PDA

View Full Version : IL-2 vs FSX



dirkpit7
08-24-2008, 02:47 PM
I know many people here fly FSX along with IL-2 and I would be interested to know how you see these two in comparision to each other. I got FSX a few months ago, along with my new relatively high end computer and have pretty much transitioned to FSX. The problems that I have had with IL-2 (due to Vista 64bit OS, maybe?) have also helped this. I was impressed enough to get a new yoke even.

Some may think that civilian flying is boring but I have found that there are many challenges there too, just different kind of them. Like managing navigation techniques, bush flying, takeoff and landing procedures etc.

Flying FSX has also pointed out some weaknesses of IL-2. For example some simplifying in the aircraft models (always identical nav lights, flap positions, lack of engine starting procedure) and strange view control (gunsight view), to name a few. Granted, there are things that IL-2 does better, too.

Graphics are certainly better in FSX, but this is not acually a major factor to me, because IL-2 is also quite good in this regard (amazing if you think it's age). But one thing I enjoy in FSX is more detailed ground and airfields, as opposed to the very generic grass strips often found in IL-2 (except the latest maps, so I've heard but haven't actually tried the 4.09b). Today I was flying FSX over Alaskan mountains and the view was simply spectacular!

About realism; this hard to me to comment, being no real pilot. Sometimes, FSX feels 'easy'. But I believe civ aircraft are notably easier than WW2 planes to fly in real life, too! And I'm quite sure that, in fact, IL-2 is also too easy, especially ground handling, takeoff and landing characteristic of many planes do not reflect what I have read (Bf 109 and Tempest, for example). I hope I will get a PPL one day, and I think flying FSX might give some clue what real flying is like. Even if it doesn't have the right 'feel', it teaches about the aircraft systems and theorethical stuff.

So what are your thoughts? Which do like better and why? I do realize these two sims are different in many ways, but it doesn't make comparision impossible, does it?

Also I'd like to hear how do you use FSX; in other words, what kind of planes you mostly fly. Warbirds we don't have in IL-2? GA aircraft? Or modern military stuff? I'm most into bush flying myself, and also like historic aircraft (like DC-3). I don't consider myself capable of piloting airliners yet, but maybe someday I'll do that stuff too. I have to say that air racing is quite good fun, too! (Acceleraton P-51 racer has over 3000 hp http://forums.ubi.com/images/smilies/winky.gif)

Lastly, maybe someone with RL flying experience has opinions to share about how these simulators compere with actual flying?

Chivas
08-24-2008, 03:59 PM
FSX is an awesome sim for general flying and the graphic are very nice above 1000ft. Personally I don't like FSX flying low as the buildings and trees seem to float on a low res ground texture. The only high res ground textures are on the airfields and probably in second party addons. I bought FSX because they were going to do a major upgrade to the graphics as seen in the Magic Screenshot but Microsoft as usuall dropped the project. I would imagine the low level terrain will be much better in their next flight sim.

You should try 409b for FB, the Slovakia map it is just beautifull.

Also check out the terrain in this combat flight sim coming out next year. http://ve3d.ign.com/videos/34584/Xbox-360/IL-2-Sturmovik-Birds-of-Prey
http://www.gametrailers.com/player/37361.html?type=
Its a console sim but if comes out before SOW I'm sure more than a few of us will try it out.

WTE_Galway
08-24-2008, 05:10 PM
I have no idea about the modded stuff because it doesn't interest me but as a former private pilot the standard IL2 FM's are much more realistic than FSX.

FSX is good for navigation and stuff but its FM when you get close to the edges (incipient stall, spin, even its behaviour in the last few seconds of a convential landing) is pretty poor.

Most modded FSX downloadable planes either use default FM's for a different plane or custom attempts that are even worse than the defaults though there are quite a few good ones if you can track them down. There was a cessna kicking about especially redone by an instructor for a aerobatic school for example.

I am told some of the commercial addon planes you pay money for have excellent FM's but have not tried any.

Stingray333
08-24-2008, 05:34 PM
yo, I fly some FSX, I too am interested in getting the PPL one day as well, so I thought it would be interested to get FSX and learn more about radio communications, navigating, instruments, etc.

I find WTE_Galways' comments about the flight model interesting. I don't have any experience flying a real plane, but I find the flight model in IL-2 to be much more fun, and feels "more realistic" to fly (again, even though I don't have any real experience flying a plane to provide a real basis of comparison). I am thinking of getting a BF-109 Add-on for FSX so that I can compare my experiences in IL-2 with FSX a little more closely with the same air craft.

I also find the FSX textures of the planes, ground and buildings to be kind of.. lacking. I am playing FSX Deluxe on Ultra High with most of the sliders going as high as they go on my new computer (E8400 dual core CPU, 4GB RAM, ATI 4870 GPU on 1900x1200 24" LCD)

That being said, the 3D cockpits are great,the detail on the airports in terms of runway signs, lights, etc is great. The weather is pretty cool as well, same with the civilian water/land/air traffic.

I bought the "Northern Adventures" addon missions and am working on them. I also bough the FSX Vancouver+ scenery, as I live in Vancouver, and thought it would be neat, and it is pretty cool.

I am interested in doing some Bush Flying as well dirkpit7, what type of bush flying do you usually do? Do you have any scenery/texture addons?

Stingray

JarheadEd
08-24-2008, 07:24 PM
I am in love with this little FSX Fokker. Free too. The Dr1 for Knights of the sky. I think its at M4T. EDIT: here it is. http://www.neoqb.com/fokker/

FSX And IL2 both have places on the HD. I enjoy FSX's weather and high altitude work much more than IL2's, and sometimes I do get a hankering to fly ruber dog shiite out of Hong Kong. Who knew.
The acceleration pack..........I do like the EH-101,...and abhor the Hornet (side note: I work on Hornets for Boeing, so I'm very critical)Get Acceleration for the air racing, it's a hoot.


http://img.photobucket.com/albums/v37/JarheadEd/DR1gennadichbetafsx.jpg

Chivas
08-24-2008, 08:38 PM
Yes, I forgot about the Knights of the Sky addon to FSX. They improved the scenery around the airfield in France , it actually looked quite good at low level.

dirkpit7
08-25-2008, 12:34 AM
Thanks for the comments, guys.

Stingray:
I have tried some of the missions for DHC-2 (wish there were more) and Maule for example, those are cool aircraft! DHC planes are my thing currently, I've considered buying Aerosoft's new DHC-6 Twin Otter. Yesterday I tried the mission where you fly Maule to the Denali Base camp and it is quite challenging! In fact, I crashed to the mountains http://forums.ubi.com/groupee_common/emoticons/icon_frown.gif, have to try it again.

No add-on scenery (or aircraft) yet.

JarheadEd: Thanks for the link, I've been thinking where to find that plane.

RockyAlexander
08-25-2008, 12:39 AM
I've never flown a tail-dragger, let alone a tail-dragger with 1500 horses under the cowling, so I can't speak for the realism of Il-2's flight modelling. However, while I haven't yet earned my PPL, I've been in the process of doing so for the past 3 years. I've logged a collective 9 hours (I know, not much) in Cessna 172s and a Tabago TB-10...that doesn't count the hour behind the controls of an old Beech Baron 55, which I didn't get to takeoff or land. Comparing that very limited experience to the virtual experience of FSX, the sim seems realistic enough to me, as far as taking off, cruising from one location to the next, and landing. But the truth is that it seems far easier to fly in RL than it is in either FSX or IL-2. In RL much is determined by the way everything feels. Your *** in the seat, your feet on the floor, your hands on the yoke, and the butterflies in your stomach: all essential in determining adjustments made during flight. When the fuel mixture is just a little too lean or a little too rich, you feel it, and therefore adjust until it feels right. If you're descending a little too quickly on your landing approach, you feel it. When everything is going well and you're doing everything right, it feels right. In virtual flight there is none of that, and it makes things all the more difficult.

STENKA_69.GIAP
08-25-2008, 06:12 AM
I fly light planes in Rl, fly a lot of IL2 and a bit of FSX.

As far as the general flight model and feel in the air FSX & IL2 are very similar and a useful practice for RL as far as you can go with no seat of the pants feedback.

The great weakness of the IL2 model is that its treatment of wind and weather is hopeless - as is its rendering of sideslip.

You can not practice a crosswind landing in IL2. You can do so in FSX, its weather control alows you to set up exactly what you want and you realy can do a crossed up approach.

Also the terrain looks far superior in FSX - not so lego toytown.

However, the way that FSX deals with the last few feet before touchtown is a bit suspect. I'm not sure if they modelled ground effect and as far as damage modelling you could call it non existent. What would be normaly a bit of a bumpy landing is transformed into a major disaster.

I suspect that the physics in FSX are biased towards a big Jet making a long straight shallow descent. It does not like a tight turn into finals, steep descent and flare to touch down. IL2 does this bit better although it is also more difficult than in real life with a light plane. Note that in both cases I am talking about light planes so in FSX I typicaly use the little Cessna and in IL2 the Otsu.

I find both products an interesting sim experience but am expecting that the next generation of software will address these known weaknesses.

sw25th
08-25-2008, 08:23 AM
FSX doesnt have problems relating to Wobbles. This game has some wobbles. Although very slight.

FSX cant do spins. IL2 Can. Flight sim has never been able to do spins and yet they have a larger budget and more money. Ever try to do a typical spin in a Cessna 172 in FSX? Wont happen.

I personally use IL2 for practicing landings and general flying awareness as VMC and tailwheel landing and numerous other aspects are very helpful for real life.

mortoma
08-25-2008, 08:59 AM
I used to fly MSFS ( as it used to be called ) all the way back when you installed it from two floppy disks. I think Microshaft had just acquired it from another company. This was about 1992 and then I bought it on and off all the way through FS2002 or so. I started buying X-plane ( www.x-plane.com (http://www.x-plane.com) ) about 1998 and got into it a lot more. It's got a lot more accurate flight modeling than MSFS ever had. The newest edition to come out is X-plane 9 and the graphics are starting to get better but still not suberb. I think the FM is more important than graphics and X-plane has versions that are approved by the FAA for flight training and those version have the same FM as the non-approved versions. Microsoft's sim is not and never will be approved by the FAA. They probably laugh at it.

dirkpit7
08-25-2008, 09:18 AM
Thank you gents, very interesting comments.

Talking about the landing flare, I have found it to be easier to do in FSX. When I rise the nose of a Cessna it still goes firmly to the ground, no bounces or floating usually. In IL-2 however many aircraft keep floating above the runway. This makes 3-point landing very difficult.


FSX cant do spins. IL2 Can. Flight sim has never been able to do spins and yet they have a larger budget and more money. Ever try to do a typical spin in a Cessna 172 in FSX? Wont happen.

I'm no expert, but I have understood that C172 was built 'not to spin', to fly out of a stall by itself. As opposed to Piper Tomahawk for example, which was purposely made able to spin, to offer a less forgiving trainer.

But I have seen this pointed out before so I think you're right about the spin modelling. I have to try the P-51 and see if I can get it to enter in spin.

dirkpit7
08-25-2008, 09:26 AM
Originally posted by mortoma:
I used to fly MSFS ( as it used to be called ) all the way back when you installed it from two floppy disks. I think Microshaft had just acquired it from another company. This was about 1992 and then I bought it on and off all the way through FS2002 or so. I started buying X-plane ( www.x-plane.com (http://www.x-plane.com) ) about 1998 and got into it a lot more. It's got a lot more accurate flight modeling than MSFS ever had. The newest edition to come out is X-plane 9 and the graphics are starting to get better but still not suberb. I think the FM is more important than graphics and X-plane has versions that are approved by the FAA for flight training and those version have the same FM as the non-approved versions. Microsoft's sim is not and never will be approved by the FAA. They probably laugh at it.

X-Plane is good no doubt and I believe those who say it's more realistic. I read an interview of X-Plane developers in PC Pilot where they explained the sim's engine and how it works, it was impressive.

But FS series is still preferred by most people, and has a huge number of third party add-on developers. Why is that?

waffen-79
08-25-2008, 10:36 AM
Dirkpit7 mate, just adding my two cents here http://forums.ubi.com/groupee_common/emoticons/icon_smile.gif

I own and used to fly regularily LOMAC, FS9, FSX & 1946.

I still fly LOMAC and 1946, nuff said.

But truth be told, my days in simming began with FS5 and then cfs and later cfs2

FSX better graphics?, I beg to differ, as many here say, flying down low is just not pleasant to the sight and above 2k feet, the so called photo-realistic scenary it's just ugly, and those clouds...well shader 3.0 in the planes making the crome shine, is the only thing good in fsx.

In FSX you can't do acrobatic stuff, even when flying the planes that are supposed to perform them.

My dream flight sim, would have a mix of the lomac and il2 graphics engine with ps 3 capabilities and a comprise of both FM's and plane sets

regards

dirkpit7
08-25-2008, 11:06 AM
Thanks for your comment waffen-79.

I think the way people think about FSX depends largely on hardware (especially when graphics are concerned). If you play IL-2 and FSX on older computer IL-2 may well look better because FSX needs a lot of resources to get the best out of it. FSX clouds certinly seem far better to me.

Why can you not do aerobatics?

Warrington_Wolf
08-25-2008, 11:11 AM
Originally posted by RockyAlexander:
I've never flown a tail-dragger, let alone a tail-dragger with 1500 horses under the cowling, so I can't speak for the realism of Il-2's flight modelling. However, while I haven't yet earned my PPL, I've been in the process of doing so for the past 3 years. I've logged a collective 9 hours (I know, not much) in Cessna 172s and a Tabago TB-10...that doesn't count the hour behind the controls of an old Beech Baron 55, which I didn't get to takeoff or land. Comparing that very limited experience to the virtual experience of FSX, the sim seems realistic enough to me, as far as taking off, cruising from one location to the next, and landing. But the truth is that it seems far easier to fly in RL than it is in either FSX or IL-2. In RL much is determined by the way everything feels. Your *** in the seat, your feet on the floor, your hands on the yoke, and the butterflies in your stomach: all essential in determining adjustments made during flight. When the fuel mixture is just a little too lean or a little too rich, you feel it, and therefore adjust until it feels right. If you're descending a little too quickly on your landing approach, you feel it. When everything is going well and you're doing everything right, it feels right. In virtual flight there is none of that, and it makes things all the more difficult. Agreed 100% I find flying circuits in FSX harder than flying in RL. When I did my first flying lesson my instructor said to me during a turn "you do a lot of flight simming, don't you", I replied that I do and asked him how he knew. He replied that when I started a turn I was focusing solely on the instrument panel and NOT where I was going, which is a danger.
I have found FS2004 and FSX useful for my training in other ways, such as practicing and memorising my emergency checklists, and navigation with FS2004 and the photo realistic VFR scenery and the UK airports addon. The weather in FS2004 and FSX is superior to IL-2 and once again I've found that downloading RL weather and using it for simulated navigation flights has been useful to some parts my training (learning how to use a flight computer, calculating my heading to give me the correct ground track and calculating ETAs for example).
An area that can be improved with FSX (apart from the FM) is having RT voices that suit the country where you are flying, I have "flown" into the airfield where I do my training in FSX and heard an American accent on the radio instead of a North West England accent such as Mancunian, Liverpudlian or Lancashire (I don't have anything against Americans before anyone thinks that, it just ruins the immersion factor a little). At least in IL-2 you have the voices that match the country that you are flying for, for example if you fly for the Luftwaffe, your wingies speak German or if you are flying for the Soviet Union your wingies will speak Russian. I know that all ATC units are supposed to speak English but the controller may still have a bit of an accent, especially if the country doesn't have English as it's native language. Another area that can be improved in FSX is the inclusion of helicopter lessons as well as the fixed wing lessons, I know that I would like to have a crack as flying the Bell 206B and I've had some luck in taking off but landing it again turns into a bit of a comedy show and I usually topple forward or back and end up crashing. FSX does include some training missions but IMHO they are not enough to actually be of any real value.

dirkpit7
08-25-2008, 11:14 AM
One thing I want to say, my intention never was to dismiss IL-2 or promote FSX over it. I like both of them. Just discuss the differences and what makes each one enjoyable.

dirkpit7
08-25-2008, 11:22 AM
Good points W_W about the language. I'd think that in all but large international airports pilots and ATC can use their native language.

I have flown the heli training missions and I think they do teach you something...The last is quite difficult, I haven't passed it yet.

Edit: About the language , have you tried to change the voice? It's possible to do in the sound options I believe.

dirkpit7
08-25-2008, 11:53 AM
More about stalls and spins...

I just flew the P-51 and it was ridiculous! No matter how much you throw it around it keeps flying nicely and executes every wildest manoevre easily. Not even a siqn of stalling let alone spinning.

I also tried low speed stall and no wingdrop or nothing...you can keep the nose up until the plane just starts falling and even after this.

So there is a problem indeed. In IL-2 you can stall the Mustang almost from full speed.

TheFamilyMan
08-25-2008, 01:14 PM
For me, FSX or FS2004 (basically MSFS) is far more comprehensive in its approach to flying. These sims cover all the facets of what it takes to fly civ. aircraft, such as flight planning, preflight checkout, engine starting, radio and navigation ops, takeoff and landing instructions, etc., along with realistically modelled terrain and a huge range of weather effects. I enjoy 'flying' MSFS for this level of experience/immersion which is somewhat lacking in IL-2. As for the actual flying, I can't say whether MSFS or IL-2 models what better for I've never flown an aircraft. As for the white-nuckled, adrenaline pumped experiences while flying, MSFS has nothing compared with what can be done in IL-2, and oh...did I mention combat situations, ordnance or damage modelling...?

As for which is a more realistic experience I find IL-2 too abstract from my reality, thus I have to say that it is a less realistic experience than MSFS, but maybe that is what makes it more fun and why I spend 10x more time flying in IL-2 than in MSFS.

mortoma
08-25-2008, 02:12 PM
Originally posted by dirkpit7:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by mortoma:
I used to fly MSFS ( as it used to be called ) all the way back when you installed it from two floppy disks. I think Microshaft had just acquired it from another company. This was about 1992 and then I bought it on and off all the way through FS2002 or so. I started buying X-plane ( www.x-plane.com (http://www.x-plane.com) ) about 1998 and got into it a lot more. It's got a lot more accurate flight modeling than MSFS ever had. The newest edition to come out is X-plane 9 and the graphics are starting to get better but still not suberb. I think the FM is more important than graphics and X-plane has versions that are approved by the FAA for flight training and those version have the same FM as the non-approved versions. Microsoft's sim is not and never will be approved by the FAA. They probably laugh at it.

X-Plane is good no doubt and I believe those who say it's more realistic. I read an interview of X-Plane developers in PC Pilot where they explained the sim's engine and how it works, it was impressive.

But FS series is still preferred by most people, and has a huge number of third party add-on developers. Why is that? </div></BLOCKQUOTE>X-plane also has a huge amount of third-party development. Why is that? MSFS only has more following because it's been out longer and has a better distribution chain because it's Microshaft. Does that make it a better sim than X-plane? I hardly think so. X-plane has been growing substantially in popularity and has taken many away from the MSFS series. The X-plane forums are chock full of people who've switched allegiance. And more coming everyday, literally every minute. Why is that???

dirkpit7
08-25-2008, 02:21 PM
Lower your rpm and open radiator man, you're overheating http://forums.ubi.com/groupee_common/emoticons/icon_wink.gif

My question was not meant to have a negative tone at all. Did you read the entire post? I was just wondering. I certainly don't claim FSX to be better than X-Plane (which I haven't tried).

I_KG100_Prien
08-25-2008, 03:35 PM
I've seen copies of X-plane at the local Wall-Marx. Imma going to pick it up if they have it the next time I'm there. Looks like fun. Especially the plane maker.. Reminds me of a MSFS version for Mac that allowed you to create your own aircraft. What a hoot.

WTE_Galway
08-25-2008, 04:31 PM
Originally posted by I_KG100_Prien:
I've seen copies of X-plane at the local Wall-Marx. Imma going to pick it up if they have it the next time I'm there. Looks like fun. Especially the plane maker.. Reminds me of a MSFS version for Mac that allowed you to create your own aircraft. What a hoot.

My limited understanding of the difference is when modding X-Plane you enter the physical parameters of the aircraft (things like wing camber and stuff) and it generates an FM.

FSX is like IL2 you directly create a FM based on your (mis)conception of what the plane should fly like.

Disclaimer: I have never tried to build a plane in X-Plane so might be totally wrong.

ElAurens
08-25-2008, 05:54 PM
I still fly 1946 quite regularly, I have FSX but have not touched it in months.

I was cruising over Lake Erie one day in some little civilian high wing thingie, A Maule if I remember correctly, and it hit me like a brick from 4 stories up. This is unbelievably boring. You can't blow anything up. There is no damage model. The ground textures and objects look like a third grader's crayon drawings. This isn't "flying", it's a button mashing procedure simulator, nothing more.

GatorSub1942
08-25-2008, 06:57 PM
I always get bored with games lik FSX because they make no attempt to immerse you into the career of a pilot. They're just a series of disconnected flights.

As for fighter planes, what's the point when you can't shoot anything?

I'm getting IL 1946 tomorrow, for only 10. Result! It's been a long time since I shot anything out of the sky!!

triad773
08-25-2008, 07:06 PM
Originally posted by ElAurens:
I was cruising over Lake Erie one day in some little civilian high wing thingie, A Maule if I remember correctly, and it hit me like a brick from 4 stories up. This is unbelievably boring. You can't blow anything up. There is no damage model. The ground textures and objects look like a third grader's crayon drawings. This isn't "flying", it's a button mashing procedure simulator, nothing more.

I quite agree El. I flew from near where I live in the midwest to California, learnt about navigating via GPS, flew around restricted zones in California. So FSX might be great for learning instrument flying and the like, it does have value in that. MS still doesn't have water right which is an immersion killer for me.

For blowing stuff up, FSX just don't float my boat. I had more fun on long patrols torpedoing unsuspecting Japanese ships in Silent Hunter IV- and the graphics are better.

In summary:
FSX = learning to fly and navigate around the world
IL-2 = learning more about history and blowing stuff up (FUN http://forums.ubi.com/images/smilies/partyhat.gif)

dirkpit7
08-26-2008, 07:25 AM
Originally posted by WTE_Galway:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by I_KG100_Prien:
I've seen copies of X-plane at the local Wall-Marx. Imma going to pick it up if they have it the next time I'm there. Looks like fun. Especially the plane maker.. Reminds me of a MSFS version for Mac that allowed you to create your own aircraft. What a hoot.

My limited understanding of the difference is when modding X-Plane you enter the physical parameters of the aircraft (things like wing camber and stuff) and it generates an FM.

FSX is like IL2 you directly create a FM based on your (mis)conception of what the plane should fly like.

Disclaimer: I have never tried to build a plane in X-Plane so might be totally wrong. </div></BLOCKQUOTE>

You are right. X-Plane has some kind of physics engine in it so the game 'knows' how the plane flies. Or so I've understood.

joeap
08-26-2008, 07:32 AM
Actually I always understood that Il-2 was more like X-wing than MSFX in being physics based rather than table based (like MSFS).

Stingray333
08-26-2008, 08:19 AM
Originally posted by joeap:
Actually I always understood that Il-2 was more like X-wing than MSFX in being physics based rather than table based (like MSFS).

In the end, they are all "physics based", there is no way to make a flight simulator that is not based on physics. They all integrate the differential equations of motion in order to give the plane motion. However, there are a number of parameters used in the equations that must be defined: area of wings and fuselage, coefficient of drag of wings, fuselage, etc. In order to solve the equations of motion these must be specified.

X-plane takes the 3D geometry of the the plane itself and all the parts,and actually attempts to model the airflow over each part of the plane to obtain the necessary drag coefficients. So, for example, the 3D geometrical model of the prop is used to calculate drag coefficients that are used in the modelling.

In IL-2, the same equations of motion are used, but the drag coefficient is a parameter that is solved such that the plane achieves the desired acceleration and speed parameters. The "correct" acceleration and speed parameters are all based on tables that Oleg had to find somewhere, and there is the possibility for mistakes.

Although the method used by X-plane sounds superior, (and obviously some think its superior), in order to get the calculation right it requires a very detailed 3D model of the plane and its parts. Even still, there is still the "physics model" that uses these parameters to determine the aircrafts motion. As there are things like turbulence and non-laminar flow, vibration, that are difficult to encapsulate in the physical model in such a way that makes them solvable. The result being that even though X-plane may use the exact geometry of the plane to try to setup the coefficients of the physics, at the end of the day, it is still requires a physics model that will contain some approximations.

They are still both "physics" based, just a little different in how they approach solving the problem.

As a physicist, I must say, that the table based approach looks to me like the best solution if your goal is to get the best flight sim model with the least amount of CPU cycles. You are using the known differential equations of motion, but are using a table of the relatively well known flight parameters of the plane to help you get some coefficients.

Although MSFS uses the same "table method", it is lacking is in the physics model itself. Some of the physics at some of the difficult areas to model, i.e. close to the ground where you have physical interaction with the ground, or at low speeds in the stalling regime.

Just because its based on a "table" doesn't mean its not based on physics.

Some will argue that one way is better and others will argue the other way. At the end of the day it is important to keep in mind that they are both just solutions to a coupled system of ordinary differential equations and that some steps must be done to solve them.

Now, many on this forum will argue that either data Oleg used is wrong, or that a certain plane doesn't "feel" the way it should. Even if the table Oleg used to determine coefficient so that the resulting motion of the plane matches the flight performance of an aircraft, the physical model is still an approximation, so the table could be right, but could yield results that do not feel right. Keep in mind that these are the same people on the board who will have a thread of 30 pages arguing about the placement of a bar in the cockpit of the FW-190 and when finally shown a photograph of a real cockpit showing that the bar is in that position argue that the photographer took the photo wrong.. anyhoo.

dirkpit7
08-26-2008, 01:38 PM
Originally posted by joeap:
Actually I always understood that <span class="ev_code_YELLOW"> Il-2 was more like X-wing </span> than MSFX in being physics based rather than table based (like MSFS).

http://forums.ubi.com/images/smilies/16x16_smiley-very-happy.gif

Like this (http://en.wikipedia.org/wiki/X-wing) ?

Just thought it was funny http://forums.ubi.com/groupee_common/emoticons/icon_wink.gif

dirkpit7
08-26-2008, 01:46 PM
Stingray, interesting stuff.

I wonder, as IL-2 and FSX both use the same technique for FM, why do they still 'feel' different?

Stingray333
08-26-2008, 02:28 PM
Originally posted by dirkpit7:
Stingray, interesting stuff.

I wonder, as IL-2 and FSX both use the same technique for FM, why do they still 'feel' different?

Hi dirkpit7,

Well, the answer of course, is that they feel different simply because they are different!

The physics model, aka flight model (FM), is the system of equations that is solved in order to determine where the plane moves and how it reacts to the various forces that are acting on it. For example, when you say input an action, like say, push an aileron, that action is passed onto the flight model. The flight model will then take that input and determine how the plane reacts by solving the physics.

Now, it is important to realize that the physics being solved in the flight model is an approximation to the real thing. The real physics is quite complex, to do it correctly would require solving the Navier-Stokes equation in 4 dimension (3 of space, 1 of time) over a complex geometry with a compressible fluid (air). This would require discretizing the plane using boundary elements, or finite elements and solving the differential equations. To do so would require a significant amount of computing horse power, and to do it on your PC would reduce the frame rate to less than 1 frame per hour (fph). Despite the amount of "realism" this method would indeed add, I don't think I would be happy playing a game at 1 fph, even though my inner mathematician would be satisfied. And again, there are some aspects of compressibility, turbulence and non-laminar flow that would have to be approximated, so, in the end, this would in some senses, still be an approximation to the real world: it is still a model, just a more complex one.

So the physics of the air passing over the wings and the various interactions of the air frame with the air are approximated in such a way that the resulting equations are much easier to solve, and give "approximately" the right behavior. Now, although Newton's laws of motion are well known, there are differences in the exact way that the physics model can implement their solution and how the various forces and influences acting on the air frame can be incorporated. So, the equations being solved by FSX, or CFS, or Aces High, or X-plane, or IL-2 will be different: they are all using their own physics model that they developed themselves. In some instances, such as the video game Cyrsis, a lot of work goes into the physics model and the development is complex, and is sometimes handled by an entirely separate company. So, because the physics models of all these games are slightly different in their approach and implementation, they will of course, be different in how they "feel"!

Now, As part of the physics model, there are some numbers that are required describing how parts of the plane interact with the air: coefficients of lift and drag, friction, mass, etc. These coefficients must be specified in order to solve the physics.

There are a few ways as to how to approach getting these coefficients. You can phone up Boeing and ask them what the coefficients are for their various planes and use them directly. Your success with getting this information from the various aircraft manufacturers will likely be a painful effort at best. Another way, would be using your knowledge of computational fluid dynamics and the known geometry of parts of the airframe, try to calculate yourself on a computer what these coefficients would be. This is the approach of X-plane. Because it is possible (in some instances) to have an idea of the flight characteristics of the plane, i.e. roll rate, yaw rate, turn and acceleration performance numbers, you can then use your physics model and use the inputs that would be for that airframe i.e. engine torgue, horsepower, wing and fuselage area, etc, and then play with the coefficients until the results from your physics simulation matches what you expect from that aircraft. This is the "table" approach used by IL-2.

Now, it is of absolutely importance to remember, that at the end of the day, the physics model in *all* flight simulators is an *approximation* to the real fluid flow problem! So, in the case of say X-plane for example, no matter how accurately you use the airframe to try to obtain those exact drag coefficients, your still going to put those coefficients into a physical model that is an approximation to real flight! This means that no matter how accurately you specify the various parameters in the aircraft to be as exact as possible, the flight model will still at the end of the day, be a "model". It will not be actual flight. This means that even if you can characterize the plane exactly, it will still not feel "exactly" like the real plane. It may, or may not, get close in approximating the "feel" of the real plane, and this depends on the accuracy, complexity and implementation of the underlying flight model. a.k.a how they solve the physics.

Now, with this in mind, this is why I like the approach of IL-2 better: Since the underlying physics is an approximation anyway, lets choose the various coefficients such that they give us the desired performance using the flight model of our game. This, in some sense, can "cancel out" some of the problems with the approximations used in the flight physics. I am sure that in their code, X-plane probably does a bit of this too.

Does that clear things up? I hope so, its an interesting topic nonetheless. Everytime I fly in IL-2 I think to myself: I am riding a differential equation.

Have you ever taken a calculus class with integral calculus? and thought: Runge-Kutta integration, mid-point approximations, integration, differentiation, ughghg!!! When the heck am I ever going to use this??? If you wanted to write a flight simulator, you would use them all.

Also, my statement about the people complaining about the FW-190 cockpits, I feel is a little harsh. I was just trying to make the point that a lot of people want the "details" right in this game, as it is a historic simulator. As such, they want the planes to "feel" a certain way, as described in historic accounts or by modern day pilots flying the warbirds. The point I was trying to make was that there are people who are never going to be happy until all the details are, in their eyes, correct. This is fine. But the fact remains that the underlying physics model will always be an approximation. Thus, no matter how accurate the information and exact as to the real specifications of these aircraft is that goes into the physics model, the physics model itself is an approximation.

This statement: "FSX is like IL2 you directly create a FM based on your (mis)conception of what the plane should fly like"

illuminates the problem that the approach of FSX and IL-2 have: you require knowledge of the exact performance of the aircraft to match some parameters of your flight model to, and if that information is incorrect, then the plane will of course not be correct in its handling.

The statement above itself implies that the approach of X-plane will solve this. It will not. The method of X-plane will require very accurate geometrical knowledge of the airframe, which may, or may not be available as the sands of time erases some of our knowledge of the past. And again, because the underlying flight model is an approximation, even putting in the right values of the various coefficients may not result in the correct feel and performance anyway!

Aaron_GT
08-26-2008, 02:38 PM
Just to emphasise the above, a 'table-based' model is, in essence, a pre-computed version whereas X-Plane uses a series of quite large finite elements (and I think FS:SDOE used a similar approach). The former would certainly be the best choice if you have to cope with lots of AI planes in the sky as doing finite element analysis on every part of every AI plane would bring a standard PC to its knees but isn't such a problem for a GA sim.

It seems that BoB:WoV uses even more approximation in that a formation, before it breaks up, uses the same calculation for the whole formation.

Given that things behave differently if you lose a wing in IL2 I presume the precomputation includes this as well.

In theory (given that you can read off the various force vectors in X-Plane) you could use X-Plane to generate a set of tables, and then change the inner loops of code in X-Plane to be 'table-based' and feed the information back in.

Stingray333
08-26-2008, 02:45 PM
From the website http://www.combatsim.com/review.php?id=736

Qutoe:

X-Planes website describes how they do this in the following way.

How it Works:
X-Plane reads in the geometric shape of any aircraft and then figures out how that aircraft will fly. It does this by an engineering process called "blade element theory", which involves breaking the aircraft down into many small elements and then finding the forces on each little element many times per second. These forces are then converted into accelerations which are then integrated to velocities and positions... of course, all of this technical theory is completely transparent to you... you just fly! It's fun!
X-Plane goes through the following steps to propagate the flight:

1: Element Break-Down
Done only once during initialization, X-Plane breaks the wing(s), horizontal stabilizer, vertical stabilizer(s), and propeller(s) (if equipped) down into a finite number of elements. The number of elements is decided by the user in Plane-Maker. Ten elements per side per wing or stabilizer is the maximum, and studies have shown that this provides roll rates and accelerations that are very close to the values that would be found with a much larger number of elements.

2: Velocity Determination
This is done twice per cycle. The aircraft linear and angular velocities, along with the longitudinal, lateral, and vertical arms of each element are considered to find the velocity vector of each element. Downwash, propwash, and induced angle of attack from lift-augmentation devices are all considered when finding the velocity vector of each element.
Propwash is found by looking at the area of each propeller disk, and the thrust of each propeller. Using local air density, X-Plane determines the propwash required for momentum to be conserved.
Downwash is found by looking at the aspect ratio, taper ratio, and sweep of the wing, and the horizontal and vertical distance of the "washed surface" (normally the horizontal stabilizer) from the "washing surface" (normally the wing), and then going to an empirical look-up table to get the degrees of downwash generated per coefficient of lift.

3: Coefficient Determination
The airfoil data entered in Part-Maker is 2-dimensional, so X-Plane applies finite wing lift-slope reduction, finite-wing CLmax reduction, finite-wing induced drag, and finite-wing moment reduction appropriate to the aspect ratio, taper ratio, and sweep of the wing, horizontal stabilizer, vertical stabilizer, or propeller blade in question. Compressible flow effects are considered using Prandtl-Glauert, but transonic effects are not simulated other than an empirical mach-divergent drag increase. In supersonic flight, the airfoil is considered to be a diamond shape with the appropriate thickness ratio... pressures behind the shock waves are found on each of the plates in the diamond-shaped airfoil and summed to give the total pressures on the foil element.

4: Force Build-Up
Using the coefficients just determined in step 3, areas determined during step 1, and dynamic pressures (determined separately for each element based on aircraft speed, altitude, temperature, propwash and wing sweep), the forces are found and summed for the entire aircraft. Forces are then divided by the aircraft mass for linear accelerations, and moments of inertia for angular accelerations.

5: Get Back to Work
Go back to step 2 and do the whole thing over again at least 15 times per second. Aren't computers great?

end quote.

Although the language may be intimidating, it is just going through the steps that X-plane uses to approximate and solve the physics of flight in its flight model. Also note that it also uses a "lookup" table method in its flight model.

Stingray333
08-26-2008, 04:23 PM
also, one more thing, I don't think I will be able to sleep at night unless I mention this:

Even though it is confusing, keep in mind that the words "finite elements" can mean two separate things depending on which scientific/engineering discipline is doing the talking.

The finite elements mentioned by Aaron_GT above, and in the description of the X-plane physics, is the notion of finite elements used by engineers. It is the idea that to compute the response of a complicated objected, you break it up into smaller simple pieces, and compute the response of each piece and add them up:

http://en.wikipedia.org/wiki/Finite_element_method_in_structural_mechanics

This can be done in real time, and it is used in some of the flight physics models. Although the wikipedia article above is for structural mechanics, a similar approach is used in aeronautics, see for example: http://www.aeromech.usyd.edu.au/aero/propeller/index.html

Now for the second definition of "finite elements":

The "Finite elements", which I speak of in my second extremely long and boring post above, where I mention that to solve the physics properly, you would need to solve the Navier-Stokes equation in 3D using finite elements. Finite elements in this instance is a framework for solving PDEs (partial differential equations).

http://en.wikipedia.org/wiki/Finite_element_method

It is used in computational fluid dynamics (CFD). It is not possible to do this in real time, and the technique is not used in the flight models. It used extensively in aircraft design and modelling, but it takes too long to solve for real time applications at this point.

Just as some people on this forum get their feathers ruffled if someone gets the details of the differences between a BF-109E4 and an BF-109E4B wrong, this type of mathematical detail gets my feathers ruffled... I apologize http://forums.ubi.com/groupee_common/emoticons/icon_smile.gif

Anyhoo, a very interesting topic. I hope the long and boring <STRIKE>dissertation</STRIKE> posts above were not too boring, and at least brought some insight into the discussion.

WTE_Galway
08-26-2008, 05:01 PM
Very informative information.


For me personally, in essence the only significant advantage I can see for X-plane is if you change the 3D model that you see visually you change the way it flies.

If you own an ultralight and want to see what putting massive floats under it will do then model it in xplane with and without the floats and take notes. You still need to test it in real life but according to all the claims you will still get an indication.

If you have all the physical dimensions of a rare prototype from WWI or WWII and no performance information, building it in xplane should indicate some things about how the real aircraft did fly.

In my opinion the table system is more suited to competitive but controlled sim environments like IL2 where the FM's are the result of extensive research and argument. On the other hand you could argue that MSCFS with its open ended unlimited modding where in theory you could build a Storch and give it a Dora flight model would have been better with the Xplane system.

joeap
08-26-2008, 05:14 PM
Originally posted by dirkpit7:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by joeap:
Actually I always understood that <span class="ev_code_YELLOW"> Il-2 was more like X-wing </span> than MSFX in being physics based rather than table based (like MSFS).

http://forums.ubi.com/images/smilies/16x16_smiley-very-happy.gif

Like this (http://en.wikipedia.org/wiki/X-wing) ?

Just thought it was funny http://forums.ubi.com/groupee_common/emoticons/icon_wink.gif </div></BLOCKQUOTE>

Hahaha, btw I am sure I read that Il-2 is NOT table based.

Well a quick search brought up these:

Tully post 1 (http://forums.ubi.com/eve/forums/a/tpc/f/23110283/m/9651080523?r=3901077523#3901077523)

Tully post 2 (http://forums.ubi.com/eve/forums/a/tpc/f/23110283/m/916008327?r=469005627#469005627)

Tagert!! (http://forums.ubi.com/eve/forums/a/tpc/f/23110283/m/1141088595?r=6161023695#6161023695)

Yet another? (http://forums.ubi.com/eve/forums/a/tpc/f/23110283/m/5711059124?r=9731098324#9731098324)

Ok so why are you guys suddenly saying Il-2 uses "tables" like FSX when ALL the info I posted before (I still can't find what Oleg said I think it may have been a loooong time ago or maybe elsewhere SimHq??) compares its method to X-plane???

joeap
08-26-2008, 05:32 PM
Another one

Early thread. (http://forums.ubi.com/eve/forums/a/tpc/f/23110283/m/85810694?r=50910694#50910694)


OK, let me add my take on this issue. This sim is not table based. In a table based sim it is possible to create a spreadsheet such as what the 1% group uses in CFS, and then use published data to engineer the data tables required for the flight models. The sim (CFS and virtually all other combat and civilian flight sims) provides the physics model, to which the 3rd party developers usually do not have access, while the data that give each airplane its separate characteristics is saved in separate files that can be edited by the community. In some cases, such as Falcon 4.0, the community did get a hold of the source code, but even there virtually all flight model edits (for instance in the SuperPAK series) is done in the data with very few changes to the actual physics model. Some sims have this data hidden in the executable files or binary files, but the key here is that the data and the physics model is kept separate.

In IL-2, "data" makes up only a small portion of the flight models (according to what Oleg has posted on the forums and also stated in some private conversations I and others have had with him). The rest is hard coded into the sim, so that you cannot really separate the physics model from the individual flight models. In other words, to edit the FM's, you need access not only to the littlebit of data there is, but to the full physics model and FM source code, and you need to be able to re-compile with your edits in order to test it. So, 1C does not only have to "unlock" the flight models, but they would have to share their inhouse development tools and source with the 3rd party developers to allow them to edit the flight models. For these reasons, I highly doubt it will ever happen that someone outside the 1C team will be in a position to edit the flight models directly.

I think the closest we will get is for people to compile data (such as what SkyChimp did with the P-47 roll-rate), and then send it to the developer to use as he sees fit, or within what is possible due to time and priority constraints.

The Man himself: got to read a couple of times (http://forums.ubi.com/eve/forums/a/tpc/f/23110283/m/6631080852?r=6631080852#6631080852)

Stingray333
08-26-2008, 06:32 PM
Originally posted by joeap:
Another one

Early thread. (http://forums.ubi.com/eve/forums/a/tpc/f/23110283/m/85810694?r=50910694#50910694)

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">OK, let me add my take on this issue. This sim is not table based. In a table based sim it is possible to create a spreadsheet such as what the 1% group uses in CFS, and then use published data to engineer the data tables required for the flight models. The sim (CFS and virtually all other combat and civilian flight sims) provides the physics model, to which the 3rd party developers usually do not have access, while the data that give each airplane its separate characteristics is saved in separate files that can be edited by the community. In some cases, such as Falcon 4.0, the community did get a hold of the source code, but even there virtually all flight model edits (for instance in the SuperPAK series) is done in the data with very few changes to the actual physics model. Some sims have this data hidden in the executable files or binary files, but the key here is that the data and the physics model is kept separate.

In IL-2, "data" makes up only a small portion of the flight models (according to what Oleg has posted on the forums and also stated in some private conversations I and others have had with him). The rest is hard coded into the sim, so that you cannot really separate the physics model from the individual flight models. In other words, to edit the FM's, you need access not only to the littlebit of data there is, but to the full physics model and FM source code, and you need to be able to re-compile with your edits in order to test it. So, 1C does not only have to "unlock" the flight models, but they would have to share their inhouse development tools and source with the 3rd party developers to allow them to edit the flight models. For these reasons, I highly doubt it will ever happen that someone outside the 1C team will be in a position to edit the flight models directly.

I think the closest we will get is for people to compile data (such as what SkyChimp did with the P-47 roll-rate), and then send it to the developer to use as he sees fit, or within what is possible due to time and priority constraints.

The Man himself: got to read a couple of times (http://forums.ubi.com/eve/forums/a/tpc/f/23110283/m/6631080852?r=6631080852#6631080852) </div></BLOCKQUOTE>

I like this quote from Oleg the best:

"User psychology is most often impossible to change. How soon they all forget what I have actually said, and instead misinterpret and misquote me to claim many things I have never said."

joeap
08-27-2008, 01:48 AM
Well what he said still proves you wrong to say Il-2 is "table-based."

Like this:
"3.. Flying won't become more complicated. It's only complicated to fly in the "trigger" or table-based or arcade flight simulators. I'm mentioning arcade here who those don't see it as such and do not see the physics errors which go against a pilot's nature. It's only easy to fly arcade for those who don't know how an aircraft really flies, or in short physics and a bit of aerodynamics. For example, I see absolutely disastrous errors in physics modeling in some sims, which makes it very difficult for me to fly and _understand_ as an aviation engineer.
...

So if some of you will continue to set a few other sims as examples of inertia modeling, I just can't take them seriously. For some of them them, the relationship is inverted - inertia changes seen in there are hard-coded for certain scenario, and are not a direct result of the airflow modeling or an
overall aerodynamics / flight model. Our sim is just not built this way, and will never be."

FatCat_99
08-27-2008, 07:00 AM
Originally posted by joeap:
Well what he said still proves you wrong to say Il-2 is "table-based."

Like this:
"3.. Flying won't become more complicated. It's only complicated to fly in the "trigger" or table-based or arcade flight simulators. I'm mentioning arcade here who those don't see it as such and do not see the physics errors which go against a pilot's nature. It's only easy to fly arcade for those who don't know how an aircraft really flies, or in short physics and a bit of aerodynamics. For example, I see absolutely disastrous errors in physics modeling in some sims, which makes it very difficult for me to fly and _understand_ as an aviation engineer.
...

So if some of you will continue to set a few other sims as examples of inertia modeling, I just can't take them seriously. For some of them them, the relationship is inverted - inertia changes seen in there are hard-coded for certain scenario, and are not a direct result of the airflow modeling or an
overall aerodynamics / flight model. Our sim is just not built this way, and will never be."

Every sim is Table based to some extent, there is no other way. Misunderstanding is here most likely in definition what is "table based" sim.

For example you can change drag coefficient of plane in Il2 and that will change it's performance.

Oleg probably think that some sims have predefined performance numbers and that in Il2 you are changing aerodynamic coefficients and than FM gives correct performance numbers.

FC

Aaron_GT
08-27-2008, 01:49 PM
The finite elements mentioned by Aaron_GT above, and in the description of the X-plane physics, is the notion of finite elements used by engineers.

That and the system for making PDEs soluble have the same inspiration - chopping up the problem into small pieces which are simpler problems with simpler boundary conditions. In a sense the engineering version is just a simple case where the equations to be solved are simpler, although the case of the stress of a plate with a hole in was always more challenging until recently solved.

In a flight sim I doubt there is time to do a full solution, every cycle, to do a numerical solution in fine detail, of the fluid flow given that typically it is done in non-realtime on the sort of cluster that might get into the top 500... So you end up doing simplified solutions on larger blocks and assume the values vary smoothly between the boundaries. Or that's my reading of what X-Plane does.

You could imagine precomputing values that say 'if airspeed is X, AoA is Y, then wing lift for this element is Z'. Lookups are faster than doing multiplications, but they are just cached multiplications in essence. It means more data to be stored in RAM.

The worst approximation would just look at the lift as a whole, not per lift element.

What IL2 uses I don't know, but it is easier to know what X-Plane uses as it is published in outline.