ORIGINAL: darock
In another thread I made the argument that if the videos look right,
then the physics are right. I'll stand by that claim. I guess the
counter claim would be if the videos look wrong ... then the physics
are right??? That does not make sense to me.
Michael,
First off, thanks for the work that's helped us so much.
And then let me ask a somewhat pushy question. In the quote I pulled out, "if the videos look right, then the physics are right" seems perfectly right and true. But for guys who're asking about the realism of the sim, doesn't the answer have to have a little broader reach than the physics? I understand I'm broadening the context of a statement I pulled from context, but how would you cover the broader question? Physics certainly is one source for the realism, but most of us judge the realism with how the airplanes look (as in the video) but also connect the look in our "videos" (on our monitors) to how we've had to push the sticks to get that look.
When y'all did those videos, did the flyers performing for the videos find that they were pushing their TX sticks pretty much like they would at the flying field while flying those models in flights that looked like the videos they got back home? how fast and how much they pushed the sticks?
I worked for major airlines since back when they/we were starting to develop computer applications for systems such as flight planning and weights and balances and the problems of developing RC simulation have intrigued me since I first saw them. I'd hazard a guess that the problems you guys face are probably pretty much the same "real" ones we faced. Although I'd bet y'all didn't have to deal with a Flight Engineers' union or ALPA.
BTW, I'm looking forward to benefitting from your work, yet again.
whew....... long winded again......
[1] To answer your question about stick input, this is an edit from
another post I made previously:
I think the videos can be used to judge the physics. The only input to
the airplane (motion) is through the sticks: elevator, ailerons,
rudder, throttle, etc. So if the physics were wrong but a sim video
looks right, then that motion could only be achieved by bogus control
inputs in the sim. Flying a sim with wrong physics and bogus control
inputs (to make it look real) implies that a pilot would have some
serious "negative training", i.e. puts in some very wrong stick inputs
to get out some correct looking motion in the video. That is harder
than getting the physics right in the first place. It also implies
that a pilot would be trained to mimic the motions of real airplanes
w/o the help of the correct physics - again this is super
hard. Airplanes in real fly the way they do largely because of the
physics (e.g. snap rolls, spins) - Newton's Laws of Motion at work. So
I think if an airplane looks right in a sim, it looks that way not
because the pilot is jamming in all sorts of weird inputs to make it
look real while constantly compensating for some bad computer
simulation (i.e. wrong equations/bad aerophysics) but it looks real
instead ... because Newton's Laws of Motion (involving the airframe,
prop, fluid and all) are properly modeled and heavily leveraged!
That's the wordy version. The short version ... if it looks, walks
and talks like a duck, it's probably a duck.
[2] The other questions relate generally to what do pilots
think. We've gotten a lot of feedback from many Horizon people, our
developers, beta testers outside Horizon and vendors to-be (who gotten
samples). From what I'm hearing, things feel right and that's my
experience too.
[3] I think developing an RC sim is very much like a full-scale sim
except we don't have to deal w/ the cockpit, navigation,
communication, certification and other things unique to full scale.
We did not have to deal w/ ALPA (thank goodness), but we did work very
closely w/ the people at Horizon and when things did not fly right
... we heard about it! My experience is to trust pilot input
(especially the ones who have flown for many years, e.g. 10+), but the
challenge sometimes is trying to convert that pilot input into
something I understand and can use from the aero/flight-dynamics side.
Michael