Discussion in 'Surveying & Geomatics' started by Darrell Andrews, Feb 4, 2011.
Are there any GOOD Least Squares Adjustment programs that I can download for FREE?
Yes, But You Cannot Use Them Commercially
Paul in PA
Yes, But You Cannot Use Them Commercially
I won't tell on you!
Darrell- Maybe http://www.microsurvey.com/products/starnet/download_form.php
not as easy as star net. but free, good ,right, etc.
some programs compute 100 station, other module even more:
good for learning, but again, don't use commercially
It seems like everyone forgets that Star*NET has a demo mode. Maybe because before MicroSurvey bought the program this past year, there wasn't even a website for it. Download the full version of the program and run it in DEMO mode. The only restriction is that you are limited to 10 points when you perform an adjustment. It should otherwise be functionally equivalent to the "Professional" version of the program (although, I notice that the converters seem to be disabled as well, but these didn't use to be included as a regular part of Star*NET before it was purchased by MicroSurvey anyway). You can still learn and do a hell of a lot with it. Maybe you will find it useful enough to buy a copy. I have to wonder how many people know about this and will find it useful?
Yeah, I am considering Microsurvey and Star*NET, but for now, I need something that can handle more than 10 points and doesn't rape my wallet.
This will be my first attempt using a Least Squares program, so it will be educational! Usually, the amount of work I have doesn't require it, but I have a bunch of control points in a job I want to test the concept on.
You can download a free version of survce for your desk top from Carlson. But I don't know if you can do LSA with it.
It isn't "FREE" (although there is a free DEMO), but it won't "rape" your wallet.
$725 is a GREAT price for a Program with COLUMBUS' power and ease of use.
I like that it seems to import OPUS and OPUS-RS solutions. Now what would I do with this...
Does not your GPS processing software (LGO, TGO, Topcon Tools etc.) have adjustment functions? Assuming you are needing it for GPS. Civil 3d has this as well, but I hear it sucks and it's DEFINITELY not easy on the wallet.
I use LGO, it's not that pricy for the processing and adjustment modules. Sorry, I know this was not what you asked but just curious.
No GPS here. Old school survey techniques!
Free Least Squares Adjustment Program
The link to Penn State for the LS program is the Wolf-Ghilani LS program that is included with their Least Squares adjustment text. I first used it as a DOS program that came with an early Wolf-Ghilani Elementary Surveying text. Subsequently I bought the LS text with a Windows version. I have submitted the exact same data to that program as to full blown Star Net with equal results. It is worth downloading and is an excellent learning tool.
Truth be told, I have yet to find it necessary to do a Least Squares adjustment of any of my actual surveys. I am so satisfied with my field traverse techniques and results that I can accept any included error as within my error budget. I make enough redundant measurements and checks with GPS that LS would only change the numbers, but in essence not change the facts. I do however have blunders that require reobservations and I do that. I would rather spend extra time in the field than chasing miniscule numbers in the office.
My GPS software does an LS adjustment because that is the only way it can output data. I accept that without complaint. I have set up traverses that would allow me to do a check with GPS and found at the end of the traverse that the check step is unnecessary.
You can use Least Squares to massage a pile of doggy doo and you will have a nice neat pile of doggy doo. Notice however that it is still doggy doo.
Paul in PA
> No GPS here. Old school survey techniques!
Ah. I do use LGO for my conventional adjustments as well. Not free like you're looking for though. Good luck with that!
Free Least Squares Adjustment Program
The primary purpose of LS is not to adjust things.
I think of it as a processing tool. It is a way to process all your measurements simultaneously to get the best results.
You can get coordinates out of a static network without doing LS but no one does because it is so simple to do.
If you put all of your measurements into one big adjustment you get a better answer, in my opinion. Also full LS allows you to use all of your tools to best advantage. For example you can close your conventional traverse with a static vector rather than being forced to have a closed GPS survey and a closed conventional survey. LS allows you to use all of your tools to get the measurements you need in any way that is most efficient for you.
The price if Star*Net could not be considered rape.
Star*Net has a multitude of uses, best tool in my bag of technical tricks.
Why waste days learning some funky dunkey package?
But, you know the value of your own time and effort best
Seriously, get the demo and work with it a bit.
The NGS has a venerable package ADJUST that is free, rigorous and is distributed with sample data and guidelines. I consider it a good tool to learn about adjustments. I provides rich numeric output allowing close scrutiny of the processing and results.
That said, creating the input files (constraints and parameters, a priori positions and observations) for your data might be more trouble than you want. Also running the various steps in the process are all command line operations.
There is even a NOAA TM 47 describing the program and its use.
Download page is:here
The information below was copied from that page.
ADJUST_ALL [2011/01/04] Download a zip file with all PC executables, support files and documentation. Does not contain source code files.
adjust_all.zip contains the Adjust suite of programs for the PC.
ADJUST [version 5.6 - 2008/10/27] Performs least squares adjustment on horizontal, vertical angle, and GPS observations.
adjust.exe is the PC executable.
adjust_suppl52.zip contains supplemental documentation for versions 5.2 and later
samples.zip has sample data files, common to adjust, chkobs, obschk, obsdes
adjust_v56.zip contains program source code
Constrained Adjustment Guidelines -- (2010/03/04) recommended adjustment procedures and step-by-step approach for adjusting GPS projects.
This is definitely a bad first LSA program. I even took a 5 days NGS PAGES and Adjust seminar, and wouldn't recommend these programs to anyone...
I find it hard to believe that every surveyor does not have a good least squares adjustment program (and is not proficient in its use). To me it is an absolutely indispensable tool. I would no more think of looking for a free LS program than I would try to get some free knives if I were a surgeon. Maybe this is indicative of the state of our profession...
This is not a personal attack on Darrell. I just think that a LS program is a fundamental tool that all surveyors should be using.
I use Geolab, and it has worked for many years for me. Very powerful. But, it does have a few bugs and is basically a dead program (i.e. no one is updating it). I have used Star*Net, but do not have it now. I didn't like it because it does things in 2D/1D rather than ECEF as Geolab does. I also have Columbus, but my way of operating is to write my own input files, and I have had problems doing that with Columbus. To be honest I have not spent a lot of time messing with it, but the ASCII text files that I made didn't work, and I haven't had time to try to fix them. Let's just say I have problems with it.
I have been interested in MOVE3, because, like Geolab, it does true 3-D adjustments. It also has capability to put in constraints like collinearity, parallel lines, perpendicular lines, etc. It is expensive, $2499. But, I believe that is in the range of what I paid for Geolab many years ago. If it works well, I am willing to spend that to get a really good program with good support. I use a least squares program way more than I use CAD, but that is because of the peculiar types of projects that I do. I downloaded the MOVE3 demo today, and tried to load a small GPS net, but it exceeds the limits of the demo (not sure what the limits actually are). They do have an option for $149 for a 60 day trial version.
I do use ADJUST from NGS, but only for GPS and only on projects which I submit for bluebooking. Not at all hard to use ONCE you get the input files correct (B files and G files).
I should add that I write a lot of software, and it is very time consuming and tedious at times. Good software is worth paying for.
Some (possibly out-of-date) info:
Global Cogo has DOS Burkord and WBK, both free.
Sayed R Hashimi (SRH) has Ez-Adjust, and Convergent Computations has Geoida, both under $400.
Don't know if they're "good" enough for you.
Separate names with a comma.