But this will also depend on the GPS Receiver you are using.

For example, when you are driving forward wit a left side error of 0.5 meters.
Then drive parallel backwards you will get (worst case) 1 Meter difference.

If you then look at the GPS inaccuracy this error will also be added.
That means you can get a little better or worser Result, depending if the error in the GPS was nearer or farer than the real Point :-)

And on the other side of view, if you have a receiver with accuracy less than 1 meter
you will throw this accuracy away while not correcting.

So for all normal GPS Receivers, which are in accuracy between 1 and 3 meters,
specially if your GPS says it is only 2 or 3 meters accurate,
dont mind about correction or not.

But if you have a better Receiver which says less than 1 Meter accuracy, ...don't throw this away by not correcting.

On my old Humminbird i had an External Receiver which gives me often an accuracy about 1 Meter.
Its minimum reachable error was described under 1 Meter.
But i saw only a few times and not for long an error less than 1 Meter.

Now for my Lowrance Unit i have bought some new GPS Chips from uBlox.
They promised error rates less than 1 meters, which are reproducible been tested by some GPS Cracks.
Sometimes they reached about 0.3 Meters accuracy while testing.

Hi

you are making the assumption that the GPS position error is a constant. It is not , and is variable by the second. You can expect to be somewhere within a 2.5 m radius circle centred on your actual position 95% of the time if you have WAAS or EGNOS and everything else is perfect. It normally isn't.

Trevor

Luke,

you cannot obtain the repeatable accuracy you are talking about with consumer GPS units. The physical limit is determined by the frequency, and is about 2m, EGNOS is stated to provide only "better than 3m". Only by using RTK can you hope to get better repeatable accuracy.

Trevor

The Precise Point Positioning (PPP) technology from ublox reduces the errors more than only the 3 Meters.

I dont reflect only of the technical issues,
i look at the tests i have seen.
And they say it is much more accurate than the "normal Receivers"
Also specially the Position Hopping while standing still was much less.

Even the old Receiver from Humminbird with an previous Chip Generation (sold in 2009) was much better than the actual Receiver located in the HDS 7 Touch. (Tested at the same Installation Position.)
Give a faster First time to Fix, less hopping, faster less position Errors,
faster Egnos/Waas Detection and processing and so on.
So i hope that the new Chip Generation is slightly better.

Myself has not tested the New Chips, because my Project with it will start in Jan or feb. next Year.
First i have to build up the New Boat.
But i have seen Test with them, referenced with an RTK system.
This has been done while testing those Chips against an RTK system,
to get an alternative to an RTK setup.
To give the Farmer´s an less price System with almost the same precision.
.....ok, the new Chips never reached the precision in cm like RTK/dual system/Ref Stations,
but they are good enough to let the Tractor drive alone an accurate Line, like having an RTK System.
So this accuracy might be much better than a normal Receiver.
And for me, i dont care if it gets an error rate under 1 Meter, but i appreciate most of time getting an error near one Metzer instead of 2-3.

Without Structurescan/Sidescan, its hard to find an Object when the Water is only a few meters deep.
So here, every half meter i get a better Position makes me happy.

But thinking about GPS errors is also a religion thing. :-)

Luke,

I wasn't aware that the slow convergence problems (15 - 20 minute startup?), the lack of real-time algorithms and data stream problems had been sorted to the extent that real-time mobile use was now possible with PPP. ? I understood that this still relied heavily on post-processing. Perhaps this has all been resolved in the last year or so ? Does this rely on a third party (subscription) correction supplier ?

Trevor

I cant tell you exactly what they did, because this runs out of my knowledge.
I try to describe what i know (Its no pro Stuff !)
I understood that the Chip used in the Humminbird in 2009 might act and get its best results with Post Processing and then Output the data.
This was a test project from HB to clarify what accuracy was possible.
But HB didn't tell anything about it and actually this receiver is only legacy.
It was to expensive so i think not much People bought it.
After opening the receiver and taking a look at the chip i found uBlox as producer of the Chip.
And they told me this Chip was only designed for HB and was not found on other manufacturers than HB.
And uBlox will not tell more about it due to secret/contract related.
But after my investigations i found last year the infos about the the new PPP Stuff from uBlox.
And then i found some GPS Cracks (i described in the last post) who have tested the following Chip Generation.
The New Generation of uBlox Chips might do this with a mix of real time algorithms and Post Processing.
I know that there was an EEProm in the Layout which stores some data collected from the Sats.
But not only the Positions and stuff which allows a faster detection and action after a Warmstart.
There will be also stored Informations about latency, Atmosphere Correction Parameters,
Definition of Sats which might be faulty, correction factors for some purposes i don't know and a lot of other things.
..stuff which let the Chip correct the Signal errors from the Sats much better than before.
Also there as a step, if you feed the chip the correction data before,
there was no time needed on the Water to collect this data.
If you have not this possibility,
..normally for the First Time, startup,(totally cold start)
you have to collect this data with the Receiver itself.
This tooks some time.
I believe that some of this data runs over a period of about two weeks.
Also the positions and raceways of satellites i think will be stored,
to allow a faster pseudo warmstart.

But as i said, i was not a GPS pro.
I told this with my own words as i understood it from the Things i have read from the GPS pro´s.
End of last year I could get two prototypes from one of those cracks for about his self costs.
and this was almost more expensive than normal Receivers.
I believe actually there are not so many Receivers using the PPP-Line Chips from uBlox.
uBlox sells more cheaper chips which allows Receiver Prices about 50 Euro which might be the Products "Geo Cacher" and other People wants to buy.

Unfortunately i cant test the receivers, because i have sold my old humminbird unit,
bought a new Lowrance, sold my old boat, bought a new one and so on.
Therefore everything takes that long, i have not enough free time.

And really, one of those pros is using this Chip within a project on a tractor as i described before.
and this let me believe that the Receiver must be better than most others.
And honestly,
i do not expect every time errors less than 1 meter, but i expect getting the best values ​​i can Get.
And as I said, already for the HB-designed chip, I got much better values​​ feeding to the Lowrance unit.
There was almost never a situation in which the HDS has switched from the external to the internal receiver. (Both where mounted/located at same height and view to the sky, )

In 2014 all the GPS will be compatible with Gallileo for less than 1m (version standart) accuracy as well as the new version of the American GPS, it will update the GPS software. Lowrance and Humminbird have an gps antenna with compass integrated providing a good accuracy and real displacement of the vessel to drift due to wind and current.
Gallileo will be operational late 2014.
On my boat, I use a probe AIRMAR TM258 with an angle of 5 °. More accurate if more time on water. With DD I 'd compared the results with a tranducer at 20 ° and 5 ° of a wreck, the results are more realistic in 5 °.