Great questions, Mark. I know we've discussed dynamic CORS coordinates, but presently I believe we're using the published positions of the CORS and we do not model for tidal loading. I'm sure it does make a difference in some areas. One thing I do like is that we have access to some of the individual CORS networks, so we can operate exclusive to the NGS entirely. I use the TxDOT CORS network most of the time because there are more stations in that network that NGS has excluded from the US CORS network.
I've attached a report from a short (2.5 hour) session I ran for a job on Saturday, May 19, 2018. I've also attached the raw data file (zipped) if you'd like to submit it to OPUS. The vertical HI is 1.5248m. For elevations, I'd typically prefer more than 8 hours, but for the purpose of this project, the 2.5 hour occupation was good enough. In this example, you'll see in the report that the DPOS solution used TxDOT CORS stations that were much closer than the CORS that OPUS would select for the same solution.
?ÿ
Shawn,
Does "EPHEMERIS: brdc" mean that the Broadcast Ephemeris was used?
Why not the igs Rapid?
Loyal
yes it's broadcast. not sure about the reasoning.
Such comparisons have been made and results posted. Nothing I??ve seen that is very recent.
See:?ÿ http://gpsworld.com/a-comparison-of-free-gps-online-post-processing-services/
I also did such a comparison and linked them in prior postings on this thread.?ÿ http://geodesyattamucc.pbworks.com/w/file/64369423/GPSsolutionsCompared_2013.pdf
Looking at the Javad results I see that it addresses the desire of many for SPCs. I went to the Javad site and could not find a description of how their processing works. I generally resist creating accounts and did not do so there. Anyone have a descriptive summary of its operation? BTW, rooting around their site I did find some efforts to deal with spoofing which I did not realize had become a significant issue.
Like the GPS World article, I found reassuring agreement in processing site results. The close agreement between PPS and differential solutions was also reassuring.?ÿ
Finally, the use of broadcast orbits while sub-optimal is no longer the large error source it once was. Remember that orbit error over distance to satellite is equal to baseline error over baseline length. In other words a one meter orbit error in a 20 million meter distance to the satellite yields a 4.5 millimeter baseline error on a baseline of 90 kilometers. Hope I did the math right...
I can't offer much, technically speaking, about DPOS except that the general approach is that vectors are processed using the Justin processing engine. We use 5 CORS sites, processing both GPS and Glonass, and adjust the vectors. As far as I know, we only use the published position of the CORS sites for the adjustment. One thing I like about our report is the "residuals of position" portion. This shows the difference from the endpoint of each vector from the overall average, making it easy to spot outliers and check for goodness of fit.
Remember that orbit error over distance to satellite is equal to baseline error over baseline length. In other words a one meter orbit error in a 20 million meter distance to the satellite yields a 4.5 millimeter baseline error on a baseline of 90 kilometers. Hope I did the math right...
I think you did the arithmetic right, but I don't understand your math.?ÿ Why is it proportional and not additive??ÿ
I'd think whatever orbit error resulted in a 1-unit error in the path length would make an error in the vector of (1 unit * trig functions) to convert to the components at the ground, reduced because both the base/CORS and rover see some of the same shift, and then averaged with the vectors from other satellites.?ÿ Complicated, and probably less than proportional.
?ÿ
See:?ÿ https://www.novatel.com/assets/Documents/Waypoint/Reports/Static_Accuracies.pdf
For a nice detailed discussion of orbit error impacts in the differential case see:?ÿ http://www.navlab.iit.edu/uploads/5/9/7/3/59735535/fang_master.pdf
I use the relationship I posted as a ??rule of thumb.? More detailed computation is a more academic exercise for me.
It seems they fool with the program, screw it up, it's Friday 5:00PM, time to go home. Come Monday it miraculously works again.
Why can't they fool with a program offline debug it themselves and then put it online.
It gets me pretty frustrated dealing with these people.
"Polite version of what I really feel!"
How much do you pay for this service?
Seems like an inappropriate question for any government service. I don't directly pay for most government services I use but it seems irrelevant. The question implies there should be no accountability of any government service unless it is directly funded through user fees. Furthermore, it implies that those who pay the most in fees or taxes should have the most (only?) accountability over that service.
You've implied all government services are free. They're not! Every April 15 I write a check to pay for them. If they can't competently run the programs they should get out and give the money to private enterprise. There you produce or get fired.
Every April 15 I write a check to pay for them.
The amount of your tax payment that goes to support OPUS is minuscule, likely too small to measure unless the mint brings back the mil.?ÿ I reckon you've used up about a decade's worth of complaints already on this thread.
I consider the service to be an outstanding value.?ÿ I regard the rare occasions when it goes offline to be no more an inconvenience than, say, when TBC (the big-dollar software made by people who either produce or get fired) crashes or slows to a crawl because it can't process more than a few hundred vectors at a time without overtaxing the OS.
?ÿ
Jim,
What about all the other millions of Americans that pay to run these bureacracys?
The service is not the complaint. It's the incompetent management that is the complaint. OPUS is a great program. Why does it keep crashing and doing stupid things one day and not the next. Poor management!
I love OPUS. I do not like it when it crashes over a stupid IT manager's incompetent actions or his furlough for stupid politics.
My use of OPUS is so minimal I?ÿ a mill would cover 5 years worth. How many users are processing 100 of vectors? Very few.?ÿ
Try and stay logical on your arguments. Some of you guys go way radical.
Gripes don't cost a dime. They're what made this Country great.
This post started out with someone simply wanting to know if others had experienced similar problems last weekend with OPUS.?ÿ No blame tossed about, just an inquiry to see if the problem might be with him or if OPUS was experiencing problems.
Others decided that because the outage screwed with their productivity that a good flogging of a
was in order.
They shared no facts, only their rage that they lost time because they were convinced their problem was caused by lazy government employees.?ÿ OPUS is a free service of the NGS.?ÿ I am sure the NGS would appreciate more money to be allocated to it and several other programs that they administer.?ÿ IMHO they do an excellent job with the resources that they are allotted.?ÿ I only use OPUS as a means of entering data into OPUS Projects, which is an exceptional program.
Gavin makes a very good point.?ÿ The NGS is very interested in user feedback and input.?ÿ I have received lots of very good information from the NGS folks.?ÿ I would imagine that an email to the OPUS folks asking about the outage or reduced performance over the weekend (as shown by the graph Loyal posted) would elicit a courteous reply describing what "happened".?ÿ
In the meantime, I will patiently wait for Wendell to implement the ignore poster feature.?ÿ?ÿ ??ÿ
Gene,
There are no posts here griping about lazy government employees, but your post. OPUS has worked flawlessly for close to over 20 years. Now lately it rejecting input files for rediculous reasons, using Stations outside the range it should be. It use to be said "It doesn't get any better than OPUS." Now the Guru's are double checking it's results and finding sporadic problems. The budget cut problem is BS. If you're working you're getting paid. I see it as a management problem. If you have pride in your job why would you walk out and leave a mess you know someone else is going to be using. If it's bad results they should be held accountable for it. If it's unreliable nobody will use it. It's a great program that could go down the drain because of mismanagement. You can bet there are NGS employees reading this forum every morning. If my gripes inspire them to do their jobs better, than yours, I've earned my mil of OPUS use.
I guess I'm not smart enough to detect when I'm being trolled on this Forum, because I seem to end up spending more time trying to clarify the fact and deny unfactual remarks. The real Guru's of OPUS have been suspecously quiet. If I was submitting a 100 vectors a day and kept running into these SNAFUs, I'd probably be down at the NGS Office raising hell with the head person.
I think you can block posters. I offended Wendall once and he sure blocked me.?ÿ
?ÿ
OPUS is free, I don't think many of the other services?ÿare cause you have to pay for the program. I don't use OPUS for much of anything except to check processing. I can't see complaining about OPUS when you send it to them and they do the work.
If OPUS is down then process it with your processor. It's not like the old days sitting in a room all day going over the results.
Processing takes barely more time than sending it off for someone else to do it for you. Actually less since you have to wait for the email return. .
It's technology. It's not a matter of if it will fail, it's a matter of when. Stuff happens, people fix it, life goes on. Some appreciate the service, the price, and the work that goes on behind the scenes that none of us truly understand -- because we aren't personally involved.
It's easy to stand outside the box and bitch about the box when you don't know what's inside the box.
Like what we do here? Donate
Need a new or refreshed website? Five Point Web Solutions
Looking for a web host? Website Hosting & Management
Processing in TBC is not only fast and easy, but also affords you much greater flexibility than OPUS. I can get a reliable fixed integer position almost anywhere in Louisiana with a ten minute session.
I've never been held up by OPUS, I just do it myself.
See:?ÿ http://gpsworld.com/a-comparison-of-free-gps-online-post-processing-services/
GeeOddMike, you realize that I am the author of that link? I had that article in my pocket when the government shut down and EricG put it on his web portal with a byline to me. I was told that it generated a lot of traffic for months. Editor could not figure out how I came up with it overnight, but I had actually worked on it for over a month. Glad to see it is still getting hits 🙂
...
It's easy to stand outside the box and bitch about the box when you don't know what's inside the box.
Be careful; it could be a dead cat
