tuomas

With crosswinds all is well up to, say, 40 knots or so. Then the module starts to output C000 for track, and of course the GPS responds to that. With moderate crosswinds the stuff is fine. You can see it from the serial port data too - looks like the GPS is responding to the protocol correctly. FS2004 GPS of course displays correctly even with insane (100kts) crosswinds, so it might make sense to detect FS2004 and use that value instead of your own computed one when it is available. But for "normal" crosswind values where one dares to go flying with a GA plane, the thing seems to work fine. I'll try to test on some real scenarios once I get my homecockpit to a flyable state again.. Tuomas

Ok. The last test version seems to hit the spot. The track is within a few degrees which is OK, since there's a slight delay (it's a different thing than the aircraft heading anyway) So summary: The "dont send track" did leave the track direction to the last known value, so the plane moved "sideways" or even backwards if you did an U-turn :) So it needs to be part of the sent packets (or, one thing we didnt try yet: send packages with track "---" (unknown) - it *might* then figure it by itself. Just leaving it out didnt work here. But setting magnetic course seems to give sensible results. I will do some more through testing and I'll let you know. If you ask me, this is "good enough" for the public version, stuff seems to work and I managed to find a few airports in marginal VFR with it :) And yea, now that the track is more correct, the location seems more precise too. A big thanks! This is awesome! Tuomas

The Garmin manual does not seem to explain anything like this, but at least on the FS GPS the Track is equal to the magnetic heading, since it shows the same as the plane's heading gyro and compass. Plus the bearing, track etc actually do have a small "m" next to the degree sign. Both on the FS2004 GPS and on my Garmin. When navigating on the air, one does take the magnetic variation into account. So when I do the track calculations at home before the trip, I first measure the TRUE track heading from point A to point B plus the distance. I then check the magnetic variation, and calculate the *magnetic* course from point A to point B. This is what I use when on the aircraft, since all the instruments show me magnetic headings. Map grid is indeed aligned to true north. I think in this case the GPS should show me magnetic values. When it tells me to set course to north, it makes sense if I can just turn so that the turn coordinator and compass show north. This is also how the FS2004 GPS seems to work. The winds seem to be a problem - the GPS gets very confused somehow if I put a LOT of crosswind. The same situation here: SIM: HDG 340, zero wind, airspeed 117KIAS, 2000ft MSL: * FS GPS: Track 340°m, groundspeed 121kt * Garmin: Track 000°m, groundspeed 121kt SIM: HDG 340, wind 040 at 24kts, airspeed 117KIAS, 2000ft MSL: * FS GPS: Track 329°m, groundspeed 111kt * Garmin: Track 056°m (!!), groundspeed 112kt Also the track seems to change lots over time, and the plane on the GPS screen moves totally sideways. So I dont know what is going on, with light winds the error is so small one doesnt notice it much, but with strong winds it makes stuff go totally strange. I dont know if this is an issue with the track being "off" by the magnetic variation, or if it is a problem with something else. I also wonder.. what if we gave the GPS just coordinates? Would it figure out heading and airspeed by itself? This might be worth testing. Is it hard to make a version with configuration option for true vs magnetic heading, and a toggle to turn each data field sending on and off? I could play with it and see if there is a solution to the errors, and what exactly makes them..? I tested same wind in Finland (magvar 5E) and it is working much better. Somehow I think the whole weirdness might just be because of magnetic vs. true track. The GPS does some calculations by itself too and I guess the 20 degrees offset in that makes those go totally strange - since even in 24kts crosswind the Garmin vs FS offset in 5E magvar is just 5 degrees. Ok. Here's the data. FS: W122*39.64 Garmin: W 122° 39.418' But the location seems to be approximated by the GPS, when I pause FS, the gps "anticipates" the movement anyway and just takes corrections from the serial port data when it is sent, so the movement is not "jump to new place every second" but rather a smooth movement that is corrected every time a new location is received. Maybe the mag vs true confuses this as well..? Tuomas

Yes. In Seattle, with zero wind (GPS only knows about ground track, so wind messes that up since it has no idea where the plane's nose is pointing at. With the Mooney autopilot steered straight north, HDG on-screen display shows 360 and the GPS says 020. So yes, looks like it needs to be Magnetic. The GPS also seems to be offset slightly from the coordinates FS displays, but maybe it just gets confused with the coord <-> heading + groundspeed error because the heading is off by the magnetic variation. Tuomas

Actually I got the attachment fine. I had the same problem when sending the doc pdf to you - I didnt see it being attached. But you probably got it anyway. But nevertheless. THANKS! Guess what? The thing *works* right away! The feeling is magical, it updates very nicely, about once per second (which is the data sending frequency too) - looks just like in real life! Wow. I am totally psyched. Thank you so much for this. I'll do some more testing so I can tell you how it exactly works, if everything is accurate etc - but it definitely made the GPS alive in "slave"mode when flying the sim. Check out this pic: Tuomas

Don't you mean throttle axis? One calibrates throttle "idle" to be a bit "forward" so one can then pull throttles completely "back" past idle, and that becomes reverse. At least I remember doing that once. Tuomas

And check out the Home Cockpit forum here on simflight.com (just browse the "Forum Index" on top, and also check the home cockpit forum on avsim.com. Lots of information on the archives. Tuomas

Or run the Reality XP FSGarmin GNS530 on your small LCD with a hagström keyboard emulator and GNS530 -like knobs and buttons? :) I wouldnt trust the reliability too much though, but it would undoubtedly be a nifty gadget toy :shock: Speaking of GPS though.. GRRR. There *must* be a way to "slave" a Garmin from external data.. so many flight training devices do it (and they use real garmin GNS units..) Gah. I guess I should sneak in to the local flight school and dump some RS232 traffic on the laptop8) I hope the garmin tech support gives some clues about this. Tuomas

AFAIK, FreeFD compiles and runs on Linux too..? I think it uses WideFS though, but that might be worth checking out. Tuomas

There are usb->serial converters at least, that might work. Tuomas

The only inputs mentioned are DGPS (differential GPS) which are corrections from land units, using RTCM protocol, and the proprietary Garmin protocol for downloading and uploading routes, waypoints, map data and so on - not real-time control (I've got the protocol and it is for such static data as normally exchanged between PCs and GPSs, before and after journeys). Hi! Yea, the GPS doesnt want to listen to NMEA (got the gpsmap196 now for real world flying) - even though it says it supports NMEA input. Maybe it just does for some other commands than position data and thinks that it has a satellite receiver for a reason and doesnt care about position NMEA sentences? I wonder what the Garmin protocol means by the "A700 Position Initialization Protocol" and the "A800 PVT Data Protocol"? Quote from the doc: I also mailed Garmin tech support about it - I know it *can* be done somehow since for example ALSim FNPT devices (http://www.alsim.com) can accept normal real-world Garmin GNS530 and 430 units for simulator use. Hmm. Tuomas

I downloaded the manual for the GPS III Pilot from the Garmin website, and browsing through that it really doesn't look hopeful to me. I can't reproduce bits here (it's a PDF file), but for NMEA 0183 format is simply says "supported by the GPS III Pilot and enables the unit to drive up to three NMEA devices". In other words provide positional output. Yea. Check out the GPSMap 196 manual as well - the quote was from that one. Tuomas

OK I will. I guess there is just one way to find out.. Tuomas

The thing can be linked wiht a panel mount GPS (scenario: handheld GPS wiht nice moving map but bad reception because it is inside the airplane, panel mounted GPS with less features but a nice antenna outside that gives good reception) The GPSIII Pilot manual says it can work in slave mode with another GPS. And it also states: "Aviation In - allows GPS III Pilot to communicate with a GARMIN panel-mounted GPS." - this sounds like the proprietary protocol. But also: "NMEA - Supports the input/output of standard NMEA 0183 version 2.0 data." The NMEA sounds like it *might* work. I guess the only way to find out is to try it. I do need a GPS anyway when flying, so if it works in FS and GPSOut.dll it'll be a nice bonus. Tuomas

Think of this as a very good way to practice the use of your own GPS with Flight Simulator. That's the idea. Instead of building a GPS from a small LCD monitor and some knobs into my home cockpit, I might just as well get a small Garmin aviation GPS and use the same unit for both sim flying and for real. I need to check what format it can read indeed. Thanks Tuomas

Has anyone achieved this? The GPSIII-pilot manual says it can act as a slave for a panel-mounted GPS when connected to it, so I wonder if that might work? Have you tried setting the connection mode and also checked the serial connection parameters? I am considering getting a handheld Garmin and I am wondering if anyone has got these to work with GPSOut, would be very nice to practice the GPS unit at home with the sim.. Tuomas