At least in this limited test they seem to have similar performance. The F9P path is well troddened so there is much more data on how it performs in various situations. There is a learning curve with the UM982. The documentation is more scarce and some of it is ambiguous so experimentation is needed to figure out the best configuration for AOG. Hopefully other more skillful than I will experiment with the UM982 to increase our collective body of knowledge on this alternative. The price point for 1 UM982 versus 2 F9P’s is definitely compelling.
1 Like
Very good! so when do you want a dual mode test in motion?
After I complete the following; new horse stalls, new fence between pastures and barn, complete flail mower hookup to tractor, paint bedroom, fix front pasture fence, paint front pasture fence posts, re-gravel the driveway, complete new access roads around barn … 
The list is never ending. I sneak a couple hours a day to play with AOG stuff a few times a week. 
Every information is useful but for most of us here the useful GNSS receiver operating point is below one inch error. Would be nice to see some comparison there.
I think we need two things from a GNSS receiver; pass to pass consistency / accuracy and stability. Pass to pass keeps the rows evenly spaced and parallel to each other. Stability keeps the the rows straight.
I can probably do something with plotting stability and the current positioning charts show show stability but need to be larger to show how much the line varies.
Not sure I could do pass to pass. Would probably need some sort of trolley/platform on a track to make sure the antenna/receiver follows the exact same path on each pass. I don’t think driving a vehicle along the same path would give 1" or less variation of the platform.
Put antenna in fixed spot. PPP 24h survey, then messure average distance deviation (error) / max deviation from that point from PPP to real time position?
Then connect to base and use that antenna as rover and messure with rtk on. That should give rtk accuracy. Ideal base should be 3 band, as well as rover antenna. Then we shall see if um982 can get 8mm accuracy.
You must correct rover PPP cordinates to base station datum.
I already run a base station so that part is done. I don’t have a survey grade antenna, just the U-blox ANN-MB-00’s from Ardusimple. The specs show it is dual band. Best I can do is try the F9P and the UM982 with the same antenna.
If anyone would like to loan or donate a 3 band survey grade antenna I would be happy to test.
So base station is 3 band already?
You dont need survey grade antenna any 3 band antenna should work. Using 3band aliexpress antenna should give closer results to real performance, as many will use those antennas. Not many put survey grade on tractor cab.
But you can try with ann mb antenna and compare. Then full potencial of um982 is not utilised. But should give direct comparison. As 3band are cheaper then ann mb for new installs 3band should be used.
I suggest writing a batch script to configure the receiver. I don’t understand much about this myself. Here’s the beginning
echo off
echo "Enter port number, and press Enter."
set /p portNumber="COM"
echo "Port number is COM%portNumber%"
mode COM%portNumber% BAUD=115200 PARITY=n DATA=8
echo %\n%Unlogall > COM%portNumber%
echo %\n%Gpgga com1 0.1 > COM%portNumber%
echo %\n%Gpvtg com1 0.1 > COM%portNumber%
type COM%portNumber%>CON
pause
Experienced people tell me how to do it better
The base is an ANN as well. Interestingly the Uprecise software shows the UM982 is picking up signals from B3, L5 and E5 in addition to B1, GPS L1/2, Glonass L1/2 and E1. The reception strength of the “out of spec” signals is better than some of the “in spec signals”.
Even so, your base will not send l5 corrections. Best is to make um982 base and rover for testing. (Putting 3band antenna on dualband reciver will not work).
So test for accuracy should be, take 2 tri-band antennas and mount them in fixed position (one of them can be your future base). PPP survey them in same time. Make one of them base, other one rover. Log rover rtk outputed position and compare to PPP cordinates of rover survey. Log for few mins or hour, then assemble data in some software like exel and calculate average deviation away from PPP cords. You can extract also max and 95% deviation. After that exchange rover um982 with F9P and take same log of position, calculate average deviation, max… then compere results, what is better.
(You can do this with your exististing base but then you need to convert rover PPP cords to base datum @PotatoFarmer also as said um982 is at disadvantage to f9p, also enable all 4 constalations on your f9p basestation for use with um982).
Is there a post from potatofarmer with his findings for the UM982? Don’t want to plow any already plowed fields. 
Taged him for converting cords to needed datum. Seen him write hire about that stuff.
No one did true accuracy test with um982. As far as I am awere.
tried to send a command to enable mmp. failed on um982. Applicable to: UM980 writed in Unicore_Reference_Commands_Manual_For_N4_High_Precision_Products_V2_EN_R1_1.pdf
$command,CONFIG MMP ENABLE,response: PARSING FAILD MISSING FIELD,*1D
Try lower case. I have seen some commands don’t work when uppercase. I always use lowercase. What version firmware does the UM982 have?
Here is my version string.
#VERSION,96,GPS,FINE,2285,474910000,0,0,18,456;UM982,R4.10Build9984,HRPT00-S10C-P,2310415000012-LR23B2224524818,ff274098521ef48b,2023/06/01*d73ea184
lower case not helped. How to update fw version?
#VERSION,97,GPS,UNKNOWN,1,105300,0,0,18,96;UM982,R4.10Build7650,HRPT00-S10C-P,2310415000012-LR23B2224524831,ff27d4987076f87c,2022/09/01*289ad0d7
Here is the procedure I used;
Do an freset to reset to factory defaults. Make note of your current settings.
Connect with Uprecise. Do an unlogall for COM1, COM2 and COM3.
Use the update function in Uprecise. I had to do it twice before it worked. Some of the upgrade GUI is still in Chinese even though I have the GUI set to english. I cut and pasted them into Google translate when needed.
Here is version 9984.
UM982_R4.10Build9984.pkg.zip (2.1 MB)
Thanks, now mmp enabled
Glad to hear it. Enabling it improved the stability of the readings for me.