PhoenixRC radio control model flight simulation.
 

heres a new sim for yall its not released yet but check out the web page looks dam good and for 150.00 cant beat that [link=http://www.phoenix-sim.com/]http://www.phoenix-sim.com/[/link]

RE: PhoenixRC - the professional choice for radio control model flight simulation.
 

well looks like about two more weeks and it will be ready to ship out from whati have seen of it its looks to be better than g3 and fsone so i have already put my money aside for it they raised the price of it 10 bucks but still cheaper than the otherscheck it out . and let me know what yall think of it .....http://www.phoenix-sim.com

RE: PhoenixRC - the professional choice for radio control model flight simulation.
 

Yea I heard about this a loooong time ago. Back before Christmas we discussed it here, and the professional boys here picked it apart, and I believe decided it looked pretty bad, or in beta stages. Have toa dmit it looks good, but it we won't be able to tell how good it simulates until it comes out. I'm not sure why you say it is better than G3 or FSOne, especially because FS isn't even out yet.

EDIT: Price is $170 anyway, not $160.

RE: PhoenixRC - the professional choice for radio control model flight simulation.
 

Looking at the videos the physics look weak. Nothing new almost looks like reflex. I will wait and try fsone myself. The best sim so far in my book is AFPD.

RE: PhoenixRC - the professional choice for radio control model flight simulation.
 

Snicker: no 3D fields, no active terrain, no second order effects, no fluid modeling that is apparent.

Hmmm AFDP (poor "on rails" physics)... I'll pass.

Let's see what FSone has to offer, while I continue with G3.

RE: PhoenixRC - the professional choice for radio control model flight simulation.
 

I think it is going to be a great RC-Sim :D



* Approved and developed by former British Champions, industry professionals and 3D master flyers.
aa
* Stable, powerful and accurate custom-built dynamics engine.
aa
* Advanced graphics capabilities with special effects.
aa
* Over 50 beautifully modelled and realistic aircraft, including autogyros, tilt-rotors and sea-planes.
aa
* 10 unique 360°x180° photographic panoramic sceneries - no lost coverage or fake areas.
aa
* Realistic fully interactive panoramic water environments with splashes, wakes and spray.
aa
* Intuitive, easy-to-use yet in-depth graphical interface.
aa
* Incredibly powerful flight recording and playback tools with flight "shuttling" for instant motion.
aa
* Exciting Night Flying mode with realistic glow-schemes and dedicated aircraft.
aa
* Detailed professional customisation system with revolutionary in-flight tuning system.
aa
* Advanced training modes for helicopters and fixed wing aircraft.
aa
* Unique competition modes make improving your skills fun and exciting.
aa
* Support for up to four people flying simultaneously via split-screen.
aa
* Custom Plug and Play USB interface included.

RE: PhoenixRC - the professional choice for radio control model flight simulation.
 

Only thing in that list that makes it sound good is the Nght flying. G3 has everything else already. 4 player split screen? Good luck with that. Your gunna need a big monitor and lots of chair room for that one.

RE: PhoenixRC - the professional choice for radio control model flight simulation.
 

With no 3D airfields?

You're dreaming.

You'll be flying a nice plane in a static picture.

No thanks.

RE: PhoenixRC - the professional choice for radio control model flight simulation.
 

I've been flying for a fair while, and I'm certainly no expert. I intend to purchase a simulator to hone my skills.

I read opjose's posts with interest, and I would like to know a little more about:

"no second order effects" - could you explain a little further what these are - I've looked at some simulator websites but couldn't find it mentioned.

"no fluid modeling" - Is this the water, or are you referring to modelling airfoils with fluid dynamics?

Thanks very much for your time!

RE: PhoenixRC - the professional choice for radio control model flight simulation.
 

I'm a G3 user and about 2 updates or so ago they made some sort of change that makes my FPS fall generally between 20-30 fps on the 3d fields and only getting worse than that not better. My specs are:

P4 2.4 ghz
1 gig rambus ram
Radeon 9800 pro 128
7200 rpm hard drive.

I should be able to get good performance out of the 3d fields but I dont. before that update I was getting 50-60 FPS on the same fields. The 3D fields are unusable to me now and I'm stuck flying only the 2 PhotoFields that they include with their update. Sucks for me.

Personally I hate the latest update to G3 that went live and I am looking for something new. I can't stand the sim anymore and all they seem to do is make it worse. I'll try anything at this point.

RE: PhoenixRC - the professional choice for radio control model flight simulation.
 

You may have another problem in your system. I am running G3 on a 2.2GHz P4 with 768MB DDR2100 and a GeForce 2 128mb card and I still get 70-80 fps on the photofields. I havent downloaded any of the updates, so maybe thats the diference.

RE: PhoenixRC - the professional choice for radio control model flight simulation.
 

Second order effects refer to the physics or world modeling calculation.

e.g. a program can model how an airplane moves through the air using various techniques.

Wind may be considered an "effect" or rather a first order effect that affects the way the model behaves.

A second order "effect" is not quite as important but affects the behaviour of the simulation never-the-less...

e.g. windsocks moving in the wind, trees moving in the wind, dust thrown up when a model's wheels touch the ground, smoke particles, etc.

Fluid dynamics is a more advanced way of modeling the behaviour of a system in a medium.

Most simulations utilize table modeling.

Table modeling pre-calculates aircraft aspects, etc. and responses.

Since everything is pre-calculated (think of a giant spreadsheet), all the computer has to do is look up the result instead of performing extensive calculations.

e.g. the model is at 35% to the screen, look in row 2340 to get the values needed for display.

Table modeling produces unrealistic and discreet aircraft behaviour.


Fluid dynamics is far more advanced and dynamic.

Air is treated as a semi-viscous fluid, through which an airfoil/plane is moving through.

The movement of the air around the model, and the movement of the model through the fluid is calculated constantly.

Of course this requires more computing power, but today's machines are more than up to the task.

Fluid dynamics produces simulation modeling that closely mimics real world behaviour when all of the parameters are correct.


A good example of this is G3's particle system....

Land a plane on an angled rooftop and the smoke from the engine will billow out of the plane and encounter the roof top.

The smoke then will realistically follow the angle of the rooftop until it hits the edge. At the edge it falls straight down like a waterfall, when it hits the ground it "blooms" out... imagine rain running off a gutterless angled rooftop and you'll get the idea.

The particles behave like a suspension in a fluid.


This has nothing to do with water surface waves, that are usually nothing more than a visual effect handled by the graphics card's processor... even though these are fairly computationally intensive for the video processor.

RE: PhoenixRC - the professional choice for radio control model flight simulation.
 

He is talking about the 3D fields not the photofields.

While I wouldn't be complaining about 30 FPS on a 3D field given everything that is going on (see prior post), the drop in FPS is probably due to more of a reliance on DX9 enhancements... specifically the newer revision shaders being used.

If the video card on the system can't handle the effect, the workload is given to the CPU, resulting in poorer performance.

Switching to photofields (ugh, you can have them!) improves matters greatly as now the program only needs to focus on the aircraft being modeled and very little else.

RE: PhoenixRC - the professional choice for radio control model flight simulation.
 

20-30 fps with drops to 10-15 when over any type of structure is unusable. Everything becomes a slideshow.

Odd that I can play IL2, Pacific Fighters, and the latest Flight Simulator century of flight with every single setting turned up to it's highest, at 1280x1024, with 4x antialiasing. (using more simlar games as a comparison)

RE: PhoenixRC - the professional choice for radio control model flight simulation.
 

Structure!

Ah!!!

Does this particularly apply to the helis?

E.G. it's a hundred times worst with the helis?

RE: PhoenixRC - the professional choice for radio control model flight simulation.
 

Thanks for your reply, opjose.

Are you saying that G3 uses fluid dynamics for its airfoil modelling, and that all the other simulators use this "table lookup"?

I only ask because I couldn't see anything on their website about it at all, which was strange.

Thanks for your help!

RE: PhoenixRC - the professional choice for radio control model flight simulation.
 

FSOne and G3 both proport a degree of fluid dynamics modeling. FSOne's advertizing implies that they do this to a great degree.

FSOne is not out yet, so how good of a product it will be, remains to be seen.

I don't know what Reflex uses, as I didn't examine it to that depth.

AFDP seems to be table based.

RE: PhoenixRC - the professional choice for radio control model flight simulation.
 

I had a look at both G3 and FSone's sites - and they don't mention fluid dynamics anywhere - could you point me to your sources for this information? It would be a great help.

Thanks!

RE: PhoenixRC - the professional choice for radio control model flight simulation.
 

Except for perhaps some simple games, I think all flight simulators
and the people behind them these days are working out the physics as
best they know how (F=ma). I don't think there are any magic carpet
simulators left that function like "you push the sticks this way and
the plane moves that way."

Instead what's happening behind the scenes is that something is trying
to predict all the forces on the airplane at any given moment. Then
when all those get added up, the airplane gets pushed in that
direction. The same thing goes w/ all moments: add up all the torques
on the plane and rotate it a little bit in that direction. Then the
plane is now in a new position and the process starts all over again
.. and again ... and again. This might happen 50 times a sec or 500
times a sec - whatever is setup for that sim.

On the physics side, the tricky part is figuring out all the forces
and moments. The airplane is moving in the air and that sets up
differences in the speed of the fluid around the plane and through the
Bernoulli equation one gets surface pressures from the flow speeds.
All these surface pressures are what's pushing the plane around.
There's pressure on the prop, the wing, the fuselage, the tail, etc.
There's also friction on the surface because of the viscosity of the
air. So that's another aerodynamic force. There are also gyroscopic
forces from the prop, and so on.

If you just look at a basic airfoil, to compute the forces on an
airfoil over a range of angles of attack (say -20 to 20 deg) it will
take about 1 min using a code called XFOIL written by Mark Drela at
MIT. XFOIL is pretty good - i.e. gives a good prediction, but we
cannot wait ~1 min or so to figure out what direction to go when we
have to figure that out hundreds of time in a sec and do it for an
entire airplane (or heli).

So the approach is that a bunch of stuff is pre-computed and it is
indeed put into tables that are then used while running the sim. This
can be pretty straightforward as a lot of people know how to run XFOIL
to get an airfoil prediction. Things get a bit tricky when it comes
to working it all out in a simulator because airplanes are not simple
airfoils that XFOIL can analyze. There's a bunch of bits to consider.
Also, the forces and moments on all these bits needs to be predicted
... hundreds of times per second ... for the flow coming at the
airplane from potentially any direction and speed. An added
complication is that there might be a propeller blasting away and that
makes a lot of the wild "3D" stuff possible, but working out how
things react in such a condition is hard. The only way to consider
all these things is to stuff them in tables, read them when the sim
starts up, and then interpolate on that data and go from there.

It's the fact that this is pretty hard that makes all the simulators
different, and some better than others.

Sorry to emphasize the "hard" bit, but that's what I call it. The
bread and butter aero engineering is nice streamlined flight and not
stall, post-stall, "3D" aero, tail slides, spins, flat spins,
blenders, parts breaking off, etc. It's all this latter stuff that's
the most difficult as the more normal straight line flight is
explained well in a variety of sources.

Now having said all that, what's going on behind the scenes is mostly
a detail to the vast majority of people flying flight simulators. So
any conclusion about what simulator to buy is not necessarily based on
the advertising of what's under the hood but instead what people think
is closest to real. And that latter point seems to be one of the
driving debates on this forum.

I should add that another side to all this is the graphics - does it
look real. That's another debate.

I think these debates are great! Every year things get better because
of it.

RE: PhoenixRC - the professional choice for radio control model flight simulation.
 

Hi Michael,

Thanks for the very interesting post.

regards,

Tony

RE: PhoenixRC - the professional choice for radio control model flight simulation.
 

Google always helps....

http://www.horizonhobby.com/Articles...ArticleID=1571

RE: PhoenixRC - the professional choice for radio control model flight simulation.
 

A lot of the posts on these forums are impassioned responses to people's own choices made at time of purchase.

Once someone has committed resources (money) to something like a sim, there is an automatic perception that the one "they" have purchased is the best one.... after all how can they make the "wrong" choice.

This is human nature.

Add to this the hype that say AFDP throws at the buyer about physics, and people do buy into this stuff.

So there are a lot of people that say that AFDP has "accurate" physics, when it's plane's tend to fly as if on rails.

An empirical analysis is far more important than what these people perceive.

---


The underlying mechanism are highly important, and must not be overlooked. While in and of themselves the techniques used to not guarantee accurate representations, things like fluid dynamics calculations do improve the odds that the response of the simulation will more closely reflect the real world behaviour.

Table based calculations are very handy to speed up programs, but are in essense a shortcut, with discreet and finite results for discreet input parameters.

Microsoft's Flight Sim uses table calculations, and looks great, but X-Plane's table calculations far exceed MS Flight Sim's accuracy... one reason X-Plane has been used as a predictive means for real airplane design. X-plane is able to predict behaviour that may be outside that envisioned by the programmer, while MS FS cannot.


If you want a sim that mirrors reality, you must pay attention to how it goes about it, but keep a degree of skepticism.

RE: PhoenixRC - the professional choice for radio control model flight simulation.
 

And don't go overboard trashing table based processes.

A problem is usually solved with an equation. Tables are produced from equations. They're basically producing the same solution but done beforehand.

My career spanned 35 years of programming. It included projects for airlines, one of which was flight planning. If you fly on a commercial airplane today, chances are good that some code being used to work out the flight plan came from or was influenced by some of those projects (or was stolen from.....). grin..... And most flight planning packages today rely on tables extensively since they're almost always far more efficient in a realtime environment to provide the results of long calculations with just as much accuracy and lots less pathlength (response time).

If you're interested in the most efficient use of your path length in real time, you're going to put as much of your computation into tables as you can. If you're good, all of it will be in tables, that is, unless the actual computations are dead simple and take no path length, or take less code than the lookup would take.

Good programmers put as much of the work offline as possible. Do your big computations "batch" and beforehand.

To judge the quality of a product simply because it's got tables or not is naive.

RE: PhoenixRC - the professional choice for radio control model flight simulation.
 

While I agree that you cannot discount a sim simply because it does use table based calculations, tables do not necessarily mean that the "same" equations are done before hand.

Rather "a set of calculations" is done before hand. These do not necessarily represent, say the resultant set of values from an airfoil in a fluid.

And there-in is the rub.

Tables are usually a shortcut to limit the amount of descreet values returned to the simulation.

They limit the performance or returned results of the sim soley to what is within the tables. But what does the table itself represent computationally?

You cannot encompass everything the user will throw at the simulation this way, nor may you need to, but what goes in, comes back out....

That said, they can sucessfully be utilized to pre-compute many of the calculations that would overload the processor but are less important to the fidelity of the sim.


And after ALL of that...

I agree with MSelig in that I'd rather give up a bit of fidelity to obtain a more realistic flying environment.

This is one reason I favor the 3D flying fields (which are CPU intensive) over photofields.

RE: PhoenixRC - the professional choice for radio control model flight simulation.
 

You betcha. And the absence of tables also does not mean the computations being done are appropriate, written efficiently, have an pertinence, or for that matter, actually are being done..... whatever......

And truth is just as you mention......... The presence of the photofields adds realism and anything that enhances the realism is worth the cost.