When you were doing JD's test 1 below you said you raised the mask to 18° and made sure you had 5 or more sv's....If I understood correctly.

After raising the mask then did you time trim out any/all portions of the session that had lower than 5 sv's, or was the 18° the highest you could raise the mask and get 5 sv's throughout the session ??

Have you tried constraining the CORS points and only processing the vectors to POST??

Jimbo

I couldn't get any redundancy over 18 degrees due to the distance involved. I felt that I should try and keep five throughout each solution, although at times there were only four in bits of a solution. I didn't worry to much about this, as long as it was just a sv dropping below the mask and not a cycle slip..I experimented around from 15-21 and 18 seemed to be the best overall fit.. Found several articles that expounded on long baselines where below 15, then the Tropo goes to heck and there's nothing you can do with it..

On each solution I provided (where I noted what was fixed as a single point), then I disable the rest of the data and only used the single vector to calculate the position. The only reason I processed between the CORS was to check their validity...Since I haven't used those particular sites, then I wanted to assure myself that the CORS were 'pretty close' and that they agreed with each other...I've heard on this board, where sometimes the CORS stations have some discrepancie, so thought it best to do a little checking..

TM

Do you have any feeling for the solution differences from the software for Yellow and Blue ?

I've got your results as posted and whan I did. When I do what I think you did.....
set time from 6:00 UTC to 14:00 UTC
set mask angle 18°
exclude all but 4 vectors(exclude Houston)
Hold 1 CORS site at a time as fixed H and V
No time trims within the data session
95% confidence level.....

I get pretty big error residuals using just one CORS (fixed) to POST, no matter which CORS site I use.

If I fix all 4 CORS then it comes down to the .06' residual error range. If I include HOUS (fixed) then it goes up to the .1' range.

Jimbo

I also got large residuals, but I think our software is just failing because of the line lengths...With this much distance and they way it computes residuals, they seem very large, but the actual coordinate doesn't seem to fluctuate much..

Also, I did cut out some start's and stops (but cut the entire segment out, sometimes this was the first or last 30-minutes to an hour)
Disabled noisy SV's and took out the cycle slips....

There's so many variables, we'd almost need to do a weekend retreat and set together and do a side by side comparison to insure that we were all doing the same thing..Hey..That's not a bad idea at some point in this experiment..

I firmly believe that both software/hardward brands are capable of the same results, given the same stipulations...

Also want to note, that there is no intent on my part to compare yellow versus blue..Just have fun and see what we can determine on the long L1's...

TM

No intention of deliberately comparing the 2 either. What I was seeing when I say large is in the 1-2' range per single vector....not bad for 125+ miles.

I agree with the vast amount of variables it is virtually impossible to duplicate efforts exactly........but then in a way isn't that what we want to be able to do ???

Without any sv cuts for noise or cycle slips, just mask angle and "lining up" the start and stop times, I am able to achieve better than 1:2million......
.06' error residual using vectors from the 4 CORS, leaving out HOUS.

That is why I am so curious as to what dual frequency does at approximately the same distance vs CORS, and what OPUS provides at those types of distances ???

DA FUN is just really beginning !!!

Crawdad season real soon !!!!

Jimbo
Modified By James Webb on 12/19/2001 at 9:25 AM

You wonder what an L1 only solution between the CORS vs the L1/L2 solution difference would be....Okay...I'll get on it and see if that provides any insight into our little (but growing) experiment...

I did re-submit a point today to OPUS now that the Precise was available for the session..Now understand that is dual frequency so...
This is all in US Feet...and on point PID:EM1092

Using Ultra Rapid;
N 476788.226
E 1851779.968
Elip 1056.156
Elev 1140.536

Using Rapid;
N 476788.226
E 1851779.964
Elip 1056.149
Elev 1140.529

Using Precise Ephemeris;
N 476788.229
E 1851779.961
Elip 1056.140
Elev 1140.522


Discounting the Elevation, which is based on the same separation calculation, then what can we assume?
N .226 .226 .229 (no appreciable difference)
E .968 .964 .961 (some variance, but still no appreciable difference)
Elips .156 .149 .140 (GOOD GOD, WE HAVE A HUNDRETH of difference)....

Published Data Sheet;PID 1092
N 476788.252 (0.02 from Precise)
E 1851779.958 (.0003 from Precise)
Elip 1056.100 (0.04 from Precise)


At this point...I'd like to ask the NGS what is the difference in the broadcast eph vs the ultra rapid (which I had thought would be the same thing)...

Baseline lengths to this session were;
109 miles to VCIO
55 miles to PRCO
149 miles to Imno

This sesssion was about 6 hours and the OPUS sheet reported 99% OBS used and 99% fixed AMB with an overall RMS of 0.012M.


Questions now arise on the premise that the precise eph is much better than the broadcast..If the Ultra is better than the broadcast, then I can buy this , but if it's the same, then something is breaking down in our plan...

Although, in our experiment, the stats improved using the precise, the actual position didn't tend to flucuate all that much...

TM

And Jimbo....Bring on the heads and I can hold my on with booray (don't have a clue on how to spell it)...

Yes, that is smoke u see rising slowy from above the collar region.......slow broil inside the cranium.

Booray.....spelled just as it sounds...like all good coonass things and names.....But the really important question is HOW did you learn to play ??? And you do then understand the SINGLE most important rule in booray...HOUSE RULES APPLY !! That can be a real stickler sometimes. Has made for a few life threatening situations in POT LIMIT.

Suck the heads, eat the tails and deal'm,

Jimbo
Crawfish Kitchen
SW LA.
O.R.G.I.

Soon be over my head I think. But at least the gears inside the cranium will cool off. Currently in overuse mode. Must be approaching end of week. Better yet, start of long holiday. About time to remove poor dead brain cells and allow the few remaining to flourish....

Plan on getting base post...O.R.G.I 2 ...built next week during time off....provided weather and buddy cooperate.

Jimbo