I may know more about magnetic compasses than some of the engineers putting them in devices...but of course, I don't know what those engineers are doing. But let me try to address the question in general terms anyway.
Consider that traditionally, huge iron ships could use magnetic compasses with fair accuracy. How did they do this? Well, you may have seen pictures of a ship's compass, with two big iron balls on either side of it? The purpose of those balls is to allow the local magnetic field to be "flattened", so the effect of all that iron, much of it magnetized by arc welding, would be balanced out and made more uniform in all directions. They are placed side-by-side, to compensate for the fore-and-aft alignment of the ship, and independently adjustable for compasses not on the ship's centerline. Then once the field was flattened, the compass would be calibrated by a specialist. As best I recall the procedure, the ship would be run back and forth in calm waters on each heading, and the discrepancies recorded, to be applied to headings and bearings taken using the compass. The ship's compass is filled with a fluid, with vanes attached below, to provide damping and to keep it level. All of this could better be done electronically, in software, working from the raw sensor data. But, of course nobody actually calibrates a phone this way. The manufacturer's could, but I seriously doubt they do. But I don't think there's any real reason for them to do so -- because how accurately can you judge the orientation of your phone relative to your heading? And without the little pin in the middle for you to sight against the distant object, how well can you peer over your phone and just the bearing of the distant object? I expect that details like this outweigh any sensor accuracy issues. As for batteries -- those should have no effect. They generally won't have ferrous metals. The currents involved are small -- and so long as they are relatively constant, on a reasonable timescale, at most they'd be something that could be taken care of in the calibration process, were there such a process. But I wouldn't be surprised to see something like a camera flash LED being lit, or the battery being charged, making a noticeable difference. That could also be addressed in software, but could also be addressed by carefully separating the sensor from the LED current. The bottom line is, that while I think phones could be quite accurate, I see no advantage in doing so, because if you need that kind of accuracy, you will be better served by a compass with a physical package designed for the purpose. These days, many of those are electronic using the same technologies as are found in the phone. You can get them built into binoculars, even. On Oct 31, 3:39 pm, DanH <danhi...@ieee.org> wrote: > I do wonder how accurate it could possibly be, however. Your standard > compass can be led astray by a steel belt buckle, so it's hard to see > how a compass inside a phone, with batteries, printed circuits, > electrons whizzing around, etc, could be very accurate at all. > > On Oct 31, 5:35 pm, Adrian Romanelli <adrian.romane...@gmail.com> > wrote: > > > > > > > > > Just to answer my own question, Compass functionality is built into > > some (all?) devices. I noticed that the up-coming T-Mobile Comet has > > compass functionality. > > > On Oct 31, 2:06 pm, Adrian Romanelli <adrian.romane...@gmail.com> > > wrote: > > > > Does an Android 'sensor' have a compass built in? I thought it was > > > just a motion sensor, portrait/landscape, etc., and not a true > > > compass? > > > > And as far as I know, the gps device is a coordinate/point location > > > thing, not a compass thing. > > > > How would the phone/device actually know its facing North? > > > > Finally, I'm not sure if this would help you out, but take a look at > > > this article (its in three parts, link is part one), that describes a > > > scrollable surface > > > view:http://www.droidnova.com/create-a-scrollable-map-with-cells-part-i,65... > > > > Not sure if it can help you or not, but maybe a place to start? > > > > On Oct 31, 1:51 am, "josef.hardi" <josef.ha...@gmail.com> wrote: > > > > > I want to create a non-generic compass that uses rotating directions > > > > instead of a rotating needle as in the conventional compass. The > > > > drawing is like the illustration below. > > > > > | > > > > ' ' W ' ' ' ' ' N ' ' ' ' ' E ' ' ' ' ' S ' ' ' > > > > > (shift a bit to the east) > > > > > | > > > > ' ' ' ' ' N ' ' ' ' ' E ' ' ' ' ' S ' ' ' ' ' W > > > > > The needle (depict as "|") is fixed and the direction string should > > > > have > > > > some sort of "rotation" effect that immediately relocates each sign or > > > > character from one tip of the edge to the opposite edge. Of course, > > > > the movement follows the reading of the Android sensor. > > > > > Does anyone has an idea how to implement this? I stumbled with how > > > > I create the rotation effect and connect the string's movement with > > > > the sensor reading. > > > > > Thanks > > > > /Joe -- You received this message because you are subscribed to the Google Groups "Android Developers" group. To post to this group, send email to android-developers@googlegroups.com To unsubscribe from this group, send email to android-developers+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/android-developers?hl=en
