RTK comparison (i80 v. R10)

Shoot yeah! I love a good shootout!

I'm looking forward to hearing about the results.

I'm running the R10s through the paces right now. So far I can't really say it's performing that much better than the R83 in and around high multipath enviros. Looks like the TS won't be collecting dust any time soon.

Recently took a hard look at the CHC stuff and was curious about this exact matchup. Also looking forward to the post-game.

Good luck to both teams.

will you have any good truth to compare it to, or just see how they track?

A seemingly good answer that is wrong is worse than no answer at all.

John Hamilton, post: 392685, member: 640 wrote: will you have any good truth to compare it to, or just see how they track?

A seemingly good answer that is wrong is worse than no answer at all.

John,

I consider this to be pretty much a simultaneous semi-independent comparison of the two units. I believe that the likelihood of BOTH units (different bases and obviously different initializations) returning the SAME BAD position is pretty small. Both units will be "dumped" between shots (as always), so all things being equal, the only real difference (if any) should be in initialization time, repeatability, and whether or not one unit can get a "fix" where the other won't. If one can, and the other can't, that certainly begs the question about the validity of the fix itself. I dunno...it would be nice to run a Total Station Traverse through all of this, but it isn't in the cards (or budget), and it would take at least a week to really get something meaningful.

Loyal

I have done some testing of the R10 in the woods, but the base was quite close (100's of meters). We did have very accurate ground truth, but I didn't compile the results. I do remember that points on the edge of the woods worked fine, but deeper in the woods it would not initialize (which was fine, I NEVER want a wrong answer, but I can live with no answer, like I said)

John Hamilton, post: 392689, member: 640 wrote: I have done some testing of the R10 in the woods, but the base was quite close (100's of meters). We did have very accurate ground truth, but I didn't compile the results. I do remember that points on the edge of the woods worked fine, but deeper in the woods it would not initialize (which was fine, I NEVER want a wrong answer, but I can live with no answer, like I said)

I agree John, and we have been pushing the i80 pretty hard the last couple of months. We HAVE HAD a couple of bad "fixes" (less than 1%), and many spots where it just would not "lock" at all. Most of these problem spots have been revisited several times (days or weeks apart), and some of them are just no joy.

Loyal

You might have an additional variable today, Loyal. The Kp index is showing some continued impact (Boulder station Kp = 4) from interaction between the solar wind and the heliospheric current sheet (See the Space Weather page for details). I was in open sky conditions (above timberline) yesterday and didn't notice any issues with Leica's GS14 (rover) and GS15 (base). Will be processing the base today to see how noisy it is compared to other days.

My test came about because we do a lot of USGS LiDAR control. One of the requirements is to do X number of checkpoints in all of the coverage classes that are present (bare ground, low vegetation, urban, forest, etc).

We do the forest points by setting a pair in the open with GPS and using a total station to shoot a point in the woods. Of course this is significantly more expensive than a point we can directly occupy (which are all of the other types). So, in their quest to reduce costs, they were trying to get us to do the woods points with GPS only, and they suggested getting it "on the edge of the woods", or "just inside the woods" using GPS only. I did some testing, and I told them I was not comfortable doing it that way (i.e. I did not want to certify the results as being .xx accuracy). They were wondering what other USGS contractors were doing, and they didn't think the others were going to the extra effort (set a pair, turn angles, etc) we were. But, I talked a company that does similar work at the Lidar conference in Denver, and they do it exactly the same as we do.

I don't mind doing it that way, the only hassle is taking along a total station (we usually fly commercial to the projects, as they could be anywhere but have typically been in the lower midwest).

Oky doky,

Here's the report from the field in a nut shell:

This is the second day that we have had these units side-by-side (last Monday & today). When one will fix, they BOTH will. When one can't fix, the other one won't either. When fixed, both units return the same coordinate +/- 1 centimeter or [usually] better. As near as I can tell at this point, there are only 3 real differences.

1. Software, Matt has been using Trimble software for a decade or two, and he is new to Carlson (which IS different). Being an old(ish) dog, he doesn't like learning new tricks.

2. Either the i80 firmware, OR the current Carlson SurvCE software, is not reporting the L1_PC to L1_PC vectors & covariance matrix. This may not be an issue for some folks, but it is for me. The X-91+ and Carlson SurvCE (which we have also been using lately) does NOT have this issue.

3. The PRICE! (BIG difference...REALLY BIG).

I'm sure that there are other differences that are of no real concern to us. We use RTK to get in the Zip Code, a probe, shovel, trowel, etc. to find the monument, and then use RTK to tie that point in. We also use RTK to SET points, but that only happens when we can't find the original, which is usually rather rare on most projects. In any case, the ‰ÛÏnumbers‰Û (coordinates) are the same, so both units work equally well for us in that regard.

Matt has already decided that he is going with the X-91+ system, so this is probably our last chance to compare the mini units. As near as he has been able to tell, the only difference between the i80 and the X-91+ is size and price (oh, and the vector & covariance thing).

Loyal

Thank you Loyal.
I was told that they BOTH have Trimble boards.....
assuming this is true....
Why would Trimble supply boards, to a Chinese GPS company?
So, the Chinese company could undercut them?
Just curious. What do you think?
N

Yes, it is my understanding that both the X-91+ and i80 use Trimble boards (different ones though). I was told over a decade ago by someone who should know, that the "boards" are considered a commodity, and are sold to MANY different "manufacturers" of GPS equipment.

I don't know (or understand) how "big" business works, I'm just a poor surveyor trying to get the most bang for the buck. My personal GPS gear is all over 15 years old now, and I don't use it very often anymore. BUT, it still does what I bought it to do (at least STATIC wise, which is all I use it for now)..

Loyal

I'd like to see a comparison between either of the 2 units you compared, and the Javad LS.
IF anybody around here has either of those units, and wants to compare, I am game. Loyal, you are far from here. But, I think it would be a great deal of fun to have you involved.
Nate

Nate

If it could be organised, I'd really like to see a comparison between Javad and R10 - especially under 'adverse' conditions.

Your descriptions of how it works have me very interested, although I dont think Javad is available down here yet

Did the R10 get upgraded to an X-91+? Where did it go?

Mapman, post: 393008, member: 6096 wrote: Did the R10 get upgraded to an X-91+? Where did it go?

The R-10 belongs to Ben, Matt has been testing the i80 and X-91+ to see which one (set) he wants to purchase. Looks like he X-91+ wins for the reasons stated above.

This project has been a "perfect" (bad news conifer & Quakie canopy) testing ground. Most of my/our work is usually out in the desert, but EVERYTING works ALL THE TIME out there!

Loyal

I thought that I should bring this back to the top, and add a quick update.

Matt decided that he liked the small size/weight of the i80 system, so that's the system he bought and has been using since late last year. There have been no further "shoot-outs" with other gear, and he (and I) have been very happy with the performance of the i80. There are places where it just doesn't work (no surprise there), and some areas (points) take quite a while to get warm-n-fuzzy solutions (multiple dumps, fixes, dumps, etc.).

I have seen the new ig8 units, and they look like a great system, maybe Mark (and maybe Matt too) and I can get out somewhere and play with them.

Loyal

How good is the onboard COGO and Stakeout with the Triumph LS?

does the tilt-tip correction work during stakeout?

Is the cogo firmware stable?