Sirf Give Better Accuracy?

[+NetJunkie]

We got a Garmin Vista CX for geocaching about two weeks ago. So far I'm "meh" with it. We geocache in pretty dense trees and usually my accuracy is in the +/- 50' range even when averaging a waypoint. Would a unit with the SIRF chipset do better? 50' doesn't sound like much but in that heavy of cover it makes for a lot of places to look.

[+Tharagleb]

The Sirf chip just plain locks on to satellites better than others. My old Etrex Legend would often get NO signal in the woods, I have never had that problem with my 60CSx. It gets more sats in general and thus has better accuracy. It also does the WAAS thing, which also makes for better accuracy.

[+El Diablo]

I just got a 60 CSx and used it for the first time today. It's the most awesome thing I've seen in the GPS world. This thing has a lock 5 seconds after powering on.

 

My wife has a 60CS and while we were hunting today her's kept losing signal and mine was constantly locked in.

 

El Diablo

[+NetJunkie]

I just got a 60 CSx and used it for the first time today. It's the most awesome thing I've seen in the GPS world. This thing has a lock 5 seconds after powering on.

 

My wife has a 60CS and while we were hunting today her's kept losing signal and mine was constantly locked in.

 

El Diablo

 

How about accuracy when they are both locked? I haven't had an issue of losing signal, just fluctuating accuracy (which I guess could be due to losing a sat or two). We were looking for a micro cache yesterday in a thick area with tons of places to hide and the best I could do on averaging a waypoint was ~20' and even then I didn't trust it because it would jump from there to 70' over a few seconds.

[robertlipe]

*IF* they both have an identical view of the sky with an identical signal strength on identical satellite constellation, the accuracy will be very similar. Comparing the Sirf-based Garmins to their predecessors with Garmin's home-brew receiver, you will typically just plain get more sats with a stronger signal in most cases you would describe as a "thick" area.

 

The C units are also notorious for not readily divulging when they don't really have a lock - most owners quickly recognize the "hmmm. I'm standing still; my compass is frozen and my reported speed is unwavering at 2.7 (or whatever) miles an hour" as an indicator that in about a minute, the GPS will finally divulge that it's lost lock.

 

See http://www.mtgc.org/robertlipe/showdown

[+budd-rdc]

I don't own a SiRF-Star III Garmin, but have gone on group caching/hikes in deep forest cover a few times with people who have 60cx's and 60csx's.

 

My observation so far is that the SiRF Garmins tend to zero-in to a cache location faster in the woods than other GPSr's, because of better reception in tree cover. It also helps that most caches in the area are hidden by people who took time to post good coordinates with Magellans and Garmin 60 series with Gilsson external antennas.

 

My eTrex got close, too, but due to inferior sensitivity, it would point near, but still drift around for a while.

[+park2]

I had an interesting observation a couple weeks back. A group of us were hiking up a fairly deep, narrow canyon. My 76CS didn't get a lock. My buddies 76CSx got a lock, but looking at the data after in MapSource, in most places his was obviously out to lunch by 70+ meters.

 

I'm thinking his greater sensitivity was picking up signals after they bounced down off the rock walls (multipath) a couple times. So the greater sensitivity in this situation didn't really get him anything. (But of course in other situations it would)

[+YuccaPatrol]

I just got a 60 CSx and used it for the first time today. It's the most awesome thing I've seen in the GPS world. This thing has a lock 5 seconds after powering on.

 

My wife has a 60CS and while we were hunting today her's kept losing signal and mine was constantly locked in.

 

El Diablo

 

You are supposed to let your wife carry the good GPS!!!!!

[kb9nvh]

This is the drawback to greater sensitivity is the multipath issue.

 

In my book, some lock is better than no lock as long as you know whats happening...

Knowing how to use your tools is the key and when your information may be degraded.

 

 

I had an interesting observation a couple weeks back. A group of us were hiking up a fairly deep, narrow canyon. My 76CS didn't get a lock. My buddies 76CSx got a lock, but looking at the data after in MapSource, in most places his was obviously out to lunch by 70+ meters.

 

I'm thinking his greater sensitivity was picking up signals after they bounced down off the rock walls (multipath) a couple times. So the greater sensitivity in this situation didn't really get him anything. (But of course in other situations it would)

[+hogrod]

Using my 60cx a few days ago I noticed that when I was under a heavy tree cover area near a openside metal roof shelter I was getting some signal bounce(using most current firmware). When I say bounce I mean couldn't seem to get any closer than +/- 40ft. The unit had a much better claimed accuracy than that, but just couldn't decided where ground zero was.

 

one time in a poor signal area (heavy tree cover in a 200+ ft deep narrow canyon) the legendC had 80+ ft claimed accuracy and could only get me within 100ft of the mark. At the sametime the 60cx had 40+ft calimed accuracy and could get me within 20ft.

 

Comparing these two situations I would say that I am still happy with the 60cx, though I wish there was a way to turn on some sort of position average for some difficult spots near buildings ect.

Edited July 31, 2006 by hogrod

[+Searching_ut]

Being the sort of guy who judges GPS units mostly by the tracks they record, and the accuracy/repeatability I see there, I'm pretty much becoming convinced that overall, the new Sirf units are a little more accurate than the older units were. As others have mentioned, they more often than not lock on faster, and hold a lock better as well. Having logged a little over 500 miles now using an assortment of Sirf units at walking or running speeds then looking over the tracks and when possible comparing them with tracks from other units recorded at the same time, I'd have to say the Sirf units recorded a more accurate track most of the time. That said, once in awhile they can get it way wrong in the canyons, and unlike most other Garmin units it can be a little bit difficult to tell that it's way off the true position. (They can be way off and fairly stable unlike the older Garmin units that useually jump all over when they're struggling with a fix)

 

As for upgrading from the Vista CX you currently have, myself I wouldn't find it worthwhile to go from the eTrex form factor to the 60 Form factor for the noticable, but in my opinion not that significant reception/accuracy improvement. My advice would be borrow a Sirf unit and do a little side by side of your own before you make the jump.

[+dumketu]

Not really. The Sirf chipset uses an order of magnitude more "correlators" in its lookup table than the older GPS chips. That is how they get the sensitivity. It is not so much that they "receive" more signal, the lookup tables help it make more use of weaker signals. By the same token, the more robust lookup tables will normally make it more accurate as well under the same conditions as the older chips. This has also been my experience in practice.

 

 

*IF* they both have an identical view of the sky with an identical signal strength on identical satellite constellation, the accuracy will be very similar. Comparing the Sirf-based Garmins to their predecessors with Garmin's home-brew receiver, you will typically just plain get more sats with a stronger signal in most cases you would describe as a "thick" area.

 

The C units are also notorious for not readily divulging when they don't really have a lock - most owners quickly recognize the "hmmm. I'm standing still; my compass is frozen and my reported speed is unwavering at 2.7 (or whatever) miles an hour" as an indicator that in about a minute, the GPS will finally divulge that it's lost lock.

 

See http://www.mtgc.org/robertlipe/showdown

[8mmag]

*IF* they both have an identical view of the sky with an identical signal strength on identical satellite constellation, the accuracy will be very similar. Comparing the Sirf-based Garmins to their predecessors with Garmin's home-brew receiver, you will typically just plain get more sats with a stronger signal in most cases you would describe as a "thick" area.

 

 

*IF* is the operative word. With the increased sensitivity of SIRF comes increased numbers of satellites received. Not only is there more opportunity to position average (overcalculate) a solution, but the weaker signals being received are from SVs closer to the horizon as well thus increasing the positional accuracy of the solution further. So as long as the correlators in the chip are doing their job effectively processing the weaker signals received, it seems reasonable to conclude they do indeed offer better accuracy (in spite of the C(S)x's higher reported EPE's compared to the C(S)'s).

[robertlipe]

Correct. That's why I emphasisized the *IF* in that sentence. Because of the increased sensitvity, these designs will typically get a usable view of a larger number of birds. That allows increased choice in geometry of the SV's used which can lead to increased accuracy.

 

"More use of weaker signals" corresponds to being able to use more SV's.

 

But it doesn't do that in any mysterious way. When this happens, look at the sat screen and you'll "see" more satellites which would make the conditional of my "if" false.