I understand what you are saying and do not disagree. I have in the past held the published values in most cases as the gospel. The concept challenged my past practice. I have primarily used OPUS in remote areas where monuments were far away. As I have thought about this, I see that the OPUS solution is from a source that is running continuously where the referenced position is constantly monitored. The monument is based on a historical observation that may have lasted a few hours. Being a Second Order monument, it was probably not occupied for several adjustments. The OPUS solution would most likely be the closest to the State Plane Network. This is without even considering the affect of velocities as Jim mentions below.
Honestly, this ongoing discussion of RTK has caused me to review my procedures. In this area, a boundary survey with just RTK is generally impractical due to tree cover. Our primary use of GPS in a boundary survey is for control. We rarely perform a boundary survey that is not referenced to SPC. Most boundary control is based on a networked GPS RTK solution based on three observations spread out through the day as much as is practical. We determine an RPA for each control point. The idea of taking one RTK observation on a property corner is not defendable IMHO. Yes a static survey in many cases may be even quicker; however, leaving an unattended receiver is not worth the risk.
Kent I appreciate you and others challenging my practices. We should all know what we are doing and why we are doing it. If we can not defend it, we need to review what we are doing and why we are doing it. After 40 plus years, I am still learning.
> > So I take it you never actually turned on the RTK feature of that 5700 you bought and ACTUALLY gave it the good old college try then?
>
> When Trimble specs a horizontal accuracy of +/-(8mm + 1ppm) RMS error for RTK positioning with an R8, what would YOU expect to get out of it? Note that doesn't include antenna centering error, either. This is why a claim of +/-(2mm) activates the laugh function. :>
So that's a no then, right? You've actually never used RTK then, right?
Just wanted to be clear.
> > Is that question coming from the same guy that determines his own error estimates for angular accuracy for his theodolites and total stations? Suddenly the equipment manufacturer specs are golden? IMWTK
>
> News flash: GNSS receivers are fundamentally different from electronic total stations and there is no incentive for a manufacturer to underspec the accuracy of their equipment in the top rank. This is quite unlike the situation where total stations are being offered as lesser grade instruments, but with the same circles and centers as are used in high grade instruments.
5700 is NavStar only. Why don't you turn it on and test it? Oh, that's right. You'd hate to be wrong.
:good:
> In general, I am finding most of his called for monuments to be within a hundredth or two of his published location.
> Was cannabis involved?
:good:
> 5700 is NavStar only. Why don't you turn it on and test it? Oh, that's right. You'd hate to be wrong.
M'kay, you think that a Trimble 5700 receiver is going to have an RTK horizontal accuracy much better than the manufacturer's specification of +/-(1cm +1ppm) RMS? Does anyone else you know think that?
> > When Trimble specs a horizontal accuracy of +/-(8mm + 1ppm) RMS error for RTK positioning with an R8, what would YOU expect to get out of it? Note that doesn't include antenna centering error, either. This is why a claim of +/-(2mm) activates the laugh function. :>
>
> So that's a no then, right?
No, that's finding the optimistic belief that a major manufacturer like Trimble intentionally claims that their equipment performs much, much worse than it actually does to be humorous. Just think about that one.
> > 5700 is NavStar only. Why don't you turn it on and test it? Oh, that's right. You'd hate to be wrong.
>
> M'kay, you think that a Trimble 5700 receiver is going to have an RTK horizontal accuracy much better than the manufacturer's specification of +/-(1cm +1ppm) RMS? Does anyone else you know think that?
So you haven't used the RTK function of the 5700 and multiple observations to review positions? Okay. Just checking.
> > > When Trimble specs a horizontal accuracy of +/-(8mm + 1ppm) RMS error for RTK positioning with an R8, what would YOU expect to get out of it? Note that doesn't include antenna centering error, either. This is why a claim of +/-(2mm) activates the laugh function. :>
> >
> > So that's a no then, right?
>
> No, that's finding the optimistic belief that a major manufacturer like Trimble intentionally claims that their equipment performs much, much worse than it actually does to be humorous. Just think about that one.
So you're still admitting that you have no practical knowledge of how to use RTK to achieve the same specs as a static point. Okay. We were just checking.
> > M'kay, you think that a Trimble 5700 receiver is going to have an RTK horizontal accuracy much better than the manufacturer's specification of +/-(1cm +1ppm) RMS? Does anyone else you know think that?
>
> So you haven't used the RTK function of the 5700 and multiple observations to review positions? Okay. Just checking.
You realize how crazy your premise sounds, I hope, if you're thinking that the Trimble spec for any of their GPS receivers is much more inaccurate than the equipment reliably delivers. :>
> So you're still admitting that you have no practical knowledge of how to use RTK to achieve the same specs as a static point. Okay. We were just checking.
I think your obvious problem is with Trimble if you think they are misrepresenting the accuracy of their equipment. Why not call them up and set them straight? They might even pay you for your services as an expert consultant who will enable them to re-spec all their equipment four times better than that they have felt at liberty to do. :>
> > > M'kay, you think that a Trimble 5700 receiver is going to have an RTK horizontal accuracy much better than the manufacturer's specification of +/-(1cm +1ppm) RMS? Does anyone else you know think that?
> >
> > So you haven't used the RTK function of the 5700 and multiple observations to review positions? Okay. Just checking.
>
> You realize how crazy your premise sounds, I hope, if you're thinking that the Trimble spec for any of their GPS receivers is much more inaccurate than the equipment reliably delivers. :>
So, to recap, my premise is that you do not use RTK and do not understand how to make repeatable observations with it. Is it that crazy, or did I miss the memo?
> > So you're still admitting that you have no practical knowledge of how to use RTK to achieve the same specs as a static point. Okay. We were just checking.
>
> I think your obvious problem is with Trimble if you think they are misrepresenting the accuracy of their equipment. Why not call them up and set them straight? They might even pay you for your services as an expert consultant who will enable them to re-spec all their equipment four times better than that they have felt at liberty to do. :>
No, my obvious problem is that I had the misfortune to have taken statistics and understand what multiple observations do to those specs and how to evaluate them. It is unsightly I know, but hey, it's the reason we doubled the angles so that we could improve on DIN specs. :>
> So, to recap, my premise is that you do not use RTK and do not understand how to make repeatable observations with it. Is it that crazy, or did I miss the memo?
It seems remarkably obtuse. "Crazy" is so permanent and I'm an optimist.
> > I think your obvious problem is with Trimble if you think they are misrepresenting the accuracy of their equipment. Why not call them up and set them straight? They might even pay you for your services as an expert consultant who will enable them to re-spec all their equipment four times better than that they have felt at liberty to do. :>
>
> No, my obvious problem is that I had the misfortune to have taken statistics and understand what multiple observations do to those specs and how to evaluate them.
In other words, you aren't questioning the basic Trimble specification of +/-(8mm + 1ppm) RMS for RTK vectors via an R8 system? If so, I'm glad to hear it.
> > > I think your obvious problem is with Trimble if you think they are misrepresenting the accuracy of their equipment. Why not call them up and set them straight? They might even pay you for your services as an expert consultant who will enable them to re-spec all their equipment four times better than that they have felt at liberty to do. :>
> >
> > No, my obvious problem is that I had the misfortune to have taken statistics and understand what multiple observations do to those specs and how to evaluate them.
>
> In other words, you aren't questioning the basic Trimble specification of +/-(8mm + 1ppm) RMS for RTK vectors via an R8 system? If so, I'm glad to hear it.
No. In other words that you can't get around, I was speaking about your Trimble 5700, not an R8, and the "magic" that happens when you have multiple measurements to the same point and the statistical evaluation of those numbers.
> > In other words, you aren't questioning the basic Trimble specification of +/-(8mm + 1ppm) RMS for RTK vectors via an R8 system? If so, I'm glad to hear it.
>
> No. In other words that you can't get around, I was speaking about your Trimble 5700, not an R8, and the "magic" that happens when you have multiple measurements to the same point and the statistical evaluation of those numbers.
Well, considering that I don't have a 5700 receiver, I was trying to figure out what you have in mind. Basically, what you're saying is that if you take an RTK system such as a Trimble R8 that can position something with a horizontal uncertainty of +/-(8mm + 1ppm) RMS, then if you want to position something with it by RTK with a horizontal uncertainty of +/-(2mm) RMS at 5km from the base, you'd expect to repeat the measurement 36 times from independent intializations and form a weighted mean of all.
Who do you know who does that with RTK on a regular basis?
> > > I
> Well, considering that I don't have a 5700 receiver, I was trying to figure out what you have in mind. Basically, what you're saying is that if you take an RTK system such as a Trimble R8 that can position something with a horizontal uncertainty of +/-(8mm + 1ppm) RMS, then if you want to position something with it by RTK with a horizontal uncertainty of +/-(2mm) RMS at 5km from the base, you'd expect to repeat the measurement 36 times from independent intializations and form a weighted mean of all.
>
> Who do you know who does that with RTK on a regular basis?
Huh. I could have swore that you posted that you bought one and were considering firing up the radio. That would be my mistake if I dreamed that up.
To answer your last question, probably more than you think. Considering that each shot has "multiple" measurements in it already, then really, the amount of measurements for each shot is dependent on the amount of time one observes the point. An Observed Control Point has 180 observations that are then averaged. How would one be able to determine the residual of a point without multiple observations. Also, the independent initilization is but half of the equation. Observing under a different (albeit slightly) constellation of SV's is just as important and is illustrated by every diagram I've ever used when processing static data.
That being said, we do that a few times for critical points. Each case is different, but at least three different points, under three different initilizations, allowing the constellation to change. Do I care if a Center of a County Road shot is that tight, well, no. Do I care if my corners and traverse points are, well, yes. Multiple observations of the same point over time are, as the NGS promulgated, in their paper on the subject, necessary to achieve standards they are good with. So, if I follow those procedures, then I should, be able to look at 5mm + 5 ppm over a 95% interval vs. 68% or RMS for each shot. Clearly this is what is done with total stations in observing multiple observations to the same point.
So, yes, I do know VERY many people who do exactly what I'm recommending and practice myself. How many surveyors, other than those on this board, have you questioned about their methods for determining their positions using RTK? Clearly you have met with many.
> To answer your last question, probably more than you think. Considering that each shot has "multiple" measurements in it already, then really, the amount of measurements for each shot is dependent on the amount of time one observes the point. An Observed Control Point has 180 observations that are then averaged. How would one be able to determine the residual of a point without multiple observations.
Well, am I the only one of us to actually read the Trimble specification for the R8?
They claim an RTK horizontal accuracy of +/-(8mm + 1ppm) RMS with the following qualifications in a footnote:
"Precision and reliability may be subject to anomalies due to multipath, obstructions, satellite geometry, and atmospheric conditions.
"The specifications stated recommend the use of stable mounts in an open sky view, EMI and multipath clean environment, optimal GNSS constellation configurations, along with the use of survey practices that are generally accepted for performing the highest-order surveys for the applicable application including occupation time appropriate for baseline length.
"Baselines longer than 30 km require precise ephemeris and occupations up to 24 hours may be required to achieve the high precision static specification."
In other words, the spec they quote is for a result from an occupation of the recommended duration for the baseline length, not a single epoch.
> > To answer your last question, probably more than you think. Considering that each shot has "multiple" measurements in it already, then really, the amount of measurements for each shot is dependent on the amount of time one observes the point. An Observed Control Point has 180 observations that are then averaged. How would one be able to determine the residual of a point without multiple observations.
>
> Well, am I the only one of us to actually read the Trimble specification for the R8?
>
> They claim an RTK horizontal accuracy of +/-(8mm + 1ppm) RMS with the following qualifications in a footnote:
>
> "Precision and reliability may be subject to anomalies due to multipath, obstructions, satellite geometry, and atmospheric conditions.
>
> "The specifications stated recommend the use of stable mounts in an open sky view, EMI and multipath clean environment, optimal GNSS constellation configurations, along with the use of survey practices that are generally accepted for performing the highest-order surveys for the applicable application including occupation time appropriate for baseline length.
>
> "Baselines longer than 30 km require precise ephemeris and occupations up to 24 hours may be required to achieve the high precision static specification."
>
> In other words, the spec they quote is for a result from an occupation of the recommended duration for the baseline length, not a single epoch.
Or in other words, that was for the static function of the R8 because no one I know makes RTK shots at 18.6 miles.