I add number in reg 139 now look like this:
Part Number:006-D2617-00
Rev: 01
Flash this region over SD and let me know if accuracy change.
Printable View
I add number in reg 139 now look like this:
Part Number:006-D2617-00
Rev: 01
Flash this region over SD and let me know if accuracy change.
@lolypop000
I have clarified by editing my previous post. In html file part number has no value.. That is why I have quoted part no from GarminDevice.html
If you flash with Attached Files from my previous post you will have part number in html ;)
One time i already add number in reg 139 i think for 23xx and after nuvi show part number, then something work better, but i forget what.
I remember now what i do :)
It was muvi 22xx and after add Part Number from 23xx or 2250 to nuvi 22xx reg 139, then nuvi 22xx show JCV.
Loly, I decided to be the guinea pig in our "lab", lol .. but before I was careful to backup the region 139.
Now the part number appears on "supported_features.html" file (is "part number" of what?)
I tested before and after the alteration and the best accuracy was 4 meters. I also tried with the option "predicted eph and cold start", but the best result was also 4 meters (static measurement, open sky, almost cloudless, 10/11 satellites). It would then be a good result? I think not. First because it is a static measurement and the most important is the dynamic, beyond what many factors can influence the accuracy. Second, because I believe that only with a 3790 original side by side we can be assured of installed implements.
Supposition? Not so much .. Yesterday I made a comparative test, static and dynamic, between my 3790 and a modified 3790 original, with and without "cold start" .. and the results were always worse values for the 3790 modified, ranging from 40% (static) to over 200% or more in motion, always with both side by side. Before the alteration of the region 139 nothing was done and still I obtained 4 meters today, which seems to be just a more favorable condition of the day and location. As after the alteration of 139 rgn I obtained 4 meters too, so I can say that nothing has changed in relation to the measurement more careful what I did previously. I have video and screenshots and publish here until tomorrow..
Anyway let's collaborate to the extent of our possibilities for the solution of the problem. Loly, its initiatives are great and we just have to thank you for your collaborative and selfless effort.
Beyond the question of correct fixation of Ephemerides by the 3490 firmware in the hardware of 3790, I still believe that the issue of imprecision also has to do with the a constant of correction or the own algorithm itself, in the two different firmwares. Knowing where this algorithm and how to change it, even if for just insert the constant, is a task for the few, so we rely on you and whoever else wants to collaborate, of course
Angelozip will not be the only one guinea pig in Lolypop000's lab.;))
My new tests results from today. The number appears in html - no problem.
Unfortunately nothing significant changed in accuracy. As you may see, I have like before two cases shown below:
Case one: outside test, cloudy sky, no close buildings, hills etc.
[Only registered and activated users can see links. Click Here To Register...]
Case two: in the car driving in the city. Best reached accuracy 7m., worst – 18 m. average 12 m.
[Only registered and activated users can see links. Click Here To Register...]
I should say that again the accuracy was too much influenced by buildings, trees etc. The effect of not clear view is immediate 15 sec. max, and I'm getting worse accuracy.
Like Angelo my opinion is that both hardware and algorithm for accuracy and positioning corrections has been changed. And most probably it was easiest way not to change old algorithm, but to make entirely new one.
If I remember well there was introduced Garmin Guidance II two years ago. I’m not sure that this has something to do with the accuracy so those who know better – please do not laugh on me.
@Lolypop000 – let’s make a plan for test and to try whatever you think that might work. I’m not happy to have a unit with good software and doubtful accuracy so I’m ready for hard test. In the end of the day - new series are coming – maybe it is time to change the unit ;).
So Loly - if you have time to learn me dancing I'm all yours;)).
Hi Loly,Quote:
Originally Posted by lolypop000 [Only registered and activated users can see links. Click Here To Register...]
When I bought my Nuvi it was.
Part Number:006-D2617-00
Rev: 0A
Now it it same as yours Rev: 01
Do you think this is important ?
Should I change it back to 0A ?
I add number in reg 139 now look like this:
Part Number:006-D2617-00
Rev: 0A
Flash this region over SD and let me know if accuracy change.
Done, no big change in accuracy and time to improve it. It looks a bit better because from my balcony - 8 meters accuracy with half of the visible sky is OK. Tomorrow morning will make one more test from the car, but I don't think that this is the decision.
Should I revers SW to 3790 and back to 3.50 on 3490 and then 3.60 up to 7.10 and then to 7.60? Nobody did that I think or I'm missing something?
I flashed it and it changed in html and Inventory.
Maybe someone with a converted unit can try.
Both Angelo and isg have converted units. No point trying in mine as i have acceptable accuracy with no value for part number, only showing Rev: 01. Maybe problem is elsewhere or a combination of factors?