This discussion has been locked.

You can no longer post new replies to this discussion. If you have a question you can start a new discussion

GPS Accuracy + satellite count data fields

(Continued discussion from here)

In older models, and in line with most GPS capable devices, there was a data field or a view where the user could see the accuracy level of the device (+/-5m for example) and also a view of the visible satellites and tracked satellites. These views could be used to decide whether a fix is good enough to start an activity, or know how much trust to put in the data displayed (such as lap pace) based on the current accuracy during an activity.

Nowadays, these fields are gone, and I wonder why. They're not even available through an API for developers of 3rd party applications.

The threshold for achieving the "Green circle" is very low, perhaps tracking 3 or 4 satellites, and the accuracy is far from acceptable - try looking at 'current pace' right after obtaining the green circle while standing still. Then there's the so-called "3D Fix" (I'm not sure if this an official term but it has been mentioned on this forum several times). So when *do* we get a 3D Fix? There's no indicator for it. It seems like we have to guess - wait 30 seconds? 60 seconds? 2 Minutes? Does it not seem strange to anyone that we have to play guessing games?

And what about transparency? I don't know of many GPS enabled devices in which GPS accuracy is not something that can be viewed (whether out-of-the-box or by 3rd party applications).

0 jim_m_58 over 6 years ago

Here's the definition of the numbers from the SDK:

QUALITY_NOT_AVAILABLE = 0
GPS is not available

QUALITY_LAST_KNOWN = 1
The Location is based on the last known GPS fix.

QUALITY_POOR = 2
The Location was calculated with a poor GPS fix. Only a 2-D GPS fix is available, likely due to a limited number of tracked satellites.

QUALITY_USABLE = 3
The Location was calculated with a usable GPS fix. A 3-D GPS fix is available, with marginal HDOP (horizontal dilution of precision)

QUALITY_GOOD = 4
The Location was calculated with a good GPS fix. A 3-D GPS fix is available, with good-to-excellent HDOP (horizontal dilution of precision).

So "Green" could even be 2 or 3. (The FW knows more than CIQ can see, and people like to see a quick fix, so I'm thinking "3" in CIQ terms.)

0 WillNorthYork over 6 years ago

For what it's worth I did a super-quick test with my 935 and a running activity. I used the native GPS field (5 bars) and a CIQ field which shows the GPS quality as a number from 0 to 4.

Before I went outside, I had a partial red circle, 0 bars on the native GPS field, and a GPS quality of 0 or 1 (I tried this a couple of times, switching to a non-GPS activity between tests to make sure the GPS turned off). Within seconds I simultaneously got a green circle, 5 solid bars and a quality value of 4.

If it's true that you need to wait 30 seconds to get a "real" 3D fix, then it's not possible that five bars or a quality of 4 is indicating that to you.

So I could be wrong, but I don't think the native field or an app will tell you what you want to know.

I do think that for the native field, the 5 bars probably don't correspond exactly to the internal quality number from 0 to 4. No bars are probably shown for qualities of 0 and 1, while one to five bars probably corresponds to a quality of 2 to 4. (There may be more granular quality information available internally, who knows.)

0 Former Member over 6 years ago

What you have mentioned is a workaround to knowing if you have 3 or 4 satellites. What if I want to know if I have 7 because I know that once I go under tree cover I'll lose a few?

And then what? Let's say for argument sake you have the data field displaying the number of satellites. You run under tree cover and drop a few. Now what? Do you abandon your run? Are you now going to change your course where there is less tree cover? You can't in a race. It's already common knowledge that tree cover can and will affect your GPS signal. Are you just looking for something to give you a visual confirmation on something you already know?

0 jim_m_58 over 6 years ago

But those 5 bars don't map to the levels seen in CIQ. If they did, seems there would only be 3 bars, as 0 and 1 would be no bars. I know in my apps I wait for 4 and that can take longer than "green" in native apps.

0 WillNorthYork over 6 years ago

jim_m_58 thanks, I thought about that too and updated my post to reflect that. Doesn't explain the fact that I got a "4" within seconds, at the same time as the green circle. In my experience the green circle has never been good enough to start running.

But I didn't spend more than a couple of minutes on it and I admittedly tested under different conditions than I usually run.

0 Former Member over 6 years ago

Having some sort of data field showing how many satellites your connected to is important for starting out and knowing your location. I'm confused how the OP would need to know how many satellites they're connected to when running under tree cover, tunnels, etc. It's already common knowledge these things will affect GPS. Let's say I do a run and the first half was under tree cover and the 2nd in the clear. Who's saying "The 1st half of my run had an erratic GPS track, with erratic pace and elevation. I wonder why? I sure wish I had a data field telling me if I had lost any satellites to explain this"? I know the closer I get to a bonfire the hotter it will get. It's common knowledge. I'm not saying "Man, it got hot as I approached that bonfire. I sure wish I had my tempe sensor with me to explain why."

0 talsela over 6 years ago

Having some sort of data field showing how many satellites your connected to is important for starting out and knowing your location. I'm confused how the OP would need to know how many satellites they're connected to when running under tree cover, tunnels, etc. It's already common knowledge these things will affect GPS. Let's say I do a run and the first half was under tree cover and the 2nd in the clear. Who's saying "The 1st half of my run had an erratic GPS track, with erratic pace and elevation. I wonder why? I sure wish I had a data field telling me if I had lost any satellites to explain this"? I know the closer I get to a bonfire the hotter it will get. It's common knowledge. I'm not saying "Man, it got hot as I approached that bonfire. I sure wish I had my tempe sensor with me to explain why."

I find this answer childish. You know reception will be worse under tree cover so why do you need to know how bad it is? Not everything's black or white. I would like to know this because even under tree cover it could be ±9m bad and it could be ±25. So if, for example, my lap pace starts slowly dropping but accuracy is steady at, say ±4m, then I know I'm probably slowing down, but if accuracy drops or is quite bad, then I can account this to the GPS not really knowing where I am.

Here's another use for you: if I had an accuracy field like that, it would be much easier to compare devices. "Hey, I ran with the 645M the same route I run every day, and accuracy went from ±9m to ±17m. With my previous device this was usually ±3 to ±8."

Of course the trivial "I start running when I have 9 tracked satellites and ±4m".

Just three examples off the top of my head.

And how about answering my question: Why not show this information? Why remove this field, that existed in the past?

0 WillNorthYork over 6 years ago

So I tried another quick GPS test on my 935 in an area where I usually run (and where, anecdotally speaking, a green circle is usually not enough to start running).

Seconds after I started my activity I got five bars and a quality of 4, coinciding with the orange circle. The green circle came afterwards.

I know data is not the plural of anecdote but I personally wouldn't trust either the GPS bars or the quality info. I feel like there's a huge range for the "max quality" that Garmin shows you.

As for why Garmin got rid of the detailed info, I'd guess:

They think it's unnecessary and possibly confusing
The accuracy data itself was probably not even that accurate
They don't want you comparing watches that way

0 talsela over 6 years ago

They think it's unnecessary and possibly confusing
The accuracy data itself was probably not even that accurate
They don't want you comparing watches that way

Fair enough - then don't show it by default - allow a data field to be added manually. If not that, at least allow developers to access this information via API.
What makes you say this? There are known methods/calculations to determine this
Exactly what I called 'lack of transparency' and if that were the case, I'd be quite upset as a long time Garmin fan

0 Former Member over 6 years ago

I find this answer childish. You know reception will be worse under tree cover so why do you need to know how bad it is? Not everything's black or white. I would like to know this because even under tree cover it could be ±9m bad and it could be ±25. So if, for example, my lap pace starts slowly dropping but accuracy is steady at, say ±4m, then I know I'm probably slowing down, but if accuracy drops or is quite bad, then I can account this to the GPS not really knowing where I am.

Here's another use for you: if I had an accuracy field like that, it would be much easier to compare devices. "Hey, I ran with the 645M the same route I run every day, and accuracy went from ±9m to ±17m. With my previous device this was usually ±3 to ±8."

Of course the trivial "I start running when I have 9 tracked satellites and ±4m".

Just three examples off the top of my head.

And how about answering my question: Why not show this information? Why remove this field, that existed in the past?

You just answered your own question in your first paragraph. If your pace starts slowing and accuracy is still good, then why do you need to know if you're locked to 7 or 5 satellites? If accuracy drops dramatically, you just stated that you can say your GPS is weak. Again, why do you need to know if you went from 8 to 3. You already stated you know why. As far as to why not show it or why remove it? Do you really want an answer based on speculation since we both know only Garmin can answer the question? Ask Garmin if you want to know for sure.