tr_s

Sirfstar 3 was market leading a few years ago. Nowadays it is actually inferior in sensitivity compared to most new chipsets from for instance Mediatek, or the Teseo in the new eTrex series. There are however, apparently, new Sirf chipsets (4 and 5). Anyway, what the original poster is looking for is both a sensitive receiver and a large selective antennae. If on a budget, only the first one is okay. The new eTrex series for instance (10/20/30) has a very sensitive receiver, but I have unfortunately seen it to be rather prone to multipath problems. Probably, the rather small quite omnidirectional antenna filament in conjunction with the extremely sensitive receiver hears any reflection from a mountain wall almost as good as the direct LOS signal from the satellites. This, however, might not pose a problem at all to the original poster depending on use; also, switching Glonass on is said to help away multipath issues.

Just wondering - why did you stroke those out? It's still clearly a bug in the receiver softwares; not something that can even with friendly lenient reasoning be explained to be "by design". The implementation of logging is flat out wrong, and Garmin should fix it.

False, even if what should be done to do so is unnecessarily inobvious to the average user. Setup -> Altimeter -> Barometer Mode: Fixed Elevation. Reboot the device and you are done.

True actually, even if significantly better in the latest firmware revisions. What's worse, there is a bug such that in the rare case the magnetic compass direction will couple into the device even if the magnetic sensor is completely disabled in the menus. Sometimes, it can even result in the direction arrow rotating while the compass rose does not (yeha...) Here's a youtube video of one moment when I experienced it: http://www.youtube.com/watch?v=UpivT2WTevQ The magnetic direction is supposed to be ignored completely though if you are moving faster than a few miles an hour. Doubt any planes are that slow.

At least in my case... it seems that this problem has been rectified since a couple of firmware revisions. The heading low pass filter seems okay tuned to not throw my direction jumpy during bushwhacking. At circa 2.8-2.9 something, it was truly terrible. Otherwise I second the poster above - the firmware is not mature yet and processor rather slow - if you don't care about long battery life and/or a small size don't even consider any of the eTrexes over a 62s !!!

And don't forget there's a barometer in the 30 as well.

The basemap only contains very basic detail, such as major highways and waters. Nowadays there is an elevation relief built in so you are actually able to see the contours of major mountains around you as well. For rather acceptable roadmaps for much of the world (free): http://garmin.openstreetmap.nl/

Yup, it should - in theory, give a more truthful position most of the time if properly implemented and a simple centered average of the position from both systems is what's being done. About 30% less error actually. The reality is indeed more complex. Not sure how well the GPS manufacturers have investigated ways of combining data from the two parallel systems. Kalman-filter like combining, or just, for the moment, using data from the system that has the lowest calculated EPE could be two possible methods of optimizing it

And after a factory reset my calibration works right off the bat again, just like the poster above mentioned... Talk about shady firmware...

Broken here too... Tried eight times being extremely careful. No luck.

Owner of a Galaxy pocket - cheap of the cheap - tracks to within meters of an eTrex 30. However, elevation accuracy is quite terrible unless one has a clear view of the sky.

My observation as well. Accuracy seemed better with Glonass on in hard reception conditions, and slightly worse in good reception conditions. However it's only an anecdotal feeling. I have not in any way recorded data to back this up. It can also be mentioned that with Glonass on, the unit consumes ~10% more power. It should in theory give higher general accuracy if we make the hugely simplistic assumption that the calculated position from both systems is normally distributed around the true position, and a simple averaging is what is being done. But then, god knows how well implemented Garmin's glonass reception actually is.

I'd go with the 62s any day if neither of the below is a big drawback for you: -you could live with a bit less battery life (the eTrex is hard to beat here, due to an underpowered processor that renders maps rather slowly and a tiny screen) -you have little use for GLONASS support (IMO it's not that much of a biggie, others may not agree) -that it has frontal buttons rather than the side button/joystick configuration of the eTrex -you don't mind the thing being quite much larger (can be a good thing too depending on use, easier handling and readability) My main points against the eTrex 30 is rather slow map rendering and that there is still some much needed work to do regarding the unit firmware. 62 series is more of a mature product line really.

Problem with speed sports that involve quick changes in speed and acceleration is that a normal GPS only samples once a second. Interpolating between the sample points might not give a very consistent view of what actually happened. I think Garmin has a 5 sample/second Bluetooth/USB GPS module which is much more suitable for these kinds of things.

Medium capacity, low discharge NiMH's (2000-2300 mAh) from a reputable manufacturer like Varta or GP, together with a low-current (overnight) charger, is the best option in my opinion. You can count on at a few hundred good runs from every cell by such a setup, at least. If you are going to do outings in freezing temperatures, forget NiMH though.

Not with iPhone. Airplane mode on iOS turns off satellite GPS as well. To test for satellite GPS ability (if the hardware specs aren't enough for you), then you need to turn off all cell/data functions manually in Settings. GPS should still work (from 3GS and up). As for Android/BB/Win, that I can't help with. =P Thanks for pointing that out. One could imagine that the iPhone world is less prone to those faults and therefore there is less reason to actually confirm it. In the Android world it seems every manufacturer has 30 different models and 4 variants of those models, so perhaps no surprise serious faults sometimes surface.

Again it is worth mentioning that "phone caching" is all good - very arguably more user friendly than using the Garmin devices in many cases. However, a prerequisite is that the GPS implementation in the phone actually properly works. I've seen more than one phone model where it doesn't. Set the phone to airplane mode, reboot, and see if you can still get a proper lock and position with a GPS testing app or standalone GPS application. If you can, that's a good indication.

Sure the new phones give a more than acceptable position. However a puny antenna limits accuracy in substandard reception conditions. Also, on certain Samsung machines (galaxy Gio is an example) there seems to be a bug which leads to that the receiver can never lock unless an AGPS signal is had. What this means in practice is that a 3G tower signal is needed to get a position. (This is a rare problem, just mentioning it in case it might be what's troubling the OP. If the GPS signal indicator shows a steady crosshair instead of "pumping" when trying to cache - this most likely is NOT the poster's problem).

And probably 6-7 times more expensive than what original poster is possibly acquiring. I'd not pay more than $15 for a used Legend even if in nearly pristine condition. Biggest problem with the Legend (I own one), as I see it, is that the GPS module in it is old. This means it could lose a satellite lock under even moderate tree cover, and you can't tosh it in your backpack handsfree to track your walk - it will lose lock.

I'm Not Sure, but I think for some reason, the eTrex series has a limit to the number of Geocaches that can be on the device, regardless of the amount of memory left. Yea i think it is 5000 on the 20/30. Maybe not that bad a design decision, considering that parsing even 1000 takes a few minutes. It's not like the system framework they used seems to be efficient, unfortunately...

Barometer makes the greatest difference for me. Not caring about the compass. I have been using GPS without them for so long that I prefer the heading and map not to jump when rotating the device. 3.x firmware finally filters heading calculations correctly so that the heading won't jump during slow bushwhacking too - this was previously a problem.

Reception with the 20/30 series is great with or without GLONASS, and so is battery life. There are still some firmware issues present in v 3.x. Depending on your usage scenario you may or may not have a good time with the device. The more serious ones seem to have been sorted out.

I say, if you are satisfied and it ain't broken - don't fix it. I was ready to praise the v3.x software for these receivers as I've had few problems... quite recently, however, I've on several occasions had the heading freeze (!) - that's certainly a new one...

No problems with the rubber? It has loosened within a year on all three of the old-style eTrex models I owned. Joystick on the classic Legend got quirky after three or four years as well, but not unusable.

In many countries it constitutes a binding contract, but not if the price is so low that it's obviously incorrect (like $10 for a premium line GPS receiver). Here it was only 40% off, so not very obvious I guess.