Geodetic/adjustment software

Columbus is my answer to both questions.

It is incredibly robust software. Adjustments are geodetic, instead of being projected to a plane and adjusted as some software does. Works with total station data, GNSS vectors, leveling data. All in a geodetic context.

It is also not entirely user friendly. Every time I've used it (which is on the order of 2-3 times per year) I have to relearn it and figure out how to make my data acceptable to the software. Since acquiring some easier software that is less robust, I've been opting for the lazier approach as the results have been good enough with the easier software. Results are similar even if not quite as good as with Columbus.

> What is the best geodetic/adjustment software package you have used, and why?
StarNet. Because it just works. Because it enables simultaneous adjustments of GPS vectors, total station data, and level data. Because I like the format of the report. Because of the availability of data convertors for almost every manufacturers raw data fields. And because that allows mixing data from various manufacturers. Because the data files are all easily edited text files. But mostly because after using it for 15 years I know where the bodies are buried, so to speak.

I use StarNet almost daily to reduce topo data to coordinates. I just like to view the raw data, since I am rarely the one doing the data collection these days.

> What is the worst geodetic/adjustment software package you have used, and why?
I used Leica's LGO for about a year. It worked okay, but it would only accept Leica data, data was difficult to edit, and it is "reactive". Which means that their is no "GO" button. You change data and its supposed to automatically update. Trouble is, that is not the case 100% of the time. Frustrating and dangerous. If you are using Leica GPS, TS, and levels it is okay.

Carlson SurvNet is okay. In fact it's a lot like StarNet. I wasn't able to get the report just the way I wanted.

TGO was probably the worst. I only tried it once, it bogged down badly on a comparatively small data set. But too be fair, I was sore because it wasn't StarNet.

I've never used Columbus.

>But mostly because after using it for 15 years I know where the bodies are buried, so to speak.
>
:good:

I hope Mr. Sawyer made a pretty penny from MicroSurvey. I'm still running V6. I demo'd V7 (MicroSurvey). It's almost identical. Star*Net is really user friendly, once you understand the DAT file commands. Seamless integration of GNSS, terrestrial and differential measurements. I've used TGO, TBC, MAGNET; nothing compares. I too like the format of the listing file (adjustment report). You can tweak the options to give you as much or as little information as you like. Applying geoid and deflection models are straight forward. MicroSurvey has a converter for most all raw data formats, making it much easier to input terrestrial data into DAT format (MicroSurvey's converters work well with the V6 platform). I will process GNSS data with what ever software works with the equipment I am using, but I will continue using Star*net for network adjustments. I see no reason to upgrade to V7 either.

> I'm still running V6. I demo'd V7 (MicroSurvey). It's almost identical. Star*Net is really user friendly, once you understand the DAT file commands. Seamless integration of GNSS, terrestrial and differential measurements.

And incredibly flexible. I use Star*Net V6 on every project where there are survey measurements to be adjusted, including topos and, of course, GPS networks. I've Star*Net to adjust transit and tape traverses run 100 years ago, which it did just as easily as a survey combining GPS vectors and conventional survey measurements run yesterday (figuratively speaking). I consider Star*Net to be a fundamental tool of land surveying. It's easy to learn to use and is backwards compatible in that it will run all data files a surveyor would have ever generated.

Ron Sawyer was a great developer who produced bulletproof software. I'm sorry that he's retired, although I'd understood that his associate Sean Curry helped MicroSurvey with V7 (which I've never bought since the enhancements over V6 seem fairly marginal).

When my dad first trained me on GPSurvey and Star*Net, we had GPS vectors and old California DOT field books to put a R/W back together. I remember being fascinated, watching the network grow and going out to find really old stuff right where the ol'timers told us to look!

> When my dad first trained me on GPSurvey and Star*Net, we had GPS vectors and old California DOT field books to put a R/W back together. I remember being fascinated, watching the network grow and going out to find really old stuff right where the ol'timers told us to look!

Yes, that is a cool use for Star*Net, adjusting old conventional traverses by combining some modern GPS-derived positions with them.

One of the more interesting projects I've used Star*Net on was reconstructing a traverse run in about 1910 along the Rio Grande river in West Texas. Nearly all of the stakes and bearing trees that the 1910 surveyor had tied to were long gone - either farmed out or cut down and rotted away in a salt cedar thicket - but there were enough observations in the 1910 surveyor's work such as ties to a rock mound that still existed, ties to a corner of a building that still stood, and angles taken to various features on mountains that could be identified and located by triangulation that the whole thing comprising miles of traverse could be put back together surprisingly well via Star*Net.

> What is the best geodetic/adjustment software package you have used, and why?
>
Leica Geo Office is my main platform. edm traverse, topography, leveling, GPS/GNSS all processed and adjusted simultaneously. it transforms your project to ellipsoid, then adjusts, then transforms back to grid (or whatever else you were using)

star net is good, and robust, and does these adjustments simultaneously, but i still have not found docs proving that this adjustment occurs in a geodetic sense. it is my expectation that adjustments are on grid only for starnet

> What is the worst geodetic/adjustment software package you have used, and why?

LGO, because it was very difficult to learn

> star net is good, and robust, and does these adjustments simultaneously, but i still have not found docs proving that this adjustment occurs in a geodetic sense. it is my expectation that adjustments are on grid only for starnet

Star*Net adjusts on a grid plane. According to its documentation, the approach is accurate for projects no larger than a few hundred kilometers in any direction.

I read that Star*Net will give output in SPCs or UTM.

Does Star*Net support custom projections?

> star net is good, and robust, and does these adjustments simultaneously, but i still have not found docs proving that this adjustment occurs in a geodetic sense. it is my expectation that adjustments are on grid only for starnet

Excerpt from the Star*Net Pro V6 User Manual:

"Just as with the adjustment of conventional observations, STAR*NET-PRO adjusts GPS vector observations right “on-the-plane.” During each adjustment iteration, conversions are made between the earth-centered Cartesian coordinate system and your grid system so that residuals can be calculated for the vectors and proper corrections can be applied to the stations. This allows conventional observations and GPS vectors to be combined in a very natural way. But this also means that all stations in your project must be somewhat localized so they relate to a single grid system. For example, in the United States and Canada, projects are run using a single NAD83, NAD27 or UTM zone. Jobs are not handled that span several zones."

Kevin

It's been a long time, but as I recall...YES

Loyal

Kevin

Thanks Loyal! I was hoping you would chime in on this!

> Does Star*Net support custom projections?
Yes, it does.

> > star net is good, and robust, and does these adjustments simultaneously, but i still have not found docs proving that this adjustment occurs in a geodetic sense. it is my expectation that adjustments are on grid only for starnet

Yes, Star*Net would not be the right choice for computing an arc of triangulation running from Kansas to the Pacific since it computes solutions in a plane projection. However, if the size of the project is such that the projection equations give millimeter-level conversions from Lat/Long to Grid N/E and back again (i.e. doesn't extend outside the limits of the SPCS zone by more than a few countiies) there should be no loss in accuracy.

I've never had a problem with it except for the times when I was experimenting with adjusting GPS vectors to a station in Texas from CORS sites in Canada. There is a workaround for that.