View Full Version : Defining FM

08-19-2006, 12:44 AM
People talk 'feeling of flight' and praise il2, and talk 'relative performance of aircraft' and praise TW, and both claim they are speaking to the FM differences... lot of confusion I think(just reading the 'feeling of flight' thread..)

TW I heard is 'table based', would mean in a general sense that while there will of course be some physics functions, the primary driver of aircraft performance is (essentially) look up tables. Eg aircraft A goes X kmh at Y altitude, rolls at Z rate with such and such a stick deflection. etc etc. Now it is not so simple as this I am sure, however to make a point, that would be the essence. So then to tune relative aircraft performance.. ...you change the tables... thats it. So if you know (or think you know) how one aircraft performed against another, why just change the table.

Il2 is probably (is said to be) more 'physics based'. Meaning (one hears) that the model is somewhat more of a plane and air, modelled directly. So with this general approach, to tune relative performance becomes <much> more difficult. Especially when (as in il2 from then to now) constant evolution of this physics model is ongoing, and is causing swings in relative plane performance, <in addition> to whatever tuning the developer may have in mind for an individual plane. And a tuning of the FM in il2 effects whole classes of aircraft, we've all seen it... a patch comes out and suddenly all the low wing loading aircraft gain a relative benefit when the global FM is tuned to be more sensitive to induced drag, for eg. So each patch a p40 against a zero is a new learning curve. However, in return, you get that feeling of flight, a tidy reward for the hassle for some of us.

So for 'accurate' and steady tactics patch to patch, look to the empirical model and TW... old steady... the P40 always just so against a zero... but for just flying around, and cool surprizes, and unexpected subtlety when performing complex manuevers... il2.

So when we speak of FM sometimes it is handy to distinguish between "relative aircraft performance" accuracy, vs "how a plane really flies..." accuracy, they are not the same. Any thoughts?

PS gonna be online tomorrow night probably on a small server, some of you real aces who've been posting around here forever ought to come on over and kick our butts around a bit... it is hard to get a nice fight on the WW servers and a person gets tired of always playing pit on against the sighted.... our little ntago server is an intermediate settings pit-on pure dogfight server, played with a certain discipline as to timing of takeoffs etc, to get 2 vs 2s or 3 vs 3s etc, ....if we see people taking the best we'll probably be in spits, otherwise well see what you all are flying and choose accordingly.

08-19-2006, 09:43 AM
I am hoping that BoB will have the stability of a table based syastem.. and the fidelity of the IL2 engine.

08-19-2006, 11:07 AM
Well thansk for keeping it on the main page Bearcat, your reply almost feels like a polite answer so as not to embarrass ol gkll with one of his classic 'what is he talking about now....' posts...

Really though two points maybe a little clearer..

- you can't have both. You aren't going to have stability and FM 'fidelity', they are somewhat opposed.... so is it worth it, the way il2 goes (and so will go BOB...)?

- the two approaches have different strengths, an accurate and stable 'tactical' environment (TW) vs a 'real plane' approach using raw physics, giving you better second by second flight characteristics.

08-19-2006, 04:10 PM
Originally posted by gkll:
TW I heard is 'table based'...

You "heard" wrong. Was this person, by any chance carrying pom poms and wearing a short, pleated skirt???

08-19-2006, 05:02 PM
The original table-based sims had many tables. One for each, a range of attitude and speed.
The lookup depended on your controls position and the result of the lookup gave the new attitude
and relative position and speed. Throw in some interpolations to shade out having fewer tables
than conditions and add in canned code for where the lookup tables leave off and that's what
state of the art was back 20+ years ago. And they didn't have megs of ram to throw around.

Of course the REAL work was in defining the tables and the results.

Advanced versions of that can be found in the Dynamix Aces series prior to Red Baron 2.