PDA

View Full Version : New FM quirks



ClnlSandersLite
06-20-2005, 03:07 AM
Well, there are a few things I thought I'd bring up, partly to make sure that I'm not just crazy.

I've been flying spit mk Vbs against bf-109 f-2s one on one in the qmb to get used to the new FM. I've got a complete handle on it now, still a little rough, but easilly servicible.

Anyways the goods:

1. Hispanos do more damage than they did in 3.04. I'm intimately familier with the ingame hispanos and they hit harder now, although I have no proof. I'll just say that I've mostly flown the lightnings and the spits since AEP. Maybe since they upped the damage of the german 20mm they figured that the hispano by comparison was **** so they brought it up to match? Just conjecture there. I've found that in most cases a single hispano hit on the 109 will take it out of the action by either puncturing his wings (see below) or blowing his whole tail section clean off. I give it about an 75% chance of a crippling blow per 20mm round impact up to 200 meters (not a scientific statistic).

2. The AI crashes A HELL OF ALOT MORE. Only about 1 in 5 flights do I actually validly shoot the bastard down (at all ai levels). I'll usually score light hits with my .30's but nothing that would cause a crash (I spray my .30s at long range to try for lucky hits, I only fire those twin 20mms when I get close and I'm pretty sure of the shot). Conjecture again, but I figure that the AI is expecting to pull one of it's old ufo maneuvers and can't so it augers in.

3. Wing hits are MUCH more deadly. I'm not talking about the kind of hit that snaps the wing, but the kind that puts big holes in it. If you hit another plane's wing (hard) in a high g turn, he'll spin out every time. Especially deadly at low altitudes. It also dramatically increases the chances of a stall/spin when you enter a high g turn as well.

4. I've noticed several little glitches in the FM. In almost all cases it's probably stupid things like rounding errors messing with the formulas. It is a beta flight model after all. Example; in the spit mk Vb, there have been several instances where I've been flying: 20-30% throttle, 100% prop pitch, rad open, flaps and gear down (this is a landing approach), trimmed to only need x axis input (no trim for that on the spit Vb) starting at 5000 feet (long approach), 100 mph constant IAS with a 2000 feet/min descent rate. All the sudden, (i'm going to look for a pattern in it but haven't noticed one yet) the nose will suddenly drop 10 or more degrees. No real reason to it, it just does. I have tracks of this but it's almost imperceptible in them because I correct for it very quickly now. Maybe I'll try and make one that I don't correct for it in. I've also noticed several such glitches in regular flying (or maybe it's the same one, I don't really know), but it's hard to describe, and usually I correct for it pretty much instinctively so I'd be hard pressed to name specific instances.

5. People complain alot about the qmb starts (tourque) but I really don't have a problem with them now. I just figured out about where I need to have my pedals at the mission start and correct till my plane's speed and engine stabilize.


Anyways, any thoughts? Any of it really worth reporting or whatever?

NonWonderDog
06-20-2005, 03:55 AM
The only really strange thing I've noticed (besides the AI crashing more) is the engine-off torque bug. The prop makes little/no torque when windmilling, but HUGE torque when stopped. It really seems like a divide-by-zero error in the torque model. Quite honestly I don't know how something this obvious got through QA, especially when lots of guys "testing" the betas noticed it before 4.00's "release." Dead-stick landings are certainly more interesting now...

The Ta-152H is also incredibly sensitive to wing hits now. I don't blame the AI if it crashes with a hole in the wing of this bird; I can barely keep it in the air, much less land. I don't know how long it's been like this, though. The late-war props are of little interest to me.

Lucius_Esox
06-20-2005, 04:17 AM
Errm, I'm not sure about the Hispanoes doing more damage.

Online things seem a litle more durable to hits from these than they were in 3.04.

ClnlSandersLite
06-20-2005, 04:18 AM
The dead engine thing IS irritating. What I do is lower full flaps asap so that I won't pick up enough speed for it to windmill. Then I just trim for dead engine.

Badsight.
06-20-2005, 04:42 AM
Originally posted by ClnlSandersLite:
2. The AI crashes A HELL OF ALOT MORE. this bears repeating

& ClnlSandersLite is noticing this from Just ONE senario

look at the coops happening online

the AI cannot handel this new FM

downplay that statement all you want , i dont care , im seeing 200% more unecessary AI crashes than before in this new FM

Stigler_9_JG52
06-20-2005, 11:35 AM
The very first thing I noticed with 4.0's new FM (and of course, I noted it with a grin) was that now the feel of flight is much closer to Targetware (http://www.targetware.net) than ever before. IL-2 planes used to have this knife-edge CoG pivot that would rock the plane along the wingline, like it was otherwise on rails. Now, the planes have more of a "wallowy" feel in all 3 axes, something Targetware has always had. Also, with the PFactor, the nose sways much more on takeoff than ever before, something that is even more evident in Targetware (since their ground model is unfinished and there isn't much friction with the ground). It's also evident in the torque effects during flight, where if you don't watch how much power you're pouring on, you can have some very uncomfortable "turtle" moments when you're nose high and attempting a switchback, hammerhead or immelman.

There have been some nice improvements to the FM overall, but I still see some old bad traits in there: such as, how planes can do long, sustained breakturns, then come out of them and go roaring up onto another plane's six when the two planes should be roughly co-E or close to it. Where does all this overtake speed suddenly come from?

Overall, engine heat and fuel considerations are still very basic; heat warnings seem arbitrary and scripted (why does a 109F overheat so much more quickly than any other 109, even if you're NOT in the desert?) and there's never any incentive NOT to go roaring around at 100% at all times. Also, all aircraft have the same radiator "steps", which is a GROSS simplification.

The FM is much better, but there are still some nettlesome issues. And one real BIG issue, the issue of no-icon visibility, still has not been properly addressed. The effect of this is that "the guy in the weeds" still holds most of the aces, the altitude advantage is for most intents and purposes nullified, and historically-proven tactics (target stalking from a distance, selection and attack from above) are next-to-impossible. You have NO HOPE of seeing a plane from above unless he is flying with a cloud in back of him. I flew a lot of "no visibility" servers this weekend, and had the usual frustrating episodes of "barely making out a distant dot" and losing it as I got [much] closer to it; failing to discern 5 or 6 planes in a fur until I was amongst them (THAT many planes in motion will be spotted during a focused vector search!), etc.

ClnlSandersLite
06-20-2005, 09:26 PM
this bears repeating

& ClnlSandersLite is noticing this from Just ONE senario

look at the coops happening online

the AI cannot handel this new FM

downplay that statement all you want , i dont care , im seeing 200% more unecessary AI crashes than before in this new FM

Most definately true. However, I believe it is more along the lines of a 400% increase.


The FM is much better, but there are still some nettlesome issues. And one real BIG issue, the issue of no-icon visibility, still has not been properly addressed.

Agree 100% I'd REALLY liked to have seen some of the long term nagging issues with the sim revamped. Especially since it has gone through 4 incarnations now.

However, I disagree with the engine heat thing. The overheat warning comes on when you reach a certain temperature. Certain engine settings X amount of heat /ms, while certain radiator settings cool down Y amount of heat per ms. I'm not 100% positive that air temperature and air speed affect it though.

Also, while the radiator system is simplified, I believe it works well enough. There is SOME variation among aircraft as well. Certain planes have NO rad settings, some have the 0-100% deployed system, some also include an auto system. I believe that on a couple of planes, closing the radiator also puts an armor plate over the system.

jarink
06-20-2005, 11:29 PM
Originally posted by ClnlSandersLite:
3. Wing hits are MUCH more deadly. I'm not talking about the kind of hit that snaps the wing, but the kind that puts big holes in it. If you hit another plane's wing (hard) in a high g turn, he'll spin out every time. Especially deadly at low altitudes. It also dramatically increases the chances of a stall/spin when you enter a high g turn as well.

I've noticed this as well. I like it; it seems to be more realistic this way.


4. I've noticed several little glitches in the FM. In almost all cases it's probably stupid things like rounding errors messing with the formulas. It is a beta flight model after all. Example; in the spit mk Vb, there have been several instances where I've been flying: 20-30% throttle, 100% prop pitch, rad open, flaps and gear down (this is a landing approach), trimmed to only need x axis input (no trim for that on the spit Vb) starting at 5000 feet (long approach), 100 mph constant IAS with a 2000 feet/min descent rate. All the sudden, (i'm going to look for a pattern in it but haven't noticed one yet) the nose will suddenly drop 10 or more degrees. No real reason to it, it just does. I have tracks of this but it's almost imperceptible in them because I correct for it very quickly now. Maybe I'll try and make one that I don't correct for it in. I've also noticed several such glitches in regular flying (or maybe it's the same one, I don't really know), but it's hard to describe, and usually I correct for it pretty much instinctively so I'd be hard pressed to name specific instances.

You weren't passing over a lake or river were you? I've noticed a "bump" now and then when passing over a shoreline at lower alt.

FritzGryphon
06-21-2005, 12:26 AM
I'm not 100% positive that air temperature and air speed affect it though.

It does, significantly. Overheat is much faster on hot maps as opposed to cold (even altitude and time of day), and low speeds cause overheat quicker as opposed to high.

For example, fly tight, slow circles at ground level on a desert map, and you'll find the motor breaking down in a few minutes.

SeaFireLIV
06-21-2005, 01:41 AM
I agree that AI are crashing more, but 400%?, I don`t think so. And what does this 200%, 400% mean exactly? Can`t you guys just say a number out of 100%? Or are we going to go 5,6,7 - 800% plus next?

I`d say AI crashes when lightly hit about 80-90% of the time. Which is too much. Sometimes, especially with Focke Wolves, they have lasted out the hit, continuing to move and avoid a crash. This could also depend on AI difficulty; I`m going to go off normal AI difficulty and see what High AI difficulty is like. Perhaps they will recover better.

But I do agree that the AI has a problem pulling out when lightly hit and he won`t bail. It`s obvious he wants to do some other manouever, but can`t so he just spirals in instead.

ClnlSandersLite
06-21-2005, 03:24 AM
You weren't passing over a lake or river were you? I've noticed a "bump" now and then when passing over a shoreline at lower alt.

Nope, but I noticed something about that landing approach I mentioned. It actually Noses UP violently. I correct for it instinctively (I wasn't noticing that part) and as soon as the effect passes, it drops sharply again.



it does, significantly. Overheat is much faster on hot maps as opposed to cold (even altitude and time of day), and low speeds cause overheat quicker as opposed to high.

I thought it did, however, I never timed it. Therefore I wasn't positive.



agree that AI are crashing more, but 400%?, I don`t think so. And what does this 200%, 400% mean exactly? Can`t you guys just say a number out of 100%? Or are we going to go 5,6,7 - 800% plus next?

I KNOW so. My numbers are derived from VERY simple estimates, and it does not always happen with light dammage. It'll happen with NO damage as well if you get on his tail, stay there, and don't fire. It also does not matter what difficulty the AI is at either. The simple fact is that the AI cannot handle the new FM because it's still counting on it's old FM.

Heres how I estimate 400% though. Before, 1 in 5 qmb one on one df's ended with the AI stupidly ploughing into the ground. Now, 4 in 5 end in said manner. Simple math.

Fractions:
1/5 and 4/5. Lowest Common Denometer = 5, no need to sort that out.
1*4 = 4
4*100 = 400
1/5 * 400% = 4/5
Erego, a 400% increase.

Do you believe that the math doesn't add up?

I do admit it's not a scientific survey. However, it doesn't take such a survey to tell the difference between a chicken pot pie and one of my mother's infamous "Chicken pot pie" like devices; or say The Wrath of Khan and Generations.

Badsight.
06-21-2005, 03:43 AM
Originally posted by SeaFireLIV:
And what does this 200%, 400% mean exactly? . 100% means twice as much

im seeing perfectly ok - unhit A/C manned by Ai going head-first into the ground

they cant handel the new FM when combined with G/A or their lead landing

its crazy , this patch is for online DF rooms

ploughman
06-21-2005, 04:20 AM
The AI's deathwish is very noticable. If I run out of ammo on a QMB I've discovered I can dispatch all remaining enemy fighters by literally chasing them into the ground. Given enough time, in a dog fight situation, they will all crash.

msalama
06-21-2005, 05:49 AM
All the sudden, (i'm going to look for a pattern in it but haven't noticed one yet) the nose will suddenly drop 10 or more degrees.

Hmm... well, FWIW: I've found out that with a c**ppy controller (like my joystick) the nose bumps violently when plane's vertical speed - regardless of arrival direction - approaches zero. And yep, it's been like this with all versions of PF I've played & no, filtering the JS signal doesn't help!

But it's not an input spike thing anyway IMO, because AC control surfaces don't flutter/move when it happens...

ClnlSandersLite
06-21-2005, 06:59 PM
mm... well, FWIW: I've found out that with a c**ppy controller (like my joystick) the nose bumps violently when plane's vertical speed - regardless of arrival direction - approaches zero. And yep, it's been like this with all versions of PF I've played & no, filtering the JS signal doesn't help!
It's not the stick. It only started happening when I installed 4.01m (I never installed 4.00). As I said though, I DO expect glitches in a beta FM.



its crazy , this patch is for online DF rooms
Most definately. The fix should be simple enough though. While I didn't write the game, I AM a programmer. Logically, the AI has a built in saftey altitude that it won't break on certain maneuvers. All that needs to be done is raise this number about 200 feet. To be impresise change:
If (alt <= 50){doClimb()}
to
If (alt <= 100){doClimb()}

zetareticulan
06-21-2005, 07:19 PM
I'm not sure if this is relevant to your discussion, but yesterday in a qmb with Beau over Crimea, I managed to make two naval vessels collide on the second pass.

ClnlSandersLite
06-21-2005, 07:33 PM
I'm not sure if this is relevant to your discussion, but yesterday in a qmb with Beau over Crimea, I managed to make two naval vessels collide on the second pass.

Now THAT'S strange. Vessels in the game are on a fixed path. Fire up the FMB and see for yourself. I'll bet that either they are unable to follow the path set for them now and you just got some wierd timing going on, or they've always done that.

If you remember where/when exactly it is, see if it always repeats itself.

zetareticulan
06-22-2005, 08:26 AM
yes, this minor quirklette is a regular occurence in a QMB over Crimea, for me leastways.

I don't remember seeing this behaviour pre http://forums.ubi.com/images/smilies/heart.gif4.01 http://forums.ubi.com/images/smilies/heart.gif. I wonder if anybody else is getting a similar picture?

If it's been reported in the bug reporting thread - apologies.

http://img181.echo.cx/img181/1728/c16yh.jpg

http://img15.echo.cx/img15/117/c43md.jpg

Cheers