IMHO, they (LR) should try to ...

[uwespeed 558]

52 minutes ago, flying_carpet said:

Either way ... they (LR) don't say, they have the PERFECT simulator, but "the most realistic". By that they mean "the most realistic currently available, but further improvements are possible and definitively will come (like during the last around 30 years)".

I couldn't have said it better 😉.

[CarlosF 1,273]

1 hour ago, flying_carpet said:

they have the PERFECT simulator

Of course not, there's no such thing as a "PERFECT simulator", anybody or any company with such statement would be fooling themselves.

Let's just flip the page and move on, no sense on beating the dead horse.

Edited December 19, 2023 by CarlosF

[uwespeed 558]

On 12/18/2023 at 3:57 PM, BostonJeremy77 said:

If I were Austin, I would do the following:

- Use OSM - the way X-World does, but 1st party

...

If only it were THAT simple ... Would be nice, of course, but ... The guy behind X-World knows what he does, i.e. it won't be possible to have X-Plane 1st party look like reality "only" with OSM.

Example: which one looks nicer, which one is closer to reality?

 

EDIT:
The pictures above are from Prague. I started somewhere at Prague airport and simply searched for a good example to show what I wanted, namely that it isn't as easy as it seems with OSM. Funnily enough ... and after doublechecking with Google Maps: just 1 month ago I drove this (main) road during a business trip - I haven't searched for it by purpose, but coincidentally chose one of the (tens of?) thousands of roads of Prague for these screenshots.

Edited December 19, 2023 by uwespeed

[Bjoern 1,206]

Ah man, the forum software ate my reply. So let's just stick to the productive thing here...

 

5 hours ago, GoranM said:

Absolutely. It can definitely be coded. But I’d be surprised if it wasn’t already in planemaker somewhere. 

Having just looked at the F-14 in PM, the inboard roll spoilers required for DLC are already implemented. However, there is no command and dataref to selectively control them. If Laminar implemented that, the rest would merely boil down to a bit of Xlua scripting.

[blingthinger 687]

2 hours ago, Bjoern said:

However, there is no command and dataref to selectively control them

Or... I wonder... since it has to go to a xlua (or similar) script for PID (or movement schedule of some sort) treatment anyway, I wonder what would happen if they were set to regular spoilers and ailerons at the same stations and set the aileron angle range to be a min of 0. Then have to be careful to subtract off a bit of custom spoiler movement when ailerons are deflected so that there's no double accounting. That latter bit might get a bit ugly though. Or wait another 6/12 months for Ben to give us a roll spoiler dref.

Edited December 19, 2023 by blingthinger

[mjrhealth 933]

Doesnt the MU2 use roll spoilers??

[blingthinger 687]

Dunno. The problem is that the F14 can smoothly move them with fine control (with the roller wheel on the control stick) whereas XP only provides 2 detents for spoiler extension. There is a speedbrake dref that can be used to do the smooth movement from within xlua, but it does nothing when they're set specifically as roll spoilers.

Played with it just now and don't even need to worry about mixing out the spoilers and aileron effects. Just set spoiler chords to 0. They still react to the dref but the aileron model does all the work because there's no surface area for spoiler forces to act upon. Set aileron min angle = 0. Then turn on "ailerons-1 with speedbrakes". Still triggers visual spoiler movement when the speedbrake dref is moved.

Now I think just need to xlua that stabilator to compensate for the pitch moment and mix in joystick inputs...

EDIT: aaand the dref's bugged (I think). Can't seem to move spoilers 1 and 2 independently. 

Edited December 20, 2023 by blingthinger

[mSparks 2,189]

3 hours ago, blingthinger said:

whereas XP only provides 2 detents for spoiler extension.

Its much more detailed than that, but it all depends how far the authors took it.

Plane maker has some reasonably powerful tools that let you connect up all the flight an control surfaces to different inputs, these are mostly aimed at "standard" aircraft, and generally do a fine job pretty quickly.

If that isn't enough you can then override all of that and set the position of every flight surface via plugin, we only recently went "that far" with the 744, effectively "wiring" all the control inputs to cables and hydraulic servos, and then hooking those into the individual flight surfaces they should drive, "by the book".

"the book" starting from diagrams like:

And going all the way down the rabbit hole till we end up with stuff like the angle of all 12 spoilers being independently controlled (6 and 7 are actually part of the speed brake, set elsewhere..)

https://github.com/mSparks43/747-400/blob/f592a2fb1dc80560e8b1fab4a88e42bec188f3e2/plugins/xtlua_keysystems/scripts/B747.19.xt.hydraulicsmodel/B747.19.xt.hydraulicsmodel.lua#L619

I enjoy delving into that detail more than actually flying them tbh, but we are still a way off any simulator being able to reduce the burden of going into that level of detail purely for the consumer market.

 

 

Edited December 20, 2023 by mSparks

[blingthinger 687]

6 hours ago, mSparks said:

all 12 spoilers being independently controlled

It looks like these all control the animations of the .obj. Those drefs are all custom for the 744. None of them directly influence how much drag is being applied to the model, correct? I am able to change a single speedbrake_ratio but that triggers the deploy function for all of them (wing and body). There's also a couple of dref arrays to move the angle of speedbrake 1 and 2 but changing those values does nothing (drag or animation). High level look at the xlua code didn't reveal LR moving those arrays at all: they're just using the default planemaker functionality as far as I could tell.

[mSparks 2,189]

31 minutes ago, blingthinger said:

It looks like these all control the animations of the .obj.

Both,

the flight model effect is set by setting the XP flight model drefs

https://github.com/mSparks43/747-400/blob/f592a2fb1dc80560e8b1fab4a88e42bec188f3e2/plugins/xtlua_keysystems/scripts/B747.19.xt.hydraulicsmodel/B747.19.xt.hydraulicsmodel.lua#L239C2-L245C77

simDR_spoiler12  = find_dataref("sim/flightmodel/controls/wing3l_spo1def")
simDR_spoiler34  = find_dataref("sim/flightmodel/controls/wing3l_spo2def")
simDR_spoiler5  = find_dataref("sim/flightmodel/controls/wing1l_spo1def")
simDR_spoiler67  = find_dataref("sim/flightmodel2/wing/speedbrake1_deg") -- array [0] left, [1] right
simDR_spoiler8  = find_dataref("sim/flightmodel/controls/wing1r_spo1def")
simDR_spoiler910  = find_dataref("sim/flightmodel/controls/wing3r_spo2def")
simDR_spoiler1112  = find_dataref("sim/flightmodel/controls/wing3r_spo1def")

Those "overrides" chain them together (e.g. the spoilers are controlled by both the speed brake and the aileron control inputs, driven to the given schedule)

Edited December 20, 2023 by mSparks

[Bjoern 1,206]

I think @mSparks just posted the key to DLC, @blingthinger. Find the drefs for the inboard spoilers, then create custom commands for DLC toggle, DLC up, DLC down in xlua with the control logic (maybe consider available hydraulic pressure and throttle posiiton as per the real thing for engagement). In theory, this should do the trick without any ACF file adjustments. Two question marks remain regarding roll input (additive to spoiler deployment with DLC?) and ground spoiler functionality (apparently for lift dumping during high speed taxi), but these sound fixable as well.

[blingthinger 687]

3 hours ago, Bjoern said:

I think @mSparks just posted the key to DLC

That was my thought also. Unfortunately I think the roll-spoiler setting is still preventing this from working the "correct" way. I've tried setting a few of those drefs to various angles from within the F14 xlua script and nothing triggers movement or force variation.

Also interesting is that I recall reading earlier (though can't find the reference again now) that IRL when DLC is first turned on, those inner spoilers move to about 7 to 8 deg deployed. The first extension detent in the sim puts those very panels at ~8 deg. Now maybe that's the correct default position when DLC is off, but it is an interesting coincidence. Maybe LR had/have good intentions for DLC. It just can't be done "correctly" yet?

[mSparks 2,189]

4 hours ago, blingthinger said:

That was my thought also. Unfortunately I think the roll-spoiler setting is still preventing this from working the "correct" way. I've tried setting a few of those drefs to various angles from within the F14 xlua script and nothing triggers movement or force variation.

Also interesting is that I recall reading earlier (though can't find the reference again now) that IRL when DLC is first turned on, those inner spoilers move to about 7 to 8 deg deployed. The first extension detent in the sim puts those very panels at ~8 deg. Now maybe that's the correct default position when DLC is off, but it is an interesting coincidence. Maybe LR had/have good intentions for DLC. It just can't be done "correctly" yet?

you have to set override control surfaces to 1 to make them writable

It comes with a health warning ⚠️ that doing so nukes an already very brittle autopilot, and its all or nothing, either XP drives all the flight surfaces or you do.

Plus side, is the 744 autopilot isnt brittle anymore, and the "biggest"/most obvious XP bug we have left open is marked as done, so should hopefully ship in 12.1

https://developer.x-plane.com/x-plane-bug-database/?issue=XPD-14585

Edited December 21, 2023 by mSparks

[blingthinger 687]

10 hours ago, mSparks said:

you have to set override control surfaces to 1 to make them writable

Ah yeah. That did it. And indeed disabled everything. Rewrite the entire control system?? Hmmmmm...

[mSparks 2,189]

33 minutes ago, blingthinger said:

Rewrite the entire control system?? Hmmmmm...

Oh gawd... what've I done 🤣

Well, if you are considering it, and it does seem you are stood on the edge of that precipice rabbit hole.

That first link earlier with all the json is some code I added to xtlua to make it easier, it creates a load of these:

https://github.com/mSparks43/XTLua/blob/0eee72294dc6fc246e2602425f34b05191f4ca0e/XTLua/src/xpdatarefs.h#L43

which run in and around here

https://github.com/mSparks43/XTLua/blob/0eee72294dc6fc246e2602425f34b05191f4ca0e/XTLua/src/xpmtdatarefs.cpp#L249

so you can easily chain lots and lots of

https://github.com/mSparks43/XTLua/blob/0eee72294dc6fc246e2602425f34b05191f4ca0e/XTLua/src/xpmtdatarefs.cpp#L168

return (c->minout + (c->maxout - c->minout) * (x - c->minin) / (c->maxin - c->minin))*scale;

to blend lots of control IO into outputs in the easiest way I could think of at the time while keeping native performance.