PDA

View Full Version : Overmodelled Torque



sunflower1
10-20-2005, 09:13 AM
There are some pretty good pilots, smart fellows, talking about how there were some ridiculous torque and departure characteristics in the previous (4.01) patch.

I do not understand how anyone can read the literature that has come out of the people who flew warbirds and come away with the idea that we've ever had anywhere nearly enough torque and nasty departure characteristics.

Corsairs, according to Boyington's book, could torque roll and would if not treated very, very carefully. I believe I've read that a P-40 would too, on approach, if the throttle were handled badly, resulting in a dead pilot.

We don't have a single airplane that CAN torque roll or a single airplane that becomes even mildly challenging to fly when the throttle is treated badly.

Now we're even further from that.

I know, there are compromises to be made when playing on a PC and Oleg Maddox is the arbiter of the CPU cycle budget.

But stop screaming about how losing what little we had in the way of progressing toward having the capability/vice of torque rolling is progress toward realism. Please.

Airmail109
10-20-2005, 09:42 AM
To add to the above

Go and read some accounts of how much real ww2 pilots had adjust for torque expecially on take off. Even read Mark Hannas account of flying the BF 109.......judging by their descriptions if anything Torque is undermoddled. If you dont like the more realistic torque turn it off or play on an arcade server!

FatBoyHK
10-20-2005, 10:25 AM
It is a game of relativity. If one plane have crazy torque while the other don't in this game, but IRL that are actually fairly comparable, we have a problem.

mynameisroland
10-20-2005, 10:33 AM
Originally posted by sunflower1:
There are some pretty good pilots, smart fellows, talking about how there were some ridiculous torque and departure characteristics in the previous (4.01) patch.

I do not understand how anyone can read the literature that has come out of the people who flew warbirds and come away with the idea that we've ever had anywhere nearly enough torque and nasty departure characteristics.

Corsairs, according to Boyington's book, could torque roll and would if not treated very, very carefully. I believe I've read that a P-40 would too, on approach, if the throttle were handled badly, resulting in a dead pilot.

We don't have a single airplane that CAN torque roll or a single airplane that becomes even mildly challenging to fly when the throttle is treated badly.

Now we're even further from that.

I know, there are compromises to be made when playing on a PC and Oleg Maddox is the arbiter of the CPU cycle budget.

But stop screaming about how losing what little we had in the way of progressing toward having the capability/vice of torque rolling is progress toward realism. Please.


You ever try switching engine off in P39 in patch 4.01? Thats a severe case of uncontrolable torque roll.

faustnik
10-20-2005, 10:38 AM
Originally posted by FatBoyHK:
It is a game of relativity. If one plane have crazy torque while the other don't in this game, but IRL that are actually fairly comparable, we have a problem.

That certainly is true. It would have been nice to add torque effects to the a/c that didn't have them rather than take torque away from all a/c. I really liked the feel of the torque in 4.01. Not, only did it make the FMs feel more alive, it put some penalty on the yank & bank. In 4.01 it's pure T&B mania. (I'm not talking realism as much as feel of the sim)

Bearcat99
10-20-2005, 10:43 AM
Torque in this sim and most sims in fact is way undermodelled. I dont mind it at all. From all accounts I have read, from Geroge Loving, Bud Anderson, several pilots of the 332nd in print and in person, Bud Fortier and others... the P-51 had so much tourque that if you slammed the throttle forward you were gauranteed to flip... You had to throttle up very slowly. On the P-40 smaller pilots had to almost stand on the rudder on take offs.

sunflower1
10-20-2005, 11:40 AM
Yes Roland, I do like the way 4.01 P-39's roll when the engine is pulled back, P-40's too. I prefer the 4.01 flight model. On the opinion spectrum, count me at the other end from Bearcat in terms of enjoying having the finer points of a/c control modelled. I'd like it if a P-51 would rock over onto a wingtip and then onto its nose if the throttle is opened too quickly. I'd dance a little dance, even. I'd like it if the Corsair did what Boyington described- going sideways to the left, etc. I can only hope the state of the art gets there while I can still see the screen. And of course realistic torque and gyro effects could be scaled to make the game learnable and playable for a lot of people- just the way it is now. I just want more difficulty than I get when I have that switch thrown. Actually, any difficulty would be good, but in reality it just will change the planes I fly and what I do with them.

thomb314
10-20-2005, 12:17 PM
About the Mustang I have heard a couple of stories:

- Pilots have torque rolled into the ground when trying to go around, due to "slamming" the throttle.

- On take off, you can't apply full power before reaching 40-50 kts, if you want to keep it on the runway.

- If you fly straight up on full power, and hang in the prop, the Mustang will eventually start to torque roll and enter a spin.

Other planes:

- From Clostermanns book: on his first flights in the Tempest he had a hard time putting in enough rudder to keep it straight (even though full trim was applied). Also, his new wingman almost crashed into him on a formation take off, due to the same problem. I remember him writing he expected it to happen, so he tried to lift off early, allowing his wingman pass under him... (which he did)

- Video of the Bearcat "Rare Bear" from Oshkosh (early 90's), found on the internet. On take off: Watch the pilot applying a LOT of rudder all the way into the air. Even though he applies throttle real smoothly.

- Though you can't really compare models to real planes, I have similar experiences from flying my 1:5 scale radio controlled Mustang. I have to apply rudder all the way down the runway. It doesn't get "trimmed out" until it has built up some speed (more than lift off speed, that is). A friend of mine crashed his Bearcat model into the trees, the plane actually yawed about 30 degrees left as he abruptly applied full power.

I agree, that IF these "facts" are true, it would be nice to see them modelled in a sim soon. If these stories are representative, then here is what I think is wrong with the current FM (at best an educated guess):

Torque - yaw-effect: Probably undermodelled.
1. You cam slam the throttle to 110% from the start, no problem keeping the plane on the runway.
2. The planes seem to be "trimmed out" too soon, long before reaching take off speed. Should be necessary to hold rudder a little longer.

Torque - roll effect: Probably undermodelled. No problem slamming the throttle in almost any configuration and at any speed. Now, this should of course not be a problem at high speed, but at low speed I think it is undermodelled. The only time I have experienced anything like this was in patch 4.01m, near stalling speed with full flaps (in the Corsair).

Being a scientist, I know that statements such as these have to be taken with a grain of salt. What pilots say about their planes is often qualitative, it is not the result of scientific research.

Also, even if a person has 1000 hrs in a real Spitfire, that doesn't mean he is necessarily able to give a correct assessment of a flight sim after 5-10 minutes, or even a few hours. That is because a flight sim not only models the airplane, it also models how input from the pilot is to be interpreted. If you don't know the details about this, the basic principles of how the sim is modelled, there is no way you can say how good or bad it is.

In fact, the only person I would trust 100% when reviewing the flight model of any flight sim, is a person who not only has a lot of time in real warbirds. He should also have a very good understanding of the programming principles behind the flight model, have a lot of stick time in the sim, and finally, he should base his thoughts on scientifically proven, measurable facts, not "feels" or "looks".

One example of the necessity of how both a real pilot and a hardcore simmer could be wrong, failing to only use scientific facts:
Think about the "bouncing" / "wobbling" effect being discussed in several threads. If a real pilot sat down in front of the computer and threw the stick around a little, he would probably say there is no way that would happen in a real aircraft. He might not know that the movement he just did on the stick would require 20-30 pounds of stick force in the real craft. The "hardcore" simmer would also complain about the nose "wobbling" around, failing to realize the fact that if he had been in a real plane, his input on the stick would have been much smoother (due to heavy stick forces, or having learned to be more careful on the stick, to avoid being throwed around the cockpit all the time).

The only correct approach to decide whether this bouncing/wobbling is correct, would be to go out in a real plane and measure it, making sure that the input on the real stick is exactly the same as what was modelled in the sim. Not having done this, I am of course unable to tell you whether the wobbling is correct. All I am telling you is that if you actually measured it, you might be surprised.


My point is, even if you made the "perfect" flight model, it may have effects that don't seem real both to simmers and real pilots. (It's not supposed to be easy, is it?)

Finally I would like to add, I'm actually quite happy with PF. It probably has the best FM in a sim so far. Oleg and his team are doing a great job, really looking forward to BoB!

And I hope we will get realistic torque effects soon, if not in PF then at least in BoB http://forums.ubi.com/groupee_common/emoticons/icon_smile.gif

(This must be one of my longest posts ever, hope someone reads it...)

darkhorizon11
10-20-2005, 12:48 PM
Originally posted by Bearcat99:
Torque in this sim and most sims in fact is way undermodelled. I dont mind it at all. From all accounts I have read, from Geroge Loving, Bud Anderson, several pilots of the 332nd in print and in person, Bud Fortier and others... the P-51 had so much tourque that if you slammed the throttle forward you were gauranteed to flip... You had to throttle up very slowly. On the P-40 smaller pilots had to almost stand on the rudder on take offs.

I concur, the 180bhp Arrows I've flown are bad enough and their tricycle gear. I couldn't imagine how bad it must be in a 1500bhp Mustang or T-bolt.

Udidtoo
10-20-2005, 01:26 PM
Even after 3 hours nap and multiple stick setting experiments I'm just as let down as when we stopped for some shuteye this AM Clark.

I know its highly unlikely to happen but since we neither ever had any troubles with.50's I would be delighted if the entire squadron voted to revert to 4.01m

danjama
10-20-2005, 02:14 PM
Id like to see ww2 birds torque modelled accurately. I feel like this(4.02) is a step backwards after 4.01! If only it was possible to have torque accurately modelled. Truly it would enhancethe Sim experience.

Friendly_flyer
10-20-2005, 02:29 PM
I jumped right from 3.o4 to 4.02, and I can attest to the torque haveing increased a lot. Now, I have to apply full rudder to get the rather sluggish Hurricane in the air.

danjama
10-20-2005, 02:35 PM
You never played 4.01! Nuff said! (or the betas of 4.01, now they were something)

msalama
10-20-2005, 04:06 PM
Oh my.

Torque = length of *****

...by which we can conclude that there's never enough, is there?

joeap
10-20-2005, 06:06 PM
Honestly I still got torque??

DmdSeeker
10-20-2005, 06:29 PM
I guess I must suck.

I downloaded the patch (just to play on HL); but I can't really notice a difference.

VW-IceFire
10-20-2005, 08:31 PM
Originally posted by Friendly_flyer:
I jumped right from 3.o4 to 4.02, and I can attest to the torque haveing increased a lot. Now, I have to apply full rudder to get the rather sluggish Hurricane in the air.
4.02 has almost no torque at all...

Archangel2980
10-20-2005, 08:39 PM
Yeah I need my torque!!!

hobnail
10-20-2005, 10:36 PM
What a torque-fest.

GR142-Pipper
10-20-2005, 10:48 PM
Originally posted by sunflower1:
There are some pretty good pilots, smart fellows, talking about how there were some ridiculous torque and departure characteristics in the previous (4.01) patch.

I do not understand how anyone can read the literature that has come out of the people who flew warbirds and come away with the idea that we've ever had anywhere nearly enough torque and nasty departure characteristics.

Corsairs, according to Boyington's book, could torque roll and would if not treated very, very carefully. I believe I've read that a P-40 would too, on approach, if the throttle were handled badly, resulting in a dead pilot.

We don't have a single airplane that CAN torque roll or a single airplane that becomes even mildly challenging to fly when the throttle is treated badly.

Now we're even further from that.

I know, there are compromises to be made when playing on a PC and Oleg Maddox is the arbiter of the CPU cycle budget.

But stop screaming about how losing what little we had in the way of progressing toward having the capability/vice of torque rolling is progress toward realism. Please. You have a point. However, the problem comes in not only with the torque but with the associated trim capabilities of the respective aircraft. Even with aircraft that have rudder trim to counter the torque, on some aircraft (Yaks, LAs, etc.) if the yaw ball is centered and then an external view of the plane is taken you'll often see that the aircraft is still crabbing in many flight conditions. This inability to fly in-trim is simply not accurate. If the ball is centered the aircraft should NOT be crabbing...but they do in this game. If REAL aircraft are flown out of trim, the stall and departure speeds are higher (i.e. stalls and departures come sooner rather than later and tend to be more violent with less warning when they occur). In addition, the overtorquing would also occur at the strangest times (i.e. during turns when there was plenty of airspeed) and the torque effects could still be overly felt.

In summary, it's not just the magnitude of the raw torque characteristics that is at issue. The companion issue is the ability/inability to counter these effects with properly modeled trim and rudder authority. It is also my understanding that the developers have limited (if any) actual acceleration data for most of the aircraft represented. If true, this absence of Ps data also significantly impacts the ability to properly model torque effects in different flight conditions. So the bottom line being that a lot of what goes into (and out of) this game is done through sheer programming trial and error. Sometimes it's better to omit a characteristic than to try and model it with insufficient data that produces very strange, unnatural, and unpredictable results. In my view, Oleg and Company made the right decision to withdraw torque implementation to more manageable levels until the modeling technology and the flight data permit a more accurate representation of these effects. It was a smart decision as sometimes less is more...for the time being anyway.

GR142-Pipper

BuzzU
10-20-2005, 11:09 PM
You could always fly the P-38 if not having torque bothers you that much.

faustnik
10-21-2005, 10:18 AM
Originally posted by GR142-Pipper:
However, the problem comes in not only with the torque but with the associated trim capabilities of the respective aircraft. Even with aircraft that have rudder trim to counter the torque, on some aircraft (Yaks, LAs, etc.) if the yaw ball is centered and then an external view of the plane is taken you'll often see that the aircraft is still crabbing in many flight conditions. This inability to fly in-trim is simply not accurate. If the ball is centered the aircraft should NOT be crabbing...but they do in this game. If REAL aircraft are flown out of trim, the stall and departure speeds are higher (i.e. stalls and departures come sooner rather than later and tend to be more violent with less warning when they occur). In addition, the overtorquing would also occur at the strangest times (i.e. during turns when there was plenty of airspeed) and the torque effects could still be overly felt.

In summary, it's not just the magnitude of the raw torque characteristics that is at issue. The companion issue is the ability/inability to counter these effects with properly modeled trim and rudder authority. It is also my understanding that the developers have limited (if any) actual acceleration data for most of the aircraft represented. If true, this absence of Ps data also significantly impacts the ability to properly model torque effects in different flight conditions. So the bottom line being that a lot of what goes into (and out of) this game is done through sheer programming trial and error. Sometimes it's better to omit a characteristic than to try and model it with insufficient data that produces very strange, unnatural, and unpredictable results. In my view, Oleg and Company made the right decision to withdraw torque implementation to more manageable levels until the modeling technology and the flight data permit a more accurate representation of these effects. It was a smart decision as sometimes less is more...for the time being anyway.

GR142-Pipper

Pipper,

I can't believe it, but, you made me feel much better about the 4.02 FMs and the loss of torque effects. http://forums.ubi.com/images/smilies/blink.gif Very nice post. Hopefully the effect can be worked on and replaced in a later patch/addon. I think the torque effect was big in reducing the yank & bank silliness but, what you said about trim effect makes a lot of sense.

antifreeze
10-21-2005, 10:55 AM
Even with aircraft that have rudder trim to counter the torque, on some aircraft (Yaks, LAs, etc.) if the yaw ball is centered and then an external view of the plane is taken you'll often see that the aircraft is still crabbing in many flight conditions. This inability to fly in-trim is simply not accurate. If the ball is centered the aircraft should NOT be crabbing...but they do in this game.


I thought that the aileron is the 'proper' way to correct for torque roll. It makes sense to me that if the rudder and elevator trim is used in the absence of aileron trim, then the aircraft will side-slip (crabbing?).

No?

idonno
10-21-2005, 10:59 AM
Originally posted by Bearcat99: From all accounts I have read, from Geroge Loving, Bud Anderson, several pilots of the 332nd in print and in person, Bud Fortier and others... the P-51 had so much tourque that if you slammed the throttle forward you were gauranteed to flip... You had to throttle up very slowly. On the P-40 smaller pilots had to almost stand on the rudder on take offs.

Then that's how it should be in this "sim," otherwise, it's not a sim, it's a game.

When getting shot down means that my monitor explodes in my face and kills me, THEN we will have too much realism.

effte
10-21-2005, 12:23 PM
Originally posted by GR142-Pipper:
If the ball is centered the aircraft should NOT be crabbing...

Provided thrust and drag are symmetrical about the longitudinal axis in calibrated level flight, etc.



Sometimes it's better to omit a characteristic than to try and model it with insufficient data that produces very strange, unnatural, and unpredictable results. In my view, Oleg and Company made the right decision to withdraw torque implementation to more manageable levels until the modeling technology and the flight data permit a more accurate representation of these effects.

You know the approximate shaft horsepower of the aircraft simulated. You know the propeller RPM. Hence, you know the torque exerted on the propeller. A good idea about the moments of inertia/products of inertia can easily be deduced by creating a crude mass property model of the aircraft. Without this data, it is pointless trying to simulate an aircraft at all.

The modeling of how a rigid body will react to a given torque is very well understood.

Just what is missing to realistically implement torque modelling?

Jumoschwanz
10-21-2005, 02:17 PM
I would like to see the sim represent things as close to real as possible, otherwise it ceases to be a sim.
Why not make it as realistic as possible? Those who don't like it can simply turn it off with a switch under difficulty settings, just like they turn off lots of other realistic sim-features.

What worries me about this, is the up and coming BoB, might not be any better a flight sim than what we have right now, if it is compromised like this one for "gamers" then all it will be is another IL2/FB/PF with better graphics.
I appreciate the graphics we have, but they are good enough for me. What I really appreciate is flight-sim realism. And if BoB is not going to be any more of a flight-sim than this, I will be really dissappointed.

Jumoschwanz

GR142-Pipper
10-21-2005, 11:41 PM
Originally posted by Jumoschwanz:
I would like to see the sim represent things as close to real as possible, otherwise it ceases to be a sim.
Why not make it as realistic as possible? Those who don't like it can simply turn it off with a switch under difficulty settings, just like they turn off lots of other realistic sim-features.

What worries me about this, is the up and coming BoB, might not be any better a flight sim than what we have right now, if it is compromised like this one for "gamers" then all it will be is another IL2/FB/PF with better graphics.
I appreciate the graphics we have, but they are good enough for me. What I really appreciate is flight-sim realism. And if BoB is not going to be any more of a flight-sim than this, I will be really dissappointed.

Jumoschwanz Well be prepared for disappointment then. A major ingredient missing from this game appears to be accurate Ps (specific excess power) data. Without that, these measurements have to be recreated (and approximated). Without being too tutorial, Ps impacts acceleration/deceleration/energy sustainability and other flight characteristics which change rapidly as flight conditions change. Many here believe that simply knowing the wing area and engine horsepower is sufficient to produce an accurate flight model. What is also critical are accurate Ps charts in which to factor available energy as well as rates of energy change into the calculations. If it's true that accurate Ps charts are not available for most of the aircraft in this game and are still absent by the time that BoB arrives, then many of these flight model matters experienced today will carry over to the new game as well. It will be interesting to see how the developers deal with this issue with future products.

To balance the scales and dispite the annoying innacuracies, it's remarkable that Ps matters are done as well as they are given the absense of authentic Ps charts. This data is clearly among the very most difficult to recreate from scratch and then implement into a dynamic flight model which will run on a PC...and often not cutting edge PC performance machines either.

GR142-Pipper

effte
10-22-2005, 02:15 AM
While Ps charts may not be available, other information which will allow you to make a qualified guess as to the actual Ps at any given point is available.

For example the POHs often give the best climb speed at various altitudes. That is your Ps maximum point then. The known climb rates are your Ps. Knowing the stall speed and the airfoil shape, you can second-guess the lift coefficient curve. You know the SHP, prop efficiency can be estimated - alas, if you can reasonably guesstimate the drag you know your Ps.

Today, we like Ps charts as they are a handy way of representing the capabilities of the aircraft. Back then, it was perhaps not as common to publish them, but to operate the aircraft successfully they needed the same information. Thus, about the same information was published, if in different formats.

Badsight.
10-22-2005, 03:01 AM
Originally posted by Jumoschwanz:
And if BoB is not going to be any more of a flight-sim than this, I will be really dissappointed. instead of using a 3rd party game & adapting it to suit conditions (FB) , the next Maddox Games CFS is a complete in-house project scratch built to replicate combat flying

BoB should be the answer to the limitations FB has

ICDP
10-22-2005, 03:19 AM
Originally posted by VW-IceFire:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Friendly_flyer:
I jumped right from 3.o4 to 4.02, and I can attest to the torque haveing increased a lot. Now, I have to apply full rudder to get the rather sluggish Hurricane in the air.
4.02 has almost no torque at all... </div></BLOCKQUOTE>

This is very strange Icefire. I still have torque on 4.02, it seems the same as 4.01. I had the twitchy controls and got it fixed thanks to your suggestions.

If you want to test for torque get slow in the P51 (or most other fighters). At around 100mph IAS slam the throttle forward. For me this causes the P51 to flip onto its back unless I use a lot of rudder and aileron to counteract the torque. I also need to use full right rudder on take off for most fighters.

I really am not notincing any difference in the torque modelling between 4.01 and 4.02.

msalama
10-22-2005, 05:15 AM
I really am not notincing any difference in the torque modelling between 4.01 and 4.02.

My feelings exactly, and there actually seems to be _more_ torque in full-power situations now if anything! Just take up a Sturmovik and you'll notice this too - almost full rudder deflection is needed to keep the plane under control in v.4.02.

And no torque, say some. W00t http://forums.ubi.com/groupee_common/emoticons/icon_confused.gif

GR142-Pipper
10-23-2005, 03:31 AM
Originally posted by effte:
While Ps charts may not be available, other information which will allow you to make a qualified guess as to the actual Ps at any given point is available.

For example the POHs often give the best climb speed at various altitudes. That is your Ps maximum point then. The known climb rates are your Ps. Knowing the stall speed and the airfoil shape, you can second-guess the lift coefficient curve. You know the SHP, prop efficiency can be estimated - alas, if you can reasonably guesstimate the drag you know your Ps.

Today, we like Ps charts as they are a handy way of representing the capabilities of the aircraft. Back then, it was perhaps not as common to publish them, but to operate the aircraft successfully they needed the same information. Thus, about the same information was published, if in different formats. At the end of the day, the Ps charts must be derived if the aircraft are to be modeled. What you're suggesting is to allow license to base the charts on quite a bit of conjecture. And it is conjecture (power, drag, G condition, etc.).
This isn't bad given the lack of actual Ps data but it's conjecture nevertheless and likely to produce very strange results (to wit: what we got in 4.01).

This complexity is EXACTLY the reason why I'm all for having LESS aircraft rather than more. The resources simply don't seem available with Oleg and Company to: 1) initially derive this kind of detailed analysis for each of the aircraft and 2) to then provide the follow-on support to further refine these charts to improve their accuracy. It's a large task for even a few aircraft let alone the many, many types we have in this game.

GR142-Pipper

GR142-Pipper
10-23-2005, 03:34 AM
Originally posted by Antifreeze:
I thought that the aileron is the 'proper' way to correct for torque roll. It makes sense to me that if the rudder and elevator trim is used in the absence of aileron trim, then the aircraft will side-slip (crabbing?).

No? The problem is that some aircraft have elevator and rudder trim but not aileron trim.

GR142-Pipper

OldMan____
10-23-2005, 06:17 AM
Well poeple.. Let´s just enumerate what we think improved or downgraded in new FM.


My opinion.

TOruqe is worse (too weak) (never felt any uncontrollable torque as some people talked about in 4.01.. if you can´t control that you would never get your wings). But it still have more torque than 3.04...
HIgh speed stall is somewhat better since most planes do it now if pushed too far. (Although I liked my plane not snap spinning as it did... I like even more to see enemies plane do it now :P )
My first tests indicate some changes on DIVE acceleration (planes gain speed faster at first few hundred kph)
Roll inertia is much smaller... strange..
FW roll is corrected now
Not the 109 neither the Spit are " pull stick with full force" machines anymore (good)