FORUMS
Remove All Ads from XDA

GPS Testing Results

2,708 posts
Thanks Meter: 3,811
 
By kennethpenn, Retired Forum Moderator / Retired Recognized Developer on 25th August 2010, 10:59 AM
Post Reply Email Thread
31st August 2010, 03:52 AM |#11  
Senior Member
Thanks Meter: 2
 
More
Quote:
Originally Posted by boomerbubba

Actually, in my experience the opposite tends to be true about SGS phones. Driving tests tend to obscure the accuracy problems.

I agree with this. Driving programs "snap" to streets obscuring lateral positioning inaccuracy.
Quote:
Originally Posted by boomerbubba

EDIT: I finally found and watched the newly linked Youtube video here, and was totally unimpressed by this reviewer. He was cluelessly misinformed about the status of the GPS problems on other platforms, and his andecdotal report of using the GPS on the Epic through Google Navigate app proved nothing at all.

BTW, for a good summary of the GPS problems on Samsung Galaxy S phones generally, see this post at the international I9000 forum. These issues are much more complex than superficial reviews engage.

I absolutely agree. Where are the important tests????:
  • TTFF (time to first fix) from RTN (factory start) with several different devices, eg touch Pro 2, Evo, etc in same location.
  • TTFF from cold, warm and hot starts with competing models in same place. IE how well is assistance and cache working
  • Number of birds and SNR, ie how good is actual autonomous gps hardware "
  • HDOP, VDOP PDOP ""
I am seeing multiple poorly done youtubes that could easily be tower triangulation which under good conditions will get you >30' standing still but blow once you start moving. These guys are not just not reporting SNR, they are not even reporting number of sats and HDOP!

The ops tests (and thank you) are not too bad, but there is no data. For example he says: "Also, the GPS occasionally took more than 10 seconds to “lock on” with a cold start." That is the average for top smartphones for an actual cold start. a cold start is no current Ephemeris or almanac stored. starting up a data connection and pulling assistance data (almanac and ephemeris) of visible birds is a agps "cold start" and typically takes at least ten seconds (a few minutes in standalone gps)./
1st September 2010, 03:42 PM |#12  
Senior Member
Thanks Meter: 126
 
More
New '98 feet' bug on Epic GPS
Now that the Epic is out, we are starting to see the usual confusing mix of anecdotal impressions of GPS performance.

Over on another forum, one interesting fact is emerging: There seems to be a new bug, unique to the Epic's implementation of GPS: The imputed accuracy is being reported by several different testers in different scenarios, always at 98 feet! It is as if this number is hard-coded into the GPS firmware.

See Testing the GPS Satellite usage on your new Epic, the right way. Have "use wireless networks" off and Epic GPS - does yours work or not?
1st September 2010, 04:49 PM |#13  
Retired Forum Moderator
Flag Layton, Ut
Thanks Meter: 299
 
Donate to Me
More
Quote:
Originally Posted by boomerbubba

Now that the Epic is out, we are starting to see the usual confusing mix of anecdotal impressions of GPS performance.

Over on another forum, one interesting fact is emerging: There seems to be a new bug, unique to the Epic's implementation of GPS: The imputed accuracy is being reported by several different testers in different scenarious, always at 98 feet! It is as if this number is hard-coded into the GPS firmware.

See Testing the GPS Satellite usage on your new Epic, the right way. Have "use wireless networks" off and Epic GPS - does yours work or not?

The gps has most of the same issues the others have but the epic doesn't lag when it does work.

Some gps videos I made while working Tuesday. Epic gps worked 1\3.

epic gps fail 1
http://www.youtube.com/watch?v=l7HUQAKg6Lc

epic gps fail 2
http://www.youtube.com/watch?v=UV1Na51Dy5A
1st September 2010, 05:09 PM |#14  
Senior Member
Thanks Meter: 126
 
More
A 'known issue' with Epic GPS?
FYI, a user on another forum is reporting that Sprint tech support acknowledged a "known issue" with the GPS.
1st September 2010, 05:23 PM |#15  
Retired Forum Moderator
Flag Layton, Ut
Thanks Meter: 299
 
Donate to Me
More
Quote:
Originally Posted by boomerbubba

FYI, a user on another forum is reporting that Sprint tech support acknowledged a "known issue" with the GPS.

Thank you for the link. I have had the same issue where it wont lock and takes over 5 mins then when it does lock it takes 3-5 mins for it to get your correct location. It looks like the only thing they improved was agps but gps still needs the new driver coming out in September. I really hope Samsung fixes the gps issues.
1st September 2010, 06:11 PM |#16  
Senior Member
Thanks Meter: 0
 
More
Quote:
Originally Posted by boomerbubba

Now that the Epic is out, we are starting to see the usual confusing mix of anecdotal impressions of GPS performance.

Over on another forum, one interesting fact is emerging: There seems to be a new bug, unique to the Epic's implementation of GPS: The imputed accuracy is being reported by several different testers in different scenarios, always at 98 feet! It is as if this number is hard-coded into the GPS firmware.

See Testing the GPS Satellite usage on your new Epic, the right way. Have "use wireless networks" off and Epic GPS - does yours work or not?

I went into my local sprint store that had 2 epics on display with my wife. We were there to decide if they're worth almost $800(waiting for price drop). So I loaded gps status on both demo epics, accuracy 98ft, they also displayed 1-2 less sats than my g1 which had @20-60ft accuracy sitting on the counter 10" lower and directly between the epics. I know this is a static test and doesn't mean much, when I asked the sales people if I could drive around....just kidding.
1st September 2010, 06:29 PM |#17  
Member
Thanks Meter: 1
 
More
I talked to a sprint district manager at work today and he confirmed the epics have a gps issue and button issue.. He said a fix is coming out for the gps issue, but the buttton issue is only on some first batch of epics and to take mine back..
1st September 2010, 06:31 PM |#18  
Senior Member
Thanks Meter: 126
 
More
Quote:
Originally Posted by kgold708

I went into my local sprint store that had 2 epics on display with my wife. We were there to decide if they're worth almost $800(waiting for price drop). So I loaded gps status on both demo epics, accuracy 98ft, they also displayed 1-2 less sats than my g1 which had @20-60ft accuracy sitting on the counter 10" lower and directly between the epics. I know this is a static test and doesn't mean much, when I asked the sales people if I could drive around....just kidding.

Yet another confrirmation of the new '98 feet' bug!
1st September 2010, 06:33 PM |#19  
Retired Forum Moderator
Flag Layton, Ut
Thanks Meter: 299
 
Donate to Me
More
Quote:
Originally Posted by kgold708

I went into my local sprint store that had 2 epics on display with my wife. We were there to decide if they're worth almost $800(waiting for price drop). So I loaded gps status on both demo epics, accuracy 98ft, they also displayed 1-2 less sats than my g1 which had @20-60ft accuracy sitting on the counter 10" lower and directly between the epics. I know this is a static test and doesn't mean much, when I asked the sales people if I could drive around....just kidding.

I got my captivate for $50 from amazon. Im sure the epic will drop to $50-100 with in 1 month on amazon. Its at $200 now with free activation.

http://www.amazon.com/Samsung-Epic-A...3358779&sr=8-4
1st September 2010, 10:33 PM |#20  
Senior Member
Thanks Meter: 2
 
More
What I am finding is poor Signal to Noise, on average about 10db lower on average for the same birds than Touch Pro2 and Treo Pro are showing while keeping them all in the same spot. For example one satellite is showing me as very healthy 41 on Treo Pro (HTC made Palm), 39 on Touch Pro 2, and same bird is 27 on Epic. It also seems to jump to usage of some of the really bad snr birds.

I am also wondering about what is going on with trigger of almanac and ephemeris population over 3g. I am getting inconsistent results on that.

There is a third problem which is that with 4 >17 SNR birds in view, which should be adequate and give hdop of <2, position still seems to jump and be incorrect by an average of 30' to 60'

Can anyone tell me what your SNR values are? best would be compared to other known good gps smartphones in same location.
1st September 2010, 10:51 PM |#21  
sling's Avatar
Senior Member
Flag Va Beach / Hatteras NC
Thanks Meter: 945
 
Donate to Me
More
Those claiming Epic GPS is great have no clue on how it works.

Satellite navigation receivers operate by line of sight with global positioning satellites. This means that at least three satellites must be in "view" of a receiver in order to calculate longitude and latitude. A fourth satellite must also be within line of sight to calculate altitude. On average, eight satellites are continuously within line of sight of every position on Earth; the more satellites in view, the more accurate the positioning.

How to Test how many satellite locs you have.
Enable telanv wait for satellite loc then go into Telanav menu & select (Share & more) , (Diagnostic Tools) & (GPS Status) You will then see pop up that provides your satellite loc info.
You will find epic is only providing 10% 1 satellite loc which is pathetic.

If Samsung considers that a fix they shouldn't be in the smartphone business..same for those giving it positive reviews
Post Reply Subscribe to Thread

Tags
epic 4g, gps, my tracks, navigation, test

Guest Quick Reply (no urls or BBcode)
Message:
Previous Thread Next Thread
Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes