A couple of years ago I downloaded a copy of Star*Net and ran the ten-day demo to evaluate its usefulness for work (employed with the state DOT), and shortly thereafter bought a copy of the Standard version for myself as the geek in me found it quite satisfying. I often e-mail myself the survey networks from work (which shouldn't be an ethics issue as I do NOT use any of the info whatsoever for personal gain), and will generate what-if scenarios, etc. just for fun. No kidding. I have a blast with it.
Is Star*Net addiction a treatable illness?
No. It isn't.
There MUST be some sort of 12-step, or similar, program available...?
😀
Unfortunately, no. Leads to harder stuff, like Columbus and the tau criterion, and early mornings pondering the difference between condition equations and observation equations.
The first step in rehab is to hang out with the fewest surveyors possible, or in Mrs. BajaOR's terms, hang out with the "least squares".
BajaOR, post: 377864, member: 9139 wrote: The first step in rehab is to hang out with the fewest surveyors possible, or in Mrs. BajaOR's terms, hang out with the "least squares".
Groan. :-D:good:
SellmanA, post: 377851, member: 8564 wrote: A couple of years ago I downloaded a copy of Star*Net and ran the ten-day demo to evaluate its usefulness for work (employed with the state DOT), and shortly thereafter bought a copy of the Standard version for myself as the geek in me found it quite satisfying. I often e-mail myself the survey networks from work (which shouldn't be an ethics issue as I do NOT use any of the info whatsoever for personal gain), and will generate what-if scenarios, etc. just for fun. No kidding. I have a blast with it.
Is Star*Net addiction a treatable illness?
I could be perceived as enabling the habit, however I will with this summary of input data types and inline options which can be a handy quick reference to have.
Star*Net is a fundamental land surveying tool, nothing else. If the State didn't think you should have a sledge hammer because of insurance issues and you had just discovered how much easier it is to drive a rod with a sledge than a 16 oz hammer, would you call the sledge an addiction? I hope not. It is a clarification or revelation, possibly, as one discovers how easy some things get with it that were a struggle without it.
In the immortal words of a surveyor named Pablo Reid : "You can have my Star*Net when you pry it out of my cold, dead hands."
I still find it somewhat strange that I actually have FUN using Star*Net, especially on the current project I'm dealing with- we referenced out 21 monuments through a three-mile corridor with a total of 45 control points. It looks like one very long, fuzzy-legged millipede with all the redundant measurements I insisted on.
I certainly agree that it is a fundamental land surveying tool, particularly in this day and age. And used properly, an outstanding one at that.
FWIW,
While in the past there were big problems with a lack of redundancy, it seem now that some add redundancy that does not contribute significantly to the projects accuracy. To what extent do software package users do scenario testing to optimize projects with only the redundancy needed to meet project goals?
Those interested in this issue might want to examine Shanking Kuang's "Geidetic Network Analysis and Optimal Design" published by Ann Arbot Press in 1996.
I rather enjoy writing my own scripts using Matlab. As I am retired I do not concern myself with making it "user friendly" as I am the sole user. That said, StarNet, Columbus and some other commercial packages are great tools. I have only used the demo versions. The in-line documentation for Columbus is very good.
Cheers,
DMM
GeeOddMike, post: 378009, member: 677 wrote: To what extent do software package users do scenario testing to optimize projects with only the redundancy needed to meet project goals?
Star*Net has a pre-analysis feature that is very useful for projects where there are lots of measurements to be made and the time spent figuring out how to most efficiently meet some uncertainty target would be warranted.
On most land survey projects I work on, the usual method is to begin with a combination of GPS vectors and conventional measurements, using the minimum number of observations necessary to give a reliable result, examining the uncertainties in points positioned, and adding additional observations as needed. The typical reasons for results not to meet expectation are sometimes due to unusual conditions that would resist modeling beforehand. That includes problematic vector solutions and abnormal seeing.
SellmanA, post: 377851, member: 8564 wrote: Is Star*Net addiction a treatable illness?
The remedy for an addiction to StarNet Standard is StarNet Pro. Start adding in your GPS vectors and you will forget all about Standard.
Giggity giggity.
Mark Mayer, post: 378019, member: 424 wrote: The remedy for an addiction to StarNet Standard is StarNet Pro. Start adding in your GPS vectors and you will forget all about Standard.
Giggity giggity.
We have Star*Net Pro at work. I don't want to feed my addiction TOO much at home!!
Yes, Star*Net Pro is definitely the next level. Being able to combine conventional survey observations with observations made by completely different methods, as those from which GPS vectors are derived, introduces a mutual reality test for both. Definitely recommend.
I don't use StarNet and also don't understand what the great appeal of it is, at least for the jobs I do. You traverse around a city block, single angles, close by a few hundredths, good. Or run around a larger parcel, get a tenth or two misclosure, what's the problem? For the typical jobs I do, just don't see the benefit of taking the time to make multiple observations so you can run it through this program to adjust everything when there is not much of anything to adjust out to begin with.
Kent McMillan, post: 377897, member: 3 wrote:
In the immortal words of a surveyor named Pablo Reid : "You can have my Star*Net when you pry it out of my cold, dead hands."
Who the heck is Pablo Reid?:-S
roger_LS, post: 378128, member: 11550 wrote: I don't use StarNet and also don't understand what the great appeal of it is .....
If your work is always a simple traverse around the block (really? that seems unlikely) and it always closes within a few hundreths (what bout when it's 5 hundreths, 6, 7 - what's your limit?) then maybe you wouldn't get a whole lot out of StarNet. But if your data is correct running it through StarNet will take 5 minutes. If it isn't correct there is no better way to isolate, weed out, and fix the problems. And just because you have StarNet (or any other adjustment package, including a spreadsheet for compass rule adjusting) doesn't mean you need to run every survey you do through it. But, once assimilated, you probably would want to.
roger_LS, post: 378128, member: 11550 wrote: I don't use StarNet and also don't understand what the great appeal of it is, at least for the jobs I do. You traverse around a city block, single angles, close by a few hundredths, good. Or run around a larger parcel, get a tenth or two misclosure, what's the problem? For the typical jobs I do, just don't see the benefit of taking the time to make multiple observations so you can run it through this program to adjust everything when there is not much of anything to adjust out to begin with.
I'm the same way. Most of my surveys are tiny little lots and my traverses are open traverses. The few times I traverse a block, it closes within a couple hundredths either way. The time I finally screw something up I'll download this and see what it says, but with all the back site checks and quickly using stake point on just shot traverse points to make sure it's dead on, it's hard to scruff something up too much when the traverse isn't a long one.
If your traverses are open I would expect that you would want to add more quality control to your operation rather than less. The traverse we run in tunneling operations are by definition open and as such we build the network with a reasonable amount of redundancy, the measurements are processed in Star*Net before the control is final and again after Gyro QAQC and traverse closure.
Traverses aren't always closed but usually you have a check on the record where you're tying into monuments and agreeing within reason with record positions so you know you don't have a blunder.
The ALTA standards talk about running surveys through least squares, I also don't completely understand this. Does this mean every single monument, control point, building corner, site feature, sideshot, etc. would be tied in from multiple control points. Isn't this the only way to get the redundancies needed? Is this what folks are doing?
