Re: [Flightgear-devel] JSBSim changes
[... Tony wrote ...]: OK, after discovering that my original implementation was horribly buggy, I have now committed a better version of the normalized control surface position code to JSBSim and FG cvs. There is a new set of properties for these: /surface-positions/flap-pos-pct [...] Hmmm, would anyone be so kind to add these controls to the 'External(NET)' FDM some time ? I am designing some flight control system for a private model aircraft project and I'd love to use FlightGear for visualization. To be honest: I found my way to the FlightGear project when I was looking for some code that I could steal in purpose to drive a remote display. Unfortunately I had to find out that I don't have the skills to do the 'real' coding myself - except from the very little time critical prototyping in Perl So I am now in the stage of examining the External interface and I would love to use FlightGear visualization to see if I'm correct. There is much more motivation in having visual feedback instead of having numbers only ;-) Martin. -- Unix _IS_ user friendly - it's just selective about who its friends are ! -- ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
Andy Ross wrote: Erik Hofman wrote: Jon S. Berndt wrote: Erik Hofman wrote: Christian Mayer wrote: Jon S Berndt wrote: In my mind, mass is in kg. and force is in Newtons. That's correct. Are you sure [...]? See? My bad Oh, dear. Time for the whirlwind tour of unit conventions. snip Well, fear not, there is. This magical unit world is called SI, and it works like this: mass: kilogram This is where it usually goes wrong. In the general language the weight is usually measured in kilos (or kg). It *is* plain wrong, but bad habbits slowly (or never?) fade away. I must say, it's been a while since I used those units, so thats the reason I said it wrong. (Sorry -) ) ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
Alex Perry writes: An earlier suggestion was -n for normalized, which is probably the most accurate (and has the advantage of brevity). Is there any standard unit abbreviation that conflicts with that and is useful for flight simulation? How about tackng -1 on the end ? That's nice, but since Tony and Andy have agreed on -norm, we may as well leave the whole discussion closed. All the best, David -- David Megginson [EMAIL PROTECTED] ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
On Fri, 2002-03-01 at 04:06, David Megginson wrote: Alex Perry writes: An earlier suggestion was -n for normalized, which is probably the most accurate (and has the advantage of brevity). Is there any standard unit abbreviation that conflicts with that and is useful for flight simulation? How about tackng -1 on the end ? That's nice, but since Tony and Andy have agreed on -norm, we may as well leave the whole discussion closed. Hey, Andy and I agreed on *two* things yesterday. A momentous occasion! ;-) All the best, David -- David Megginson [EMAIL PROTECTED] ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel -- Tony Peden [EMAIL PROTECTED] We all know Linux is great ... it does infinite loops in 5 seconds. -- attributed to Linus Torvalds ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
David Megginson wtore: Alex Perry writes: How about tackng -1 on the end ? That's nice, but since Tony and Andy have agreed on -norm, we may as well leave the whole discussion closed. Oooh, actually I really like -1. Shorter and really obvious. Alas that it's too late. I weep for what could have been. Andy -- Andrew J. RossNextBus Information Systems Senior Software Engineer Emeryville, CA [EMAIL PROTECTED] http://www.nextbus.com Men go crazy in conflagrations. They only get better one by one. - Sting (misquoted) ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
- Original Message - From: Andy Ross [EMAIL PROTECTED] To: [EMAIL PROTECTED] Sent: Thursday, February 28, 2002 5:21 PM Subject: Re: [Flightgear-devel] JSBSim changes [snip] This doesn't hold for the blasphemous engineering units. How many pounds of thrust are required to accelerate an aircraft with a mass of 3000 pounds by one foot per second per second? I dunno. Trying this in SI: How many newtons of thrust are required to accelerate an aircraft with a mass of 1500 kg by one m/s^2? The answer, immediate and obvious, is 1500. To be fair, SI isn't the only system that has this property. There is another metric system that goes by cgs (for centimeter/gram/second -- the basic units) with the same property. Those folks talk about force in dynes and energy in ergs. Lets not forget that english units do work well when one stops using pound-mass and uses slugs for mass (1 slug = 32.2 lb-mass). Then no conversion factors are required 1 lb-force = 1 slug * 1 ft/s^2. This makes life much easier and many of us don't mind using slugs (expect for maybe thermo people who just can't seem to get away from lb-mass). Now we know immediately that it takes 300 lbs to accelerate 300 slugs by 1 ft per second per second. Rob ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
re: [Flightgear-devel] JSBSim changes
On Thu, 2002-02-28 at 06:16, David Megginson wrote: Tony Peden writes: OK, after discovering that my original implementation was horribly buggy, I have now committed a better version of the normalized control surface position code to JSBSim and FG cvs. There is a new set of properties for these: /surface-positions/flap-pos-pct /surface-positions/elevator-pos-pct /surface-positions/left-aileron-pos-pct /surface-positions/right-aileron-pos-pct /surface-positions/rudder-pos-pct Since Tony and Andy think there should be some kind of official statement from me, I hereby bless these property names for semi-official use until someone thinks of something better. Go forth into the world and multiply, deus tecum, etc. Andy, I'd be grateful if you could set these properties as well, so that I can switch over the animation files. Should we use a -pct suffix for all normalized values, i.e. /controls/elevator-pct /controls/rudder-pct ? It would probably be more consistent, but some of these are heavily-used properties. I certainly like the idea of having some sort of units indicator on every property name, though I agree risk of breakage is high in this case. All the best, David -- David Megginson [EMAIL PROTECTED] ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel -- Tony Peden [EMAIL PROTECTED] We all know Linux is great ... it does infinite loops in 5 seconds. -- attributed to Linus Torvalds ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
re: [Flightgear-devel] JSBSim changes
Tony Peden writes: I certainly like the idea of having some sort of units indicator on every property name, though I agree risk of breakage is high in this case. I can change everything in FlightGear and the base package, but I'm worried about breaking people's private config files, especially for joysticks. All the best, David -- David Megginson [EMAIL PROTECTED] ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
Well, leaving them as they are certainly won't be a problem for the current developers -- the units for those have been percent for as long as I can remember. And percent are non-dimensional, so no units indicator is probably just as correct as -pct or -nd or whatever. How about having a file that contains lines like this ... /position/altitude-ft /position/altitude-m3.3 It creates a bunch of property nodes for aliasing with a scaling factor, which output a warning to the console whenever they are used (as a hint). Both will work for get and set, but only the real one can be tied()ed. This provides legacy support for people, with a reminder to change. ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
re: [Flightgear-devel] JSBSim changes
David Megginson [EMAIL PROTECTED] said: Tony Peden writes: I certainly like the idea of having some sort of units indicator on every property name, though I agree risk of breakage is high in this case. I can change everything in FlightGear and the base package, but I'm worried about breaking people's private config files, especially for joysticks. It isn't a percent unit either on the controls. Per Cent means per 100. That is getting ridiculous IMHO and I for one would prefer the control names do not get changed. Best, Jim ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
Tony Peden wrote: David Megginson wrote: Tony Peden wrote: I certainly like the idea of having some sort of units indicator on every property name, though I agree risk of breakage is high in this case. I can change everything in FlightGear and the base package, but I'm worried about breaking people's private config files, especially for joysticks. Well, leaving them as they are certainly won't be a problem for the current developers -- the units for those have been percent for as long as I can remember. And percent are non-dimensional, so no units indicator is probably just as correct as -pct or -nd or whatever. I agree in general. I'd suggest the use of -fraction or somesuch instead of -pct if the range is 0:1, as it is for most of these properties currently. Even non-dimensional numbers need units to tell you how to interpret them. Sometimes percent really is more appropriate; consider N1 and N2 turbine speeds. Andy -- Andrew J. RossNextBus Information Systems Senior Software Engineer Emeryville, CA [EMAIL PROTECTED] http://www.nextbus.com Men go crazy in conflagrations. They only get better one by one. - Sting (misquoted) ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
Andy Ross writes: I agree in general. I'd suggest the use of -fraction or somesuch instead of -pct if the range is 0:1, as it is for most of these properties currently. An earlier suggestion was -n for normalized, which is probably the most accurate (and has the advantage of brevity). Is there any standard unit abbreviation that conflicts with that and is useful for flight simulation? All the best, David -- David Megginson [EMAIL PROTECTED] ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
RE: [Flightgear-devel] JSBSim changes
I agree in general. I'd suggest the use of -fraction or somesuch instead of -pct if the range is 0:1, as it is for most of these properties currently. An earlier suggestion was -n for normalized, which is probably the most accurate (and has the advantage of brevity). Is there any standard unit abbreviation that conflicts with that and is useful for flight simulation? -n could be mistaken for newtons, although -nt is also used for this. How about -norm? I don't know of any units in any system which could be confused with that. Paul Paul R. Deppe Veridian Engineering (formerly Calspan) Flight Aerospace Research Group 150 North Airport Drive Buffalo, NY 14225 (716) 631-6898 (716) 631-6990 FAX [EMAIL PROTECTED] ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
David Megginson wrote: Andy Ross writes: I agree in general. I'd suggest the use of -fraction or somesuch instead of -pct if the range is 0:1, as it is for most of these properties currently. An earlier suggestion was -n for normalized, which is probably the most accurate (and has the advantage of brevity). Is there any standard unit abbreviation that conflicts with that and is useful for flight simulation? Forces in newtons, although the unit is capitalized in normal usage. While not a showstopper, if we do go to metric someone is going to get confused between throttle-n and thrust-N. How about -u for unit length? Or the slightly more verbose -norm or -frac perhaps? I'm fine with whatever. Andy -- Andrew J. RossNextBus Information Systems Senior Software Engineer Emeryville, CA [EMAIL PROTECTED] http://www.nextbus.com Men go crazy in conflagrations. They only get better one by one. - Sting (misquoted) ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
David M. wrote: An earlier suggestion was -n for normalized, which is probably the most accurate (and has the advantage of brevity). Is there any standard unit abbreviation that conflicts with that and is useful for flight simulation? I am not aware of a standard, currently. However, we should be careful, because what we do here could influence or even set the standard. We are being watched. :-) Jon ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
On Thu, Feb 28, 2002 at 10:25:36AM -0800, Andy Ross wrote: Tony Peden wrote: David Megginson wrote: Tony Peden wrote: I certainly like the idea of having some sort of units indicator on every property name, though I agree risk of breakage is high in this case. I can change everything in FlightGear and the base package, but I'm worried about breaking people's private config files, especially for joysticks. Well, leaving them as they are certainly won't be a problem for the current developers -- the units for those have been percent for as long as I can remember. And percent are non-dimensional, so no units indicator is probably just as correct as -pct or -nd or whatever. I agree in general. I'd suggest the use of -fraction or somesuch instead of -pct if the range is 0:1, as it is for most of these properties currently. Even non-dimensional numbers need units to tell you how to interpret them. Sometimes percent really is more appropriate; consider N1 and N2 turbine speeds. OK, if you really want to save percent for 0-100 quantities, how about something a little shorter than -fraction, -ratio (-fraction is perfectly good, just long) Andy -- Andrew J. RossNextBus Information Systems Senior Software Engineer Emeryville, CA [EMAIL PROTECTED] http://www.nextbus.com Men go crazy in conflagrations. They only get better one by one. - Sting (misquoted) ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel -- Tony Peden [EMAIL PROTECTED] We all know Linux is great ... it does infinite loops in 5 seconds. -- attributed to Linus Torvalds ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
On Thu, Feb 28, 2002 at 01:31:47PM -0500, David Megginson wrote: Alex Perry writes: How about having a file that contains lines like this ... /position/altitude-ft /position/altitude-m3.3 It creates a bunch of property nodes for aliasing with a scaling factor, which output a warning to the console whenever they are used (as a hint). Both will work for get and set, but only the real one can be tied()ed. This provides legacy support for people, with a reminder to change. I hadn't been suggesting switching everything to metric, but now that you mention it ... Oh, no! ;-) All the best, David -- David Megginson [EMAIL PROTECTED] ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel -- Tony Peden [EMAIL PROTECTED] We all know Linux is great ... it does infinite loops in 5 seconds. -- attributed to Linus Torvalds ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
Tony Peden [EMAIL PROTECTED] said: On Thu, Feb 28, 2002 at 10:25:36AM -0800, Andy Ross wrote: Tony Peden wrote: David Megginson wrote: Tony Peden wrote: I certainly like the idea of having some sort of units indicator on every property name, though I agree risk of breakage is high in this case. I can change everything in FlightGear and the base package, but I'm worried about breaking people's private config files, especially for joysticks. Well, leaving them as they are certainly won't be a problem for the current developers -- the units for those have been percent for as long as I can remember. And percent are non-dimensional, so no units indicator is probably just as correct as -pct or -nd or whatever. I agree in general. I'd suggest the use of -fraction or somesuch instead of -pct if the range is 0:1, as it is for most of these properties currently. Even non-dimensional numbers need units to tell you how to interpret them. Sometimes percent really is more appropriate; consider N1 and N2 turbine speeds. OK, if you really want to save percent for 0-100 quantities, how about something a little shorter than -fraction, -ratio (-fraction is perfectly good, just long) But if those settings don't represent a form of units--just arbitrary values, why do they need a suffix? The suffixes were intended to reflect units and make sense only when they mean something, like knowing if the value is in feet or meters, degrees or radians, knots or fps. Best, Jim ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
Jim Wilson wrote: But if those settings don't represent a form of units--just arbitrary values, why do they need a suffix? The suffixes were intended to reflect units and make sense only when they mean something, like knowing if the value is in feet or meters, degrees or radians, knots or fps. Well, strictly, the suffixes are there to provide context and tell the user how to interpret the value they get. Now, it happens that this corresponds closely to the physical notion of unit. But it's not exact. In this case, interpreting the unitless number requires knowing the range in which it lives. A percent value of 0.5 is off by a two orders of magnitude from the same value interpreted as a fraction in the range [0:1]. Applying suffixes here to disambiguate them serves exactly teh same purpose as applying suffixes to disambiguate pounds from kilograms. Units or not? You make the call, but the design impetus is the same. Also, a pedantic note: degrees and radians are both unitless in a mathematical sense. Angles have an absolute magnitude that is invariant with scale, and thus need no units. Degrees are in fact exactly like percent -- they are a convenience representation arrived at by multiplying a unitless number by a scalar constant. Andy -- Andrew J. RossNextBus Information Systems Senior Software Engineer Emeryville, CA [EMAIL PROTECTED] http://www.nextbus.com Men go crazy in conflagrations. They only get better one by one. - Sting (misquoted) ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
On Thu, Feb 28, 2002 at 07:38:39PM -, Jim Wilson wrote: OK, if you really want to save percent for 0-100 quantities, how about something a little shorter than -fraction, -ratio (-fraction is perfectly good, just long) But if those settings don't represent a form of units--just arbitrary values, why do they need a suffix? The suffixes were intended to reflect units and make sense only when they mean something, like knowing if the value is in feet or meters, degrees or radians, knots or fps. To me, no units specified does not automatically imply non-dimensional. It means I still have to ask what the units are. Best, Jim ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel -- Tony Peden [EMAIL PROTECTED] We all know Linux is great ... it does infinite loops in 5 seconds. -- attributed to Linus Torvalds ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
On Thu, 28 Feb 2002, Tony Peden wrote: I hadn't been suggesting switching everything to metric, but now that you mention it ... Oh, no! I have to admit - when I saw someone mention air pressure and in/Hg the other day I thought it was a bad thing! Even us die hard imperial brits don't use that any more ;-) Now, feet, inches, miles, furlongs, etc are another matter :-) -- Jon StockillPublic Key: C6BD585D [EMAIL PROTECTED] ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
On Thu, Feb 28, 2002 at 08:00:49PM +, Jon Stockill wrote: Now, feet, inches, miles, furlongs, etc are another matter :-) Let's not go too far. Furlongs? Next you'll want us to start using stones. :) Actually, I'd like to see everything converted to metric in FG. I never had a problem with english units until I took thermodynamics. The prof would pose a question in one system and expect the answer in the other. That borders on torture. I wish that all engineering classes used metric only these days. Unfortunately, I'm sure it's not the case. -- James (Jay) Treacy [EMAIL PROTECTED] ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
On Thu, 28 Feb 2002 16:13:34 -0500 James A. Treacy [EMAIL PROTECTED] wrote: The prof would pose a question in one system and expect the answer in the other. That borders on torture. I wish that all engineering classes used metric only these days. Unfortunately, I'm sure it's not the case. Metric isn't perfect either - it's been sort of perverted by ... I don't know who. In my mind, mass is in kg. and force is in Newtons. Unfortunately, metric mass terms are used to describe how much something weighs, i.e. a force. This leads to almost as much confusion as converting between slugs, and pounds-mass, and pounds-force. What's that about!?!? Jon ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
Tony Peden [EMAIL PROTECTED] said: To me, no units specified does not automatically imply non-dimensional. It means I still have to ask what the units are. I haven't had this problem in trying to understand these values, but maybe I've just been programming too long. :-) To me, items under the controls category intuitively indicate the potential value within a range of -1 to 1 (except sometimes its 0 to 1!). Calling them something isn't really going to increase the understanding of their meaning or use. Other items such as /autopilot/config/elevator-adj-factor and autopilot/config/integral-contribution have no semantic meaning to someone who isn't looking at the autopilot code. Perhaps integral-contribution does have a meaning, but you still need to look at the code to see where it is applied...in effect asking what it is. On the other hand it is really useful to be able to look at a number and know if it is fpm, fps, or knots! Anyway, thats all I have to say. Think I've already overextended my 2 cents worth! Best, Jim ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
Jim Wilson writes: But if those settings don't represent a form of units--just arbitrary values, why do they need a suffix? The suffixes were intended to reflect units and make sense only when they mean something, like knowing if the value is in feet or meters, degrees or radians, knots or fps. I'm still undecided myself, but the argument in favour is that the suffix does show a kind of unit (a normalized value). Otherwise, the throttle setting could be in inches, the ailerons in degrees, etc. All the best, David -- David Megginson [EMAIL PROTECTED] ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
Jon Stockill writes: Now, feet, inches, miles, furlongs, etc are another matter :-) Why don't we give this thread a rest, then resume it in a fortnight? I have half a stone of paperwork to get through first. All the best, David -- David Megginson [EMAIL PROTECTED] ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
Jon S Berndt writes: Metric isn't perfect either - it's been sort of perverted by ... I don't know who. Napoleon. All the best, David -- David Megginson [EMAIL PROTECTED] ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
Jon S Berndt wrote: Metric isn't perfect either - it's been sort of perverted by ... I don't know who. In my mind, mass is in kg. and force is in Newtons. That's correct. Unfortunately, metric mass terms are used to describe how much something weighs, i.e. a force. This leads to almost as much confusion as converting between slugs, and pounds-mass, and pounds-force. What's that about!?!? posh Pah, you'll always have troubles as soon as commons are using that stuff. /posh The problem lies in the fact that the only time a human really cares about mass is when he feels the gravity force. This is due to the fact that we live on a planet where we've got everywhere the same gravity acceleration (and the places were that's not true doesn't mater to Joe) So the metrical system is actually correct in caring about the difference between mass and force. That ordianary people don't know that is not a fault of the metric system. *I* think it's much more confusing to have pounds that stand for a mass as well as a force. I can easily imagine cases where even professionals can get it wrong (e.g. EOM on the moon) CU, Christian -- The idea is to die young as late as possible.-- Ashley Montague Whoever that is/was; (c) by Douglas Adams would have been better... ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
Jon S Berndt wrote: On Thu, 28 Feb 2002 23:25:47 +0100 Erik Hofman [EMAIL PROTECTED] wrote: Christian Mayer wrote: Jon S Berndt wrote: Metric isn't perfect either - it's been sort of perverted by ... I don't know who. In my mind, mass is in kg. and force is in Newtons. That's correct. Are you sure, for what I know mass is in Newton/m2 and weight is in kg? See? kilograms are a MASS unit - though it is incorrectly used for weight (IMHO). N/m2 is pressure - not mass. My bad it's not N/m2 but just Newton. I remembered we used newton for mass instead of kg. I guess it's just used for confinience but representing not *the* standard. Erik ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
Erik Hofman wrote: Jon S. Berndt wrote: Erik Hofman wrote: Christian Mayer wrote: Jon S Berndt wrote: In my mind, mass is in kg. and force is in Newtons. That's correct. Are you sure [...]? See? My bad Oh, dear. Time for the whirlwind tour of unit conventions. In traditional* mechanics, there are only three measurable quantities in the universe: mass, length, and time. [* The various force interactions, like electromagnetism, have units too. And in relativity, it turns out that length and time are the same thing. And I'm skipping temperature for simplicity.] You can move something around, covering some distance in some time. So we have a derived unit: velocity = length/time The velocity can be changing with time, too, so we have another derived unit: acceleration = velocity/time = length/time^2 Newton's first law leads us to a wonder conservation property. It turns out that if you take the mass of something multiplied by its velocity, and add up all the mass-times-velocity values in the whole universe, the sum never changes. Call this wonderful property momentum: momentum = mass * velocity = mass*length/time It is possible to move momentum between objects, by delivering some amount per time. This is called a force: force = momentum/time = mass*length/time^2 There's another great conservation law, too. When you apply a force for a given distance (by lifting it, or by compressing a spring) it turns out that you are trafficing in another conserved quantity. All the force-time-distance you put in can be gotten back out later on by reversing the action*, and in fact all the force-times-distance in the whole universe stays constant with a little accounting magic. We call this nifty bit: energy = force * distance = mass*length^2/time^2 [* Although it doesn't always come out in quit the way you want; c.f. the third law of thermodynamics.] What does this all have to do with units, you ask? Well, wouldn't it be nice if we could pick units where all of these relationships worked automatically without the need for conversion factors? That is, wouldn't it be nice if one unit for energy was equal to one unit of length time one unit of force? If we could get a velocity by dividing a distance by a time? Well, fear not, there is. This magical unit world is called SI, and it works like this: mass: kilogram length: meter time: second velocity: m/s acceleration: m/s^2 momentum: kg*m/s force:newton, or kg*m/s^2 energy: joule, or kg*m^2/s^2 And there are other units in there too: pressures of one pascal are the result of one newton of force acting over a square meter, etc... This doesn't hold for the blasphemous engineering units. How many pounds of thrust are required to accelerate an aircraft with a mass of 3000 pounds by one foot per second per second? I dunno. Trying this in SI: How many newtons of thrust are required to accelerate an aircraft with a mass of 1500 kg by one m/s^2? The answer, immediate and obvious, is 1500. To be fair, SI isn't the only system that has this property. There is another metric system that goes by cgs (for centimeter/gram/second -- the basic units) with the same property. Those folks talk about force in dynes and energy in ergs. Andy -- Andrew J. RossNextBus Information Systems Senior Software Engineer Emeryville, CA [EMAIL PROTECTED] http://www.nextbus.com Men go crazy in conflagrations. They only get better one by one. - Sting (misquoted) ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
Jon Stockill writes: Now, feet, inches, miles, furlongs, etc are another matter :-) Why don't we give this thread a rest, then resume it in a fortnight? I have half a stone of paperwork to get through first. 0.5 stone = 7 lb = 3.178 kgabout 635 sheets. Standard copy paper = 80g/m^2 = 5g/sheet Flightgear has 56016 source lines, 66 lines/page = 849 pages for comparison. Will it really take you that long to audit our code tree for us ? ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
Re: [Flightgear-devel] JSBSim changes
I agree in general. I'd suggest the use of -fraction or somesuch instead of -pct if the range is 0:1, as it is for most of these properties currently. An earlier suggestion was -n for normalized, which is probably the most accurate (and has the advantage of brevity). Is there any standard unit abbreviation that conflicts with that and is useful for flight simulation? How about tackng -1 on the end ? ___ Flightgear-devel mailing list [EMAIL PROTECTED] http://mail.flightgear.org/mailman/listinfo/flightgear-devel
