My quick impression of the DCS Mustang

[159th_Viper]

...the speed at which the tailwheel comes off the ground is way faster than 40-50mph...

 

Have you any corroboration for the statement that the tailwheel comes off the ground at 40-50mph? If so, would you let me have it so I can look into it.

 

....Almost like 100mph....

 

Negative. At present tailwheel leaves ground at approx 60mph and is level (or as close as you want it) at 80mph, the transition, as seen, taking all of 2 seconds:

 

 

[Tailspin45]

...my comment was directed entirely at the simulator, in which the tail wheel seems to be something you Cannot Do Without...

 

No worries, I understood that.

 

 

Would you mind elaborating, preferably with a track to illustrate the 'problem' behaviour you are experiencing?

 

Will do.

 

... everyone who set out to learn the p51 went through this initially. shocked by the torque effects...

 

As a kid back in the late '50s I somehow managed to find myself in the control tower at La Aurora Airport in Guatemala City. In those days the Guatemalan Ar Force flew AT-6s and P-51s, and thus I was privileged to watch a new Mustang pilot make his first takeoff. It was thrilling for the reasons we're discussing here.

 

The tower was on the left side of the runway, and it turned out, directly in the path of the hell bent for leather Mustang and it's new rider.

 

The takeoff started okay, but as the tail came up the beast galloped toward the runway edge. As it came off the ground it had the tower cab boresighted, and the controllers decided they'd seen enough. One started to climb down the ladder that led to the floor below. The other impatiently took one look over his shoulder and dove head first through the opening, breaking his collar bone on arrival.

 

Ignorance being bliss, I stood there in awe, and thoroughly enjoyed watching and hearing the '51 go by at eye level within a wingspan of the cab. On the backside of the tower!

 

My brother and I still talk about our days playing ground hog at that airport. We found our way through the runway drainage system and would push the grates aside and pop our heads out at runway edge to watch the DC-3s and DC-6s arrive and depart.

 

Negative. At present tailwheel leaves ground at approx 60mph and is level (or as close as you want it) at 80mph, the transition, as seen, taking all of 2 seconds

 

I'll do the same test. I wonder if we had the elevator trim set the same? I'll double check and report back here.

Edited May 27, 2012 by Tailspin45

[159th_Viper]

I wonder if we had the elevator trim set the same?

 

For the above test I never touched the trim dials - I just spawned on the RW and took off. Trim settings are accordingly at default, whatever that may be - did not check.

[Tailspin45]

Well, I've spent the last several hours flying takeoffs with TO Assist off, and I'm still not sure what the problem is—but I'm more convinced than ever that there is one.

 

The problem could be 1) me, it could be 2) my set up, or it could be 3) the flight (ground handling) model.

 

1) I don't think it's me because a) I'm sober despite the holiday, and b) I have adequate RL experience in biplanes/SNJ/C-45 to understand takeoff handling considerations. That said, the lack of proprioceptive (seat of the pants) and kinesthetic (force feed back) queues certainly works against us all. But so far no one has found an affordable solution to that for home use.

 

2) I don't think it's my TM HOTAS or P-51 Options axis set-up because I've tweaked them several times to get a good 'feel' in other flight regimes. And from just airborne to 30,000 feet the aircraft is lovely.

 

3) So that leaves the ground handling model. Here's what I've discovered, but as a friend used to say, "I don't understand everything I know." And please keep in mind none of this is a complaint, just a effort to make this wonderful aircraft turn out as realistic as possible when released.

 

Using the Instant Action | Takeoff at Batumi setup (modified from 7PM to 4PM and with two 75 gal combat tanks added) I found the following:

 

a) The tailwheel exhibits a very strange behavior, visible in the attached replay--it apparently unlocks itself and starts to rotate*. On brake release at 30"MP, and 4 units nose heavy trim per the POH, the bird tried to head off into the weeds at low speed and was unrecoverable. After the pirouette, the tailwheel is backwards, so on next attempt, now with 4 degrees tail heavy trim, it rotates to the proper caster position with a little heading bobble as a result, but she is controllable.

 

b) Even knowing about the tailwheel weirdness, with 4 to 6 units of right rudder trim and with 4 units of nose heavy trim (combat tanks and fuselage fuel per POH), or with 2 units just to see if it's better, or with default zero units pitch trim, the aircraft is very unstable almost to the point of uncontrollable in the yaw axis. And that's with the stick back to keep the tail firmly on the ground and no gyroscopic effects because the pitch hasn't changed. Yes, with repeated practice you can control it, but I just can't believe that a real Mustang handles like that.

 

c) With 4 degrees of tail heavy trim the aircraft is much better behaved on the ground. Why would pitch trim affect ground handling? That suggests to me that we have a problem in the model.

 

So with 16 recorded tracks and at least an equal number that were too embarrassing to save, I believe that with TO Assist at 100% the aircraft flies more like I would expect than it does without it. But I could be wrong and have been before. I'm always willing to change my views.

 

Happy to share other tracks or record more if it would be useful.

 

 

 

*Yes, i checked to make sure there were no other controls or keys also bound to the rudder.

TO no assist 4 NH 4TH.trk

Edited May 27, 2012 by Tailspin45

[Shaderhacker]

 

Negative. At present tailwheel leaves ground at approx 60mph and is level (or as close as you want it) at 80mph, the transition, as seen, taking all of 2 seconds:

 

 

I don't get this. I've tried several times during takeoff and it seems to come up at 100mph. Also, at that point, it's useless to continue rolling down the runway as I am surely going to crash with all the gyrating going on. I keep the nose up and just takeoff at 100mph.

 

UPDATE: I agree 100% with the previous poster. There is definitely something wrong with the model on the ground. I can't seem to keep the plane straight no matter what I try unless I keep it tail heavy at which point, when the speed is correct, the plane will just lift off the ground WITHOUT rolling on just the 2 front wheels.

Edited May 28, 2012 by Shaderhacker

[Weta43]

I usually set 2 or 2 degrees nose heavy, let the tail come up at about 80mph, roll on 2 wheels till about 120 - 130mph, then lift off. No particular trouble keeping straight...

No assists...

[ErichVon]

this is the character of the plane. everyone who set out to learn the p51 went through this initially. shocked by the torque effects. 4 degrees of rudder trim is alot imo but essential to combat the torque. i also think taking off with full power causes the propwash to be way to powerfull to the extent the tail will just not come up smoothly, instead the bird goes for takeoff and if in that situation yes all hell breaks loose.

 

ive no idea about real flight physics but things seem good to me. very happy with the character of the plane. however wrong it maybe !

 

and no amount of virtual flight expereince prepared anyone for what to expect from ed regarding propellor driven aircraft and their real character attributes..... or did i miss a sim platform somewhere that got close....

 

I am not a real life pilot.

 

Wish a real life P-51D pilot would try this DCS_World: P-51D and comment on it. And Comment On It!!

 

The other sim I fly, the P-51D has much more torque on takeoff than the current DCS_World Beta #2 does. The other sim requires much more rudder control using the pedals when taking it out of Auto-takeoff mode.

 

Here, I fly full/real sim, no cheats. I add the recommended rudder trim. I add a little right aileron trim. I experiment with positive and negative elevator trim. I usually have the manifold at 46, but rpms are full on. It will takeoff all 3 wheels on the runway until airborne as well as a two wheel takeoff, but if two wheel, I use the rudder pedals as out of habit with the other sim.

 

It is very unforgiving taking off, guns only, not heavy, using one notch of flaps. I don't use flaps taking off with no ordnance.

 

I bet later Betas and the finish product will have more torque.

 

I do pretty good with this plane all around in DCS_World. Last night online, just using the .50 cal guns I killed 10, probably lightly armored, moving tanks or similar that were firing at me. I rtb and landed a perfect 3-point landing at Batumi.

 

Wish the sim online could retain an ongoing point system at least monthly as scoring to compete against others for Ace of the Base.

 

I would think there has to be differences in performance compared to the differences in computers, their setups, their different controllers, and if online---their ISP speed and ping hops.

Edited May 28, 2012 by ErichVon

[159th_Viper]

I am not a real life pilot.

 

Wish a real life P-51D pilot would try.....

 

A RL Pilot did. He was happy.

[JimMack]

A RL Pilot did. He was happy.

 

From RL Pilot, there are still some corrections needed. That is why we are still in beta!

[VH-Rock]

It's a very impressive sim so far, I'm looking forward flying the final version. It would be interesting to see one of the Mustang pilots (Maybe Nick Grey or Ed Shipley) to fly the sim in the TFC hangar at Legends this year?

[159th_Viper]

Well, I've spent the last several hours flying takeoffs with TO Assist off, and I'm still not sure what the problem is—but I'm more convinced than ever that there is one.

 

The problem could be 1) me, it could be 2) my set up, or it could be 3) the flight (ground handling) model.

 

1) I don't think it's me because a) I'm sober despite the holiday, and b) I have adequate RL experience in biplanes/SNJ/C-45 to understand takeoff handling considerations. That said, the lack of proprioceptive (seat of the pants) and kinesthetic (force feed back) queues certainly works against us all. But so far no one has found an affordable solution to that for home use.

 

2) I don't think it's my TM HOTAS or P-51 Options axis set-up because I've tweaked them several times to get a good 'feel' in other flight regimes. And from just airborne to 30,000 feet the aircraft is lovely.

 

3) So that leaves the ground handling model. Here's what I've discovered, but as a friend used to say, "I don't understand everything I know." And please keep in mind none of this is a complaint, just a effort to make this wonderful aircraft turn out as realistic as possible when released.

 

Using the Instant Action | Takeoff at Batumi setup (modified from 7PM to 4PM and with two 75 gal combat tanks added) I found the following:

 

a) The tailwheel exhibits a very strange behavior, visible in the attached replay--it apparently unlocks itself and starts to rotate*. On brake release at 30"MP, and 4 units nose heavy trim per the POH, the bird tried to head off into the weeds at low speed and was unrecoverable. After the pirouette, the tailwheel is backwards, so on next attempt, now with 4 degrees tail heavy trim, it rotates to the proper caster position with a little heading bobble as a result, but she is controllable.

 

b) Even knowing about the tailwheel weirdness, with 4 to 6 units of right rudder trim and with 4 units of nose heavy trim (combat tanks and fuselage fuel per POH), or with 2 units just to see if it's better, or with default zero units pitch trim, the aircraft is very unstable almost to the point of uncontrollable in the yaw axis. And that's with the stick back to keep the tail firmly on the ground and no gyroscopic effects because the pitch hasn't changed. Yes, with repeated practice you can control it, but I just can't believe that a real Mustang handles like that.

 

c) With 4 degrees of tail heavy trim the aircraft is much better behaved on the ground. Why would pitch trim affect ground handling? That suggests to me that we have a problem in the model.

 

So with 16 recorded tracks and at least an equal number that were too embarrassing to save, I believe that with TO Assist at 100% the aircraft flies more like I would expect than it does without it. But I could be wrong and have been before. I'm always willing to change my views.

 

Happy to share other tracks or record more if it would be useful.

 

 

 

*Yes, i checked to make sure there were no other controls or keys also bound to the rudder.

 

To avoid any misunderstanding, is it your submission that the ground-handling characteristics you are having trouble with can be attributed to the tail-wheel alone or are there other factors at play too? If so, I assume they are only limited to B above. In that respect and with due regard to your track, when exactly does this behaviour manifest? Let me know and I'll have a wee looksee.

 

Tail wheel issue is, as said previously, been investigated. I have again enquired into the issue and will let you know all I can when I know.

Edited May 28, 2012 by 159th_Viper

[RogueRunner]

The problem with a simulation is that you react on visual cues, ie the nose going this or that way on takeoff. In the real world you'd feel that long before you see it and correct accordingly. In the sim however you are kinda behind, ie things already happened when you get round to correct it hence the unusual oscillation some make on their takeoff runs.

[159th_Viper]

The problem with a simulation is that you react on visual cues, ie the nose going this or that way on takeoff. In the real world you'd feel that long before you see it and correct accordingly. In the sim however you are kinda behind, ie things already happened when you get round to correct it hence the unusual oscillation some make on their takeoff runs.

 

Precisely.

[Yo-Yo]

The problem with a simulation is that you react on visual cues, ie the nose going this or that way on takeoff. In the real world you'd feel that long before you see it and correct accordingly. In the sim however you are kinda behind, ie things already happened when you get round to correct it hence the unusual oscillation some make on their takeoff runs.

 

That's a good point I am always trying to explain. If you use TO assistant having full info of transverse acceleration take a careful look to its input.

If you rotate plane into 2-point position its input depends on how fast you do it and at what speed.

 

By the way - where is a NACA report about flying quality of XP-51 containing a record of a TO. Note that there was lighter 3-blade prop and Allison engine. One can see how the pilot worked with pedals.

[Tailspin45]

To avoid any misunderstanding, is it your submission that the ground-handling characteristics you are having trouble with can be attributed to the tail-wheel alone or are there other factors at play too? If so, I assume they are only limited to B above. In that respect and with due regard to your track, when exactly does this behaviour manifest? Let me know and I'll have a wee looksee.

 

Tail wheel issue is, as said previously, been investigated. I have again enquired into the issue and will let you know all I can when I know.

 

No, it's not a tailwheel issue. I mentioned that in paragraph a) but see now—because of your comment—that it has been reported elsewhere as well and apparently submitted as a bug. Thats great.

 

However in c) I asked, "Why would pitch trim affect ground handling?", describing what I observed in b). If you change the trim in a real aircraft it will have nothing to do with the ground handling of the bird assuming you keep the stick back and the three tires firmly on the ground. Yes, it might make the tail come off a different airspeed if you let it, and yes it may cause you to position the elevator differently to keep the tail on the ground and thus change the amount of drag created, but those would be minor effects and unnoticeable.

 

Although I have absolutely no idea how ED models the ground handling, my experience with it suggests that somehow the trim setting affects the tailwheel's ground friction which affects the yaw stability. But that's simply a wild guess based on how it feels and may have nothing to do with how the code really works. I'd love to understand what, exactly, TO Assit does beneath the surface.

 

Regardless of my impression and other's ability to successfully get the aircraft in the air, it's safe to say that if a real P-51 behaved like this one does now it would never have become the beloved fighter that it was and is. Rhetorically, which is better: A 100% faithful model or a model that feels like the real thing? I'm sure the question has been debated amongst sim developers since the days of subLOGIC and Bruce Artwick's crude masterpieces.

 

Budd Davisson wrote back in 1990

Connie Edwards, long time sparkplug of the warbird movement and quinessential Texan is credited with saying, "Start out in a Bearcat, transition to the P-51 and then you're ready for the T-6."

 

Edwards was referring to the T-6's less-then-spotless reputation for ground handling. And he's right. Many civilians transitioning into fighter aircraft are amazed at how much easier fighters are to handle (in most situations, anyway) then the old Texan, a supposedly easily-tamed "trainer." When I got my chance to fly Mustangs, I was amazed and relieved to find this was absolutely the case. If the P-51 had been the quantum jump up from the T-6 in ground handling difficulties that it was in aerial performance, my first Mustang hop would have culminated in a spectacular fire at the edge of the runway. Even if I kept control and survived the flight, I would have drowned in post-flight adrenaline flow. Obviously it didn't happen that way because the Mustang was such a pussy cat compared to the Texan.

 

With a lot of time in the SNJ (AT-6) and a lot of dual given in it, if what Davisson writes is true, then my observations are actually conservative and the model needs a lot more tweaking that I've suggested. In fact, my first takeoff in the DCS '51 did end up as a spectcular fire on the edge of the runway!

 

Again, to be clear, I am not criticizing the beautiful work, the artistry even, that has gone into building this wonderful simulation. I've been a fan of DCS since Warthog beta days, and have nothing but selfish self-interest in mind when I hope that the ground handling can be tweaked because it will make flying this sim more fun. If that makes more fun for everyone, then I'm doubly grateful for the devs fine work.

[ErichVon]

A RL Pilot did. He was happy.

 

I thought I had read that somewhere, here.

 

I would like to see his review and his critique about it.

 

I have no problem taxiing it to the runway, but I have noticed different mission setups do behave differently in SP and MP. Some missions I do not even hold the stick back until takeoff, using just the rudder pedals to steer it. But I am very careful not to move the stick forward. My trim before taxiing is pre-set for takeoff while taxiing. Other missions needs the stick full back and sometimes while full back I need to brake one wheel or the other to turn it aggressively.

 

On landing I try to come in just above the runway end at about 110 mph, not too high, I am aiming for the runway all black tire marks, stick back in my lap at about 90 to 100 mph and just gently drop onto the runway. I don't touch the brakes at all until I am well below 50 mph as compared to the Warthog A-10C.

 

Both taking off and landing are not 100% perfectly consistent for me, yet. Sometimes I do great. Sometimes I do fair. Sometimes I really screw it up. But in real life, be it Navy carrier landings or civilian landings, that too is not 100% consistent every time. Up to Lake Placid, NY airport, a few years ago, I saw one of the mountain pilots chew out an older NJ pilot that had just made a not so great landing in his private plane.

 

I like the night vision goggles; the gain controls do not work in Beta #2 on my end.

 

I like the sense that the P-51's night time cockpit dials automatically come on when set for "Create Fast Mission" for a 2AM ramp start, when I switch the battery on. My 2003 Pontiac Grand Am has this feature, too---when going into a tunnel or a dark rainstorm in the daytime, my dashboard lights automatically come on. There is a photocell underneath my central front dashboard where the defrosting vent is that senses it being dark then initializing the lights to come on.

 

I think the DCS World P-51D is relatively modeled as a full sim product if DCS is modeling WWII. Full sim to me means full sim, everything works. Maybe this will change.

 

But I like it. I am not complaining.

 

Actually, since it is in the Beta stage of development, any comments other than possible bugs is just aesthetics just talking shop, or a complete waste of time until the finished product arrives. Then critique it.

 

I am surprised the testers are allowed to say anything while in the testing phases. Beta 2 is out so maybe that is fair game to comment about it. But the testers are probably doing the Beta 3 currently and have to be tight lipped about it until it is released.

Edited May 28, 2012 by ErichVon

[Ali Fish]

That's a good point I am always trying to explain. If you use TO assistant having full info of transverse acceleration take a careful look to its input.

If you rotate plane into 2-point position its input depends on how fast you do it and at what speed.

 

By the way - where is a NACA report about flying quality of XP-51 containing a record of a TO. Note that there was lighter 3-blade prop and Allison engine. One can see how the pilot worked with pedals.

 

re the visual cues......i discovered this through attempting to land at night. and before i really mastered taking off, and since then my approach to the learning the plane radicaly changed into letting the plane do the work, the very stubborn plane that requires this tame approach to its procedures without forcing the issue. i think this aspect is very important. or atleast it was to me and allowed me to throw away everything id learned in every other sim. and just try feel what the plane wants via its fm.

 

i had all these comments initially about the bird but it just fell into place. whilst nothing you read here is actually science of physics fm and ground haldning. once you feel what the plane wants to do you can feel what it wont do....and no amount of visual dynamics can interpret that.

Edited May 28, 2012 by Ali Fish

[Druid_]

I can't help thinking that if everyone had force feedback rudder pedals then it would be that much easier. When applying the rudder for take-off the loads can be quite considerable & thus give necessary feedback. You quickly learn what loads to expect which helps you to deflect the rudder pedals to the correct position that much quicker & more accurately.

 

Without any rudder pedal forces it's all too easy to over-correct the application of rudder necessary to counter any yaw & can quite easily result in a PIO. I have logged time in tail draggers and a few high powered aero aircraft over the years and for me this lack of tactile feedback is the major hurdle. Other simmers like some of the testers put in much more time learning to adjust without any feedback and hence they are able to cope considerably better than the rest of us.

 

This said I do think the amount of rudder required when the tail lifts is a little excessive at the moment. Thats not to say its incorrect but in my limited tail dragger experience it appears that way. I'm sure with time it'll be put right

 

From RL Pilot, there are still some corrections needed. That is why we are still in beta!

Edited May 29, 2012 by Druid_

[Ali Fish]

i dont use any rudder other than trimmed rudder (3-4 degrees) on take off. and i lock my rudder on the x52pro to ensure no major mistakes. rudder control is minimal if anything on takeoff without wind.

[Bearcat]

I can't help thinking that if everyone had force feedback rudder pedals then it would be that much easier. When applying the rudder for take-off the loads can be quite considerable & thus give necessary feedback. You quickly learn what loads to expect which helps you to deflect the rudder pedals to the correct position that much quicker & more accurately.

 

Without any rudder pedal forces it's all too easy to over-correct the application of rudder necessary to counter any yaw & can quite easily result in a PIO. I have time in a tail draggers and a few high powered aero aircraft and for me this lack of tactile feedback is the major hurdle. Other simmers like some of the testers put in much more time learning to adjust without any feedback and hence they are able to cope considerably better than the rest or us.

 

This said I do think the amount of rudder required when the tail lifts is a little excessive at the moment. Thats not to say its incorrect but in my limited tail dragger experience it appears that way.

 

Yeah I think that would be great.. I'd like to see a FFB Hotas plus pedals that were made so that they would also vibrate according to RPMs.. but a butt kickier on your seat.. and the only thing missing are the actual G forces... :pilotfly:

 

i dont use any rudder other than trimmed rudder (3-4 degrees) on take off. and i lock my rudder on the x52pro to ensure no major mistakes. rudder control is minimal if anything on takeoff without wind.

 

I remember talking to a few of The Airmen from my local chapter and they all said that they had to give a lot of right rudder .. especially on the P-40s for some reason (I would have thought it would have been similar since they both used the 1650 engine.. they were flying Cs though) One guy who passed away last year said that he had to practically stand on the rudder pedals to keep the plane straight ..

Edited May 29, 2012 by Bearcat

[cichlidfan]

i dont use any rudder other than trimmed rudder (3-4 degrees) on take off. and i lock my rudder on the x52pro to ensure no major mistakes. rudder control is minimal if anything on takeoff without wind.

 

Track?

[Gonzo01]

One thing I have notice that Rogue Runner so eloquently pointed out, is that we are creating some of our own Pilot Induced Oscillations (PIO) due to the fact that are reaction times are a little slower than what they would be in the actual aircraft. I do not have any actual flight time in Tail dragger type aircraft. But in other aircraft I have flown, I get fasters cues due to lateral g-loads and my peripheral vision that let me know when I start to drift away from runway centerline.

 

In the P-51, I set my view a little further back so that I can get a sense of what it would be if I had peripheral vision. I also do not taxi onto the centerline. I taxi left or right of the centerline, or the upwind side so that I can still see the centerline on takeoff. This provides me with direct feedback so I can quickly determine when the aircraft starts to drift and make the appropriate corrections. Remember to decrease the size of your rudder inputs as the rudder becomes more effective, which is caused by the increases of airflow over the control surface. Smaller corrections are needed as the aircraft begins to accelerate down the runway: this reduces the need to over correct, which can cause Pilot Induced Oscillations. Learn how to keep the Ball in the center by watching the Ball in the Turn Coordinator and how to react with rudder on takeoff. At the beginning of the takeoff roll add power while holding the brakes to 30 inhg then release the brakes while gradually increasing power to 50 inhg on the takeoff roll. It help's to have some of that right rudder in before you release the brakes, I do use a little nose down and right wing down trim, but this will go away once I get a better feel for the aircraft at different Gross Weights.

 

Hope this helps.

 

Bugs

Edited May 29, 2012 by Gonzo01

[Ali Fish]

Track?

yep coming soon. just did one, very happy with the lack of rudder usage, however im going to re do it as the problems im having with trackir and this extra ordinary sunny weather here in blackpool.... stay tuned. i might even do a commentated video for youtube ...

[cichlidfan]

yep coming soon. just did one, very happy with the lack of rudder usage, however im going to re do it as the problems im having with trackir and this extra ordinary sunny weather here in blackpool.... stay tuned. i might even do a commentated video for youtube ...

 

Control indicator on, please.

[159th_Viper]

Control indicator on, please.

 

You can toggle that on while watching the track.