Thread: About 37xx converted to 34xx satellites reception accuracy

I've been traveling, away from civilization, but now I hope to be able to collaborate: I will write the two rgn Kunix's with a lot of attention and public the results tomorrow, but thanks very much in advance to Kunix for your interest in helping us. In relation to Kanupus's suggestion, a question that also I had in times past, that is, that the inaccuracy is apparent, I can assure you that it is a matter totally overcome for me, ie, the imprecision is undeniable fact, not only by my tests but also those of numerous other users and has relationship with the positioning registered, yes! In addition, the relation of accuracy shown and positioning in the tracklog has a direct relationship on Garmin outdoor gps, which can be seen easily when we are following a trail, and appears to be the result of same calculation algorithm of position. So why would be different in road gps and, more specifically on 3790 modified? Does not seem to me to be very logical.

If they have different algorithms and had not required relation, which the usefulness and reliability of show accuracy, mere ornament? But it is easy to check, just enable the Trip Log (in Map Layers), let the gps in Off Road mode (Pedestrian) and see the level of accuracy in real time, at the time of use. With the original 3790 is all very accurate, but with modified I noticed some barbarities.

In previous posts, Kanupus published links that dealt of the inaccuracy of 3790. Ok, it seems indeed, but considering the quality of the hardware and the fact that Garmin main components (processor, chipset..) are widely used by other manufacturers, is very unlikely to be hardware problem or project (in this case there would be recall). Either way cases like this of imprecision also happened with other releases of Garmin, but have been fixed and has to do with the immaturity of the firmware. Today the accuracy of 3790, shown on the screen and in the practice of use is excellent and was the result of the improvement of calculation algorithms that take into consideration general and specific factors of each model (average internal temperature, etc.), as already mentioned in the initial posts in this topic.

Anyway, for who is doing such tests I suggest consider some conditions for the comparison be more reliable: first leave the gps off road mode, obviously because in normal mode (look on road) the tracklog will be inaccurate. According to leave the gps on Pedestrian mode, because only here (3490 sw) the recording of the tracklog occurs in 1 in 1 second, which equals conditions when compared with other gps. In Automobile mode the recording criteria are automatic involving time, distance, etc, and the criteria may not be exactly the same for different devices. Another important detail: the test has to be done in winding streets and irregular circuit, preferably of difficult reception areas (trees, buildings, etc.). Test on major avenues and open roads do not show significant inaccuracies.