generate graphics right? Well thats how its coded. It leaves a lot of headroom for "flight modelling" and such.

[Geofa 108]

I have to agree with Ed-I was scratching my head over that statement too. I've been studying aerodynamics for quite a while-there is nothing simple about them. Modeling them simpler than a bullet or a car?-I can't see how. Who is this Brian Beckman you are quoting?I tried the Crysis demo on my minimum spec machine-and that is what I got 1 fps using its autodetect. I didn't really expect anything different as that is what a minimum spec machine usually gets on any game-including fsx. The palm trees looked nice- I didn't really care for the deciduous ones-but admittedly I didn't really get a good look as at that rate of play I'd be here for quite a while.When I have the luxury of time next week I'll check it out more.http://mywebpages.comcast.net/geofa/pages/rxp-pilot.jpgForum Moderatorhttp://geofageofa.spaces.live.com/

[Guest]

1fps? Are you having a joke there? Try putting everything on medium for starters. My PC is over two years old, yet on med-high settings it is smooth and very good graphics. If you don't like those other palm tree shoot them down!