rrTV-PHOTO   New HD TV
HOME   rrTV-PHOTO   GALLERIES   MY GALLERY   HELP-FAQ
myHOME PM pmRR MEMBERS 710 ONLINE 46 EVENTS SEARCH REGISTER  START HERE
 
6 pages [ <<    <     3      4     ( 5 )     6     NEXT    >> ]10882 viewsPOST REPLY
Next D . Fast Lad Performance . Ace Hobby

.
.
Computer Flight Simulators > Calling all software developers
 
 
w.pasman
Elite Veteran
Location: Netherlands

hvrbyte,

Probably you will need to rework the code, it is specificly built for reverse engineering purposes so you will want to strip the useless parts from it. I guess it's fast enough for straight use, it's not efficient but current CPUs are so fast that some efficiency on this point seems irrelevant.
You may also want to recheck all failure handling points. I'm pretty careful with handling failure conditions but I know I have not covered everything, for instance I ignored possible hotplugging of the audio device.

I am responsible for the macintosh version. It contains mainly my own program code but some is (modified) code from the apple developers example code. You may have to check what apple has for licencing requirements on developer demo code and if my modifications are sufficient to consider it my code instead of apple's code.

The program is available for anyone to use as he likes. I would be honored if you refer to me in the sim's info screen. I consider the tech docs as the main result, the program is just a tool for reverse engineering support. I think it would be wise to keep a reference to me in the source code so that future editors can trace back the source.

For the Windows part, that's the work of Frederic (user name FredericG on this forum). I think it would be best if you PM him. Give him a ref to this thread and forward him my greetings
07-05-2005 Over year old.
HOMEPAGE  
 
 
slow
Veteran
Location: San Diego, CA

Hi all,

Just wanted to give you guys an update on whats happening on this side of the pond.

Work from my side has pretty much ground to a halt in the last two months, but that doesnt mean that the project is dead.. Far from it.. Actually, I've been spending alot of time thinking about the sim and planning a bit for what direction we should take it in. Planning is very important for a project this size, and it allows you to reflect on the work already done to see what was good, and what probably will hinder you in the future.

My primary plan is to sit down and try to write my own rigid body dynamics routines, as I feel that ODE provides alot of things we dont need (complexity being one), and doesnt supply some that we do need (aerodynamics, etc). Writing a RDB engine is tricky work, and I've never done it before, so I hope I'll be able to.

As someone mentioned above, yes, graphics arent all that important to begin with, but writing "fancy" graphics used to be my hobby, so getting the sim up to "XTR level" took me less than a day in pure graphics programming. Reversing the .mod fileformat took a day or two extra.
Physics.. Yes, physics is the heart in all simulators, and anyone who has seen me do the physics for heli simulators on a whiteboard knows I'm serious about it all.
Actually, I have hooked up my old PC (yes, I feel very, very dirty) to give XTR a whirl again, and I can tell you, I'm not very impressed with the physics in it, and I'm quite confident that we can archive the same level and more.

One of the things I've also been thinking of is the models physics properties.
Disregarding the actual physics parameters for now, wouldn't it be more clever to present with choices as "Red or Blue Thunder Tiger Dampeners", "Washout mixers set for normal or agressive 3D"? Also to let the user choose between a range of gyros and servos, where we have a database of their individual response-times, maximum piro-rates, etc..

Anyhow, I am starting to get the urge to get back to producing code for the sim, so dont be surprised if I do just that very, very soon.

/James

Hi! I'm a .signature virus!, copy me to your signature to help me spread!
07-16-2005 Over year old.
 
 
dabba
Senior Heliman
Location: South East UK

Superb stuff.
Yes being able to change things over parameters is great.

Will you be attending 3DMasters, as i for one wanna shake your hand.
Good luck
Dave
07-16-2005 Over year old.
 
 
Drunk Monk
rrProfessor
Location: Preston, UK

I like the parameters idea alot. Sounds alot simpler


Stephen

I only open my mouth to change feet.....
07-17-2005 Over year old.
 
 
dabba
Senior Heliman
Location: South East UK

Dont you mean you prefer ditching parameters in favour of items ?
07-18-2005 Over year old.
 
 
w.pasman
Elite Veteran
Location: Netherlands

Quote 
Reversing the .mod fileformat took a day or two extra.


Hey you reverse engineered that? I'm not sure but I thought the people busy with building new scenes for XTR are dying to know about that.
How about a short documentation somewhere (here, on your webpage, whatever) telling the secrets? Could also be handy for future reference for yourself

Quote 
Disregarding the actual physics parameters for now, wouldn't it be more clever to present with choices as "Red or Blue Thunder Tiger Dampeners", "Washout mixers set for normal or agressive 3D"?


This sounds nice for people flying thunder tiger raptor 50. But for all others they will have no clue. Actually fine-tuning the settings is a work of art, everyone's heli is flying different: servo choice, play in parts, mods that change the wind flow, etc.

I would stick with the raw parameters as in Reflex. That's the only way to make sure that all heli flyers benefit from your efforts.At later stage you might try adding an extra interface that tries to interface specifically for a particular machine, say raptor 50, and fills in the lower level params as required. I think you would need quite some smart filling-in because of issued I just mentioned
07-18-2005 Over year old.
HOMEPAGE  
 
 
dabba
Senior Heliman
Location: South East UK

its the .wlp files we wanted to know about.

d
07-18-2005 Over year old.
 
 
w.pasman
Elite Veteran
Location: Netherlands

Quote 
Also to let the user choose between a range of gyros and servos


Now that would be interesting. TruFlite has a very advanced Futaba 501 simulator built into the sim. Maybe you can hook it in?
Also I would like to have a CSM560 and a Fut611 simulator

But seriously, those gyros are computers in their own right. I guess it takes serious efford to get a good simulation of that part alone.

For the servos, It seems to me that servo slop, torque and servo speed are the main ingredients, just as what is available to Reflex. Again you might consider a higherlevel interface later, where you can select servo numbers instead of speed/torque/slop numbers
07-18-2005 Over year old.
HOMEPAGE  
 
 
LewisCobb
Heliman
Location: Fredericton, NB, Canada

Re: Reflex .mod file format - here's a second hand waving for that if you can post it somewhere. There's a few fellows over in the rc-groups simulator forum that are chomping away at the moment trying to make sceneries with domed mirrors and all sorts of hocus-pocus software and I am sure they'd be jumping for joy on this info. I have been trying to follow their progress but my head hurts just trying to absorb the info in their e-mail chatter...

Cheers,
Lewis
07-18-2005 Over year old.
 
 
NickWaanders
Heliman
Location: Vancouver, BC - Canada

"My primary plan is to sit down and try to write my own rigid body dynamics routines, as I feel that ODE provides alot of things we dont need (complexity being one), and doesnt supply some that we do need (aerodynamics, etc). Writing a RDB engine is tricky work, and I've never done it before, so I hope I'll be able to."

Hi,

I've created quite a few rigid body physics engines. It's just hard if you've never done one before, like everything. The biggest piece of advice I can give is to look at the barraff papers, they will basically give you all you need for a 'simple' RBD engine. http://www-2.cs.cmu.edu/afs/cs/user/baraff/www/pbm/

About half a year ago I started coding a rc-heli sim myself, and I tried some stuff with rotors and such. Harder than it seems. There's a lot more to it than simple rigid body dynamics, because of all the gyroscopic stuff happening with the rotor. It's probably possible to model it into a simpler model, but it might not be as 'real'.

The hardest bit of physics engines I found to be collision detection and response. This has a MASSIVE impact on the design of the dynamics code. In the case of a helicopter-sim it might not be such a big issue, since it's mostly 'hit something and you crash'. Except of course for the landing gear, to which you could add a simple spring system.

Anyway, I'd be glad to answer any questions you might have. I'd love to join in coding this beast, but I feel I don't have enough spare time as it is..

- Nick Waanders
07-18-2005 Over year old.
 
 
w.pasman
Elite Veteran
Location: Netherlands

Quote 
The hardest bit of physics engines I found to be collision detection and response. This has a MASSIVE impact on the design of the dynamics code. In the case of a helicopter-sim it might not be such a big issue, since it's mostly 'hit something and you crash'. Except of course for the landing gear, to which you could add a simple spring system.


Collision detection might be harder than you think.
Think of those tricks where the canopy nose hits the ground (heli still standing on the ground)
Tail hitting the ground does not always imply a crash too.
07-19-2005 Over year old.
HOMEPAGE  
 
 
NickWaanders
Heliman
Location: Vancouver, BC - Canada

Very true, didn't think of that.

I still think that it might be 'easier' than a full blown dynamic collision detection scheme though, since the only thing moving in the typical scene is the helicopter itself. So really only one dynamic part against a bunch of static faces. Unless there are moving things in the worl of course... I guess you can go to varying degrees of complex-ness here anyway.

As for the tail hitting, what I was thinking you could do is set up a spring kind of system. When the spring is streched or compressed too far, that part breaks. It's used a lot for destruction states of cars in games because it works well, and it's simpler compared to impulse based calculations.

Still, the collision detection/resolving scheme you go with will define the way the physics engine is architected and implemented.

Just my two cts
Nick
07-19-2005 Over year old.
 
 
vitek
Key Veteran
Location: Corvallis, OR

Quote 
as I feel that ODE provides alot of things we dont need, and doesn't supply some that we do need


Hmmm. Sorry, I feel the overwhelming need to chime in here.

If ODE supplies functionality that you don't need, nobody said you had to use it. The things that it does offer covers a few of the big areas that you need to include for a complete simulation [proximity, collision, friction].

Sure it does not have support for aerodynamics, but writing just the aerodynamics is much easier than writing a most of what ODE provides plus an aerodynamics model.

Of course I'm not saying you _have_ to use ODE, I'm just saying that choosing to write something from scratch that already exists and was written by someone who knows what they are doing is _usually_ a very costly decision.

Travis
07-19-2005 Over year old.
 
 
vitek
Key Veteran
Location: Corvallis, OR

Quote 
I still think that it might be 'easier' than a full blown dynamic collision detection scheme though, since the only thing moving in the typical scene is the helicopter itself.


Yes, this is usually a fairly reasonable assumption. Unless someone is interested in having moving objects in the simulation [balloons, aircraft for remote players, AI aircraft, windmills...].

More often than not someone will want to add a feature that will require something more sophistocated than the simple solution.

Travis
07-19-2005 Over year old.
 
 
NickWaanders
Heliman
Location: Vancouver, BC - Canada

Quote 
Of course I'm not saying you _have_ to use ODE, I'm just saying that choosing to write something from scratch that already exists and was written by someone who knows what they are doing is _usually_ a very costly decision.


Couldn't agree more.
07-19-2005 Over year old.
 
 
Tjheli
Heliman
Location: Sweden

I have tested the simulator and its near reflex graphics... i am imprest by his work with it...

TJ
08-12-2005 Over year old.
 
 
nivlek
Elite Veteran
Location: Norfolk England

Does this mean that we are close to a beta release ?

At the end of the day , it gets dark .
08-12-2005 Over year old.
 
 
vitek
Key Veteran
Location: Corvallis, OR

Quote 
I have tested the simulator and its near reflex graphics
Reflex quality graphics for the panorama aren't really that difficult, especially for a graphics guy like slow. It is really just a very cheap illusion.

Quote 
Does this mean that we are close to a beta release ?

Probably not. Last I checked the physics model wasn't anywhere near completion. Writing a physics engine, or even plugging one in from off the shelf is usually a pretty big undertaking.

Travis
08-12-2005 Over year old.
 
 
Torgeir
Heliman
Location: Norway

What is the latest news. Is there still a project?
10-20-2005 Over year old.
 
 
drmgcm
Senior Heliman
Location: Grand Cayman, Cayman Islands, BWI.

Yep please tell us all whats new and exciting with the project,
I think we would all like to know.

Regards

Dmac

IRCHA: 2528 AMA: 849513 R50V2, OS 50 Hyper, MPII, GY601, TJPro, 9CHP & Cool Power 30%
10-22-2005 Over year old.
 
 
6 pages [ <<    <     3      4     ( 5 )     6     NEXT    >> ]10882 viewsPOST REPLY
Esprit Model . Thunder Power RC . Real Raptors

.
.
Computer Flight Simulators > Calling all software developers
  UPDATE SCREEN   PRINT TOPIC Advertisers 

Subscribe to This Topic

Saturday, August 30 - 2:29 am - Copyright © 2000 - 2008 runryder.com | email | link to rr | runryder needs cookie