madd@adt.UUCP (jim frost) (02/17/89)
I take exception at the way flight rates landings. I've been playing around with the cessna lately and just a little while ago did a picture-perfect landing on the pavement in front of the hangar. A co-worker of mine did the same thing on the pavement in front of the tower. Both of us got scores of zero even though they were beautiful landings. What ever happened to good landings being those you can walk away from? jim frost madd@bu-it.bu.edu PS haven't managed to land the f15 on the strip that runs parallel to the runway yet, but give me some time....
timelord@eos.UUCP (G. Murdock Helms) (02/17/89)
(jim frost) writes: >I take exception at the way flight rates landings. I've been playing >around with the cessna lately and just a little while ago did a >picture-perfect landing on the pavement in front of the hangar. A >co-worker of mine did the same thing on the pavement in front of the >tower. Both of us got scores of zero even though they were beautiful >landings. Perhaps because you're supposed to land on the *runway*. Flight grades you on how fast you're going when you land, how much vertical descent you've got, how far off the numbers you are, and how much drift and heading error you've got...on the _runway_. Note also that you can, in realtime, take off and land on grass. Flight does not allow you to do so. Theoretically, even if your Cessna ran out of fuel, you should be able to make a gorgeous landing on all that grass no matter where you are, since the only obstructions around is the runway area itself. Alas, flight doesn't work that way... it counts all that wonderful grass as "marshes". >What ever happened to good landings being those you can walk away >from? >haven't managed to land the f15 on the strip that runs parallel to the >runway yet, but give me some time.... Try flying the 747 inverted with the cockpit under the runway sometime. -Murdock
crgabb@sdrc.UUCP (Rob Gabbard) (02/17/89)
Speaking of flight:
Does anyone know what the -z option is for ? Also, have you ever done
a T while in dog ? A couple of LARGE wireframe domes show up.
--
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Rob Gabbard (uunet!sdrc!crgabb) _ /|
Workstation Systems Programmer \'o.O'
Structural Dynamics Research Corporation =(___)=
U
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=madd@adt.UUCP (jim frost) (02/18/89)
>>I take exception at the way flight rates landings. >Perhaps because you're supposed to land on the *runway*. >Flight grades you on how fast you're going when you land, how much >vertical descent you've got, how far off the numbers you are, and how >much drift and heading error you've got...on the _runway_. Yes, I know, but it upsets me a little because it won't refuel you. A minor point, considering I probably crash a billion dollars worth of planes a day so what's a throw-away cessna, but still... >Try flying the 747 inverted with the cockpit under the runway sometime. I've never tried that but my co-worker has always been amused at the ability to do barrel-rolls and loops in a 747. I'm kind of amused at what you can do in a cessna too. I keep hoping that some new version will have the wings fall off as they should. I'm also waiting for the day that I shoot someone down in dog and the plane comes apart and falls to earth in flaming pieces. This particular effect would have the bonus of making it easier to avoid planes you've destroyed (I hate the shoot/dodge-explosion sequence). Yours for a more realistic way to waste time, jim frost madd@bu-it.bu.edu
dunlap@bigboote.SGI.COM (D. Christopher Dunlap) (02/18/89)
In article <2651@eos.UUCP>, timelord@eos.UUCP (G. Murdock Helms) writes: > (jim frost) writes: > >haven't managed to land the f15 on the strip that runs parallel to the > >runway yet, but give me some time.... > > Try flying the 747 inverted with the cockpit under the runway sometime. > > -Murdock Or flying through all four buildings in one pass. Do you folks all still use sidewinders and rockets? Try flying with cannons only, and no "head-ons". chris -- ------------------- D. Christopher Dunlap email: dunlap@sgi.sgi.com Hardware Product Support Silicon Graphics Computer Systems
madd@adt.UUCP (jim frost) (02/18/89)
>Or flying through all four buildings in one pass. Or taking off across the runway and turning inside the tower flying an f15 or f16. It's pretty easy with the f16 but the f15 doesn't perform quite so well. I also like to do 8g turns at below 100ft immediately after take-off, or to lift and immediately go into a roll. Looks impressive in airshows! >Do you folks all still use sidewinders and rockets? Try flying with >cannons only, and no "head-ons". We're not to that stage yet, although it's pretty easy to avoid sidewinders and missiles if you know they're there (ie you're watching your radar). One of my co-workers actually managed to shoot down an f15 with a 747 in real combat -- quite an achievement considering the pilot of the f15 was using missiles. I wonder just how much CPU power goes into flight compared to other applications for the SGI.... :-) BTW, has anyone given though to building real controls? Even a joystick and peddles (no throttles, etc) would be a godsend. It took me several days to get the hang of turning with a mouse and it still doesn't feel right. jim frost madd@bu-it.bu.edu
timelord@eos.UUCP (G. Murdock Helms) (02/18/89)
>G. Murdock Helms writes: >> Try flying the 747 inverted with the cockpit under the runway sometime. (D. Christopher Dunlap) writes: >Or flying through all four buildings in one pass. > >Do you folks all still use sidewinders and rockets? Try flying with >cannons only, and no "head-ons". We usually only use sidewinders and rockets when we get too excited and forget, or for revenge. The rest of the time, cannon only. Steve and I are really looking forward to the big dogfight, Chris... pick a date sometime soon. I hear Archer's just signed up for combat also.... -Murdock (Strangman's buddy)
timelord@eos.UUCP (G. Murdock Helms) (02/18/89)
(jim frost) writes: >Yes, I know, but it upsets me a little because it won't refuel you. We were getting a giggle just yesterday about the way dog refuels your jet on touchdown...even though you're still doing 300 knots! I can just see the ground crew now, running after the jet with the hose in one hand.... >I've never tried that but my co-worker has always been amused at the >ability to do barrel-rolls and loops in a 747. I'm kind of amused at >what you can do in a cessna too. Cessnas make wonderful stealth aircraft when flown against any of the jets. By the time the jet pilot's seen you, it's too late for him to set up on you before he blows past. > I keep hoping that some new version >will have the wings fall off as they should. I'm also waiting for the >day that I shoot someone down in dog and the plane comes apart and >falls to earth in flaming pieces. This particular effect would have >the bonus of making it easier to avoid planes you've destroyed (I hate >the shoot/dodge-explosion sequence). I've heard rumours that this may be in the future. Have you noticed on the GTX that the rudder moves, the flames change length according to thrust, and the landing gear fold up and down? Now if we could only have head-on collisions..... >Yours for a more realistic way to waste time, Hear hear. I'm waiting for the graphic chess game that lets you choose texture and color for the pieces and the board..including crystal...and a changing viewpoint that you could even attach to chesspieces and get a 'pawn's eye' view of the board. Hint, hint. -Murdock
dunlap@bigboote.SGI.COM (D. Christopher Dunlap) (02/18/89)
In article <8902172038.AA26677@adt.uucp>, madd@adt.UUCP (jim frost) writes: > >Or flying through all four buildings in one pass. > > Or taking off across the runway and turning inside the tower flying an > f15 or f16. It's pretty easy with the f16 but the f15 doesn't perform > quite so well. I also like to do 8g turns at below 100ft immediately > after take-off, or to lift and immediately go into a roll. Looks > impressive in airshows! Rob Mace is the king of amazing stunts. I'm not bad myself, but Rob still managed to pull a Top-Gun on me a couple times. You'd be on his tail about to nail him with your cannon, and he'd pull up and hit his spoilers. You'd undershoot him and he'd nose down, reset the spoilers, hit full throttle, and pick you off. This whole manoeuver would take about a second and a half. Rob was also real good at loosing people by pulling out of a steep dive at the last moment while the pursuer buried themselves. I used to have a good airshow where I headed down the length of the runway, lifted off, snap-roll, touchdown, and liftoff again. Took me a few takes though. Again, I think Rob originated this trick too. > >Do you folks all still use sidewinders and rockets? Try flying with > >cannons only, and no "head-ons". > > We're not to that stage yet, although it's pretty easy to avoid > sidewinders and missiles if you know they're there (ie you're watching > your radar). One of my co-workers actually managed to shoot down an > f15 with a 747 in real combat -- quite an achievement considering the > pilot of the f15 was using missiles. > > jim frost > madd@bu-it.bu.edu Mark Libby used to fly a cessna and shoot down fighters of all sorts. has anyone got into playing "arena"? chris -- ------------------- D. Christopher Dunlap email: dunlap@sgi.sgi.com Hardware Product Support Silicon Graphics Computer Systems
newsuser@LTH.Se (LTH network news server) (02/20/89)
In article <8902172038.AA26677@adt.uucp> madd@adt.UUCP (jim frost) writes: >BTW, has anyone given though to building real controls? Even a >joystick and peddles (no throttles, etc) would be a godsend. It took >me several days to get the hang of turning with a mouse and it still >doesn't feel right. We have a ``Dimension-6'' -- a steering ball that senses rotation and translation in 3 dimensions. I hooked it up to flight(1), and the controls instantly felt much more natural. I'm sorry I can't give you a detailed comparison between mouse and ball control. The main problem is that the ball is polled over RS-232, which makes the program much slower, only about 6 updates/second on an IRIS 3130. An interesting feature is that you can turn left either by rotating the ball CCW, or (after toggling a switch) CW. It feels differently, but I can't say one is better than the other. Comments please! Dag M. Bruck -- Department of Automatic Control Internet: dag@control.lth.se Lund Institute of Technology P. O. Box 118 Phone: +46 46-108779 S-221 00 Lund, SWEDEN Fax: +46 46-138118
tarolli@dragon.SGI.COM (Gary Tarolli) (02/21/89)
Its nice to see so much discussion about flight going on. I wrote the original flight/dog programs, but haven't done much in the last few years. Rob Mace has done the GT version and all recent mods. Let me clear up a few things and help set the record straight on some others. Also let me try to explain why flight doesn't always try to be real. First - the "-z" option is probably obsolote now. Flight used to set a governor of 20 frames per second so that very fast machines would not have an unfair advantage. Due to a bug in swapintervals, this was removed I think. And most recent versions now actually adjust the calculations according to how many frames per second the pgm is actually achieving. However, I think there's a bug in most of the released versions that make the G-limit and Wing-stall limits come into play much too early. Future versions should have this bug fixed. Anyway, if you're still wondering what the "-z" option did, it used to turn off the swapinterval governor. The "T" threat cones were written for Williams Air Force Base (in Arizona if I remember right). The heads-up display (-h) was also written for them. I left the "T" in there undocumented rather than take it out, just for the hell of it. Someone added code to fire SAMs (probably just sidewinders) when you get close. Apparently , this version has found its way to Canada. Here's a suggestion for you solitaire players - why not record an airshow with "dog -o" and then fly against the planes in it and see how many kills you can record. Of course, the planes in the recorded airshow wont be firing on you, but if you fly a tricky route in the airshow, it could make it challenging for someone to shoot you down. Another variation is for each person to record a 5 minute airshow and then you can swap airshows - if you kill your opponent's plane more times than he kills yours, you win. Sort of a a batch dogfight (yuck Stalls: the ballet pirhouette (sp?) reported in the B-747 is just a bad stall. Stalls are unpredictable in real-life. Since I had no good data on how to model a real stall, I decided to take the liberty of making it do something fun - if you stall really bad you get a random spin based on just how bad the stall is. The worst stall is probably when you go straight up then fall down backwards. Once I was flying at about 100 feet and pulled a real sharp turn and somehow stalled it real bad - I got sent spinning about 1000 RPM and needless to say crashed. The other day I was following someone ready to shoot him down when all of a sudden he stalled and went spinning wildly - it was fun to watch! - but since he lost so much airspeed i flew right by and didn't get to kill him. Refueling: sure its tough on the ground crew to refuel you when you're going 300 knots, but would you rather have to park your plane on the runway and really wait for the crew to service you? I think the term sitting duck is appropriate.... Landing points: what can I say? - do you think the FAA would like it if you landed on the taxiway at Logan or O'Hare? The points are just some arbitrary grading system I made up to make it challenging to land. Also, if you're dogfighting it gives you a way to recharge if you are good enuf to land with people trying to kill you. One more tip - if you type the '~' key - it turns off all fuel consumption but takes away all you rockets and sidewinders. However, you can fly forever with your canon. Since fuel accounts for a lot of a plane's weight, its best to first burn off all but 1 pint of fuel (1 quart if you are conservative or paranoid about running out) before using it. Of course, you can turn fuel consumption back on, but don't expect to get your weapons back! I think that about covers most of the topics I've seen on the airwaves. Remember - half the things in flight are meant to be real - the other half are there to make dog more fun than it would be if things were too real. Some things - like exploding planes and body parts falling to the ground are left as homework exercises ...
rhbartels@watcgl.waterloo.edu (Richard Bartels) (02/21/89)
In article <27132@sgi.SGI.COM> tarolli@dragon.SGI.COM (Gary Tarolli) writes: > > Stalls are unpredictable in real-life. > Nope. Neither are spins. The aerodynamics of both are laid out in any good pilot's instruction book. Quite logical and quite standard physics. As a university teacher I could never fathom why one of my students with a pilot's license thought the flight simulator was so uninteresting. After I got my own license, I see why. Among other things, in real life you practise stalls and spins until you get sick of them (sometimes literally to the point of severe nausea) and learn through them, and sideslips, and take-offs, and landings, and coordinated turns, and spiral dives, and etc. all of the quirks and physical personality of a plane. The behavior of the planes on the iris is completely phony. Part of this is evident in the sheer caprice of their stall and spin behavior (or lack of it). -Richard
madd@adt.UUCP (jim frost) (02/22/89)
>The behavior of the planes >on the iris is completely phony. Part of this is evident in the >sheer caprice of their stall and spin behavior (or lack of it). True. I first noticed this when playing with the cessna. Most obvious was the lack of the stall that occurs immediately after takeoff unless you do things just right. This makes takeoffs much more interesting, let me tell you. It's virtually impossible to *force* this under flight unless you try a low-power takeoff. Turns are also unrealistic since rudder on the cessna has a LOT of effect, while on flight it has little effect on the cessna and almost none on any of the other planes. Stalls also bother me since I loose complete coordination when it starts spinning unpredictably (something I'd never seen before) although I have to agree with the author that this is rather interesting. I would really like to be able to do some of the common airshow stunts that involve stalls though. Also, has anyone been able to put the cessna into a flat spin? I've tried a lot and I just can't do it. As a disclaimer, I'm not a pilot but I've spent quite some time with a cessna flight simulator in addition to using a couple of the microcomputer simulators (eg Microsoft). My judge of behavior is probably somewhat flawed, especially since the lack of real movement makes many things easier and many harder. And you just don't *try* some things in a real plane.... jim frost madd@bu-it.bu.edu
blbates@AERO4.LARC.NASA.GOV (Bates TAD/HRNAB ms294 x2601) (02/22/89)
I have seen "real controls" for PC's. If it is a serial device you might be able to change dog to use it. The hardware is a complete yoke with flap and throttle controls. It was made to be used with MS-Flight Simulator. -- Brent L. Bates NASA-Langley Research Center M.S. 294 Hampton, Virginia 23665-5225 (804) 864-2854 E-mail: blbates@aero4.larc.nasa.gov or blbates@aero2.larc.nasa.gov
tarolli@dragon.SGI.COM (Gary Tarolli) (02/22/89)
True, most minor stalls are quite predictable and can be modelled in simulation and practised in real life. But violent stalls, like a plane falling backwards, or in a flat-spin, I don't believe would be predictable. To model minor stalls you need to know the wing design - which part of the wing will stall first and what the effect on the plane will be. Since flight uses the same equationsmodel all its planes, this is beyond the scope of its equations. To properly model stalls and other effects would probably require much more than the 50 floating point operations in flight, and would probably slow things down quite a bit if you got carried away. The equations in flight are lousy. I'm not an aero engineer, I'm lucky I ended up with planes that could fly period. How about one of you aero-specialists coming up with a better flight model and posting it?
drforsey@watcgl.waterloo.edu (Dave Forsey) (02/22/89)
In article <27060@sgi.SGI.COM> dunlap@bigboote.SGI.COM (D. Christopher Dunlap) writes: >In article <8902172038.AA26677@adt.uucp>, madd@adt.UUCP (jim frost) writes: >> >Or flying through all four buildings in one pass. >> >> Or taking off across the runway and turning inside the tower flying an >> f15 or f16. It's pretty easy with the f16 but the f15 doesn't perform >> quite so well. I also like to do 8g turns at below 100ft immediately >> after take-off, or to lift and immediately go into a roll. Looks >> impressive in airshows! > >I used to have a good airshow where I headed down the length of the >runway, lifted off, snap-roll, touchdown, and liftoff again. Took me a >few takes though. Again, I think Rob originated this trick too. A hard stunt is to start off across the runway straight towards the tower, snap roll burying the cockpit below ground (without crashing), touching down and lifting off *before* you reach the tower itself. An easier but fun flight is to take off from the runway, pull immediately into a vertical climb, cut power, continue the loop and land as close as possible to the same end you took off from. Repeat at the other end without stopping. And while we're at it, why not have a version of dog that allows "manned" anti-aicraft emplacements on the tower, buildings or ground? The mouse controls elevation and orientation with a steady stream of 20mm fire while a mouse button is held down. Heck, why not make them movable AA batteries.... Dave Forsey
mg@cidam.me.rmit.oz.AU ("Mike A. Gigante") (02/23/89)
The thing that dissappoints me most is the lack of coupling b/n roll and yaw. As an ex aerospace engineer this really knocked me out, as an ex-pilot, this is what makes flight so unrealistic to fly. I suppose that the equations were uncoupled to make it moderately fast on the old hardware. Now that the machines are so much faster, it is probably worth re-writing the code to use the 'correct' equations of motion. Mike Gigante, RMIT
tarolli@dragon.SGI.COM (Gary Tarolli) (02/25/89)
I didn't leave out the equatins for coupling yaw and roll, I never could figure them out. THe book I read didn't have them and I couldn't figure out how to derive them myself. Also from what I could figure out, it only is true for dihedral wings - on fighters that have nearly flat wings (instead of a V shape) roll and yaw might not be coupled nearly as much as on a B-747. I would appreciate it if anyone could send me some data on this effect, as aeronautics isn't my field. P.S. Singer Link sells a micro-flight simulator that uses a 68000 to do the calculations and then talks thru a serial port to an IRIS. I trust their equations would satisfy most of you real pilots, but I couldn't get their code and its written is assembly anyway. So real pilots - be quite and go buy a SInger Link simulator (they take American Express...)
spike@adt.UUCP (Joe Ilacqua) (02/28/89)
<And while we're at it, why not have a version of dog that allows "manned" <anti-aicraft emplacements on the tower, buildings or ground? <The mouse controls elevation and orientation with a steady stream of 20mm <fire while a mouse button is held down. Heck, why not make them movable AA <batteries.... < < Dave Forsey G.E.V.s, armed with ground to air missiles and machine guns, is the way to go. And good 'tones' on the 4D/20 sound hardware would be nice. But, of course, someone has to program this and I get the impression there are too many divergent versions already. Joe Ilacqua Associative Design Technology