GPS Chipset 4.80 what`s new ?

+1

Change Log for GPS version 4.80:

Fixed bug to improve GPS reliability.
Fixed bug to improve time to first fix in certain conditions.

Thx Matthew

Garmin-Matthew

Thank you Matthew and also to the team behind this.

I did a short test ride with the new FW 6.25 and new GPS FW 4.80.

I can see small improvements in the GPS tracking accuracy in some places where it used to struggle, but still there are issues.

Passed 3 times at about 10 minutes distance on the bike path (2 meters wide).

One side is the park with a tree line and the other side a 3 lane boulevard with 4 story buildings on the other side of the boulevard. Not a challenging environment I would say.

Looking at the scale it's about 30 meters deviation.

Next is going on a bike path on a 4 lane boulevard with 7-10 story buildings each side.

GPS track totally erratic for the entire length of the boulevard:

(I wonder how the GPS track looks in New York for example between skyscrapers) 

I think it will never be good , i suppose the Sony chipset has terrible bad multipath error correction (and/or the Edge antenna/pcb design is bad and there is huge noise on the board and noise filtering is also weak), the older MTK chipset is outperforming in every case this Sony (i made tons of triple comparisons with a 3rd reference mobile with dual-band GPS chipset which highly outperforms every Garmin units...  )

it is just a small part of the yesterday ride #pic1 if there are no obstacles , the track is usually good, but when there are trees or house near to me, the track is pretty bad , not everytime but in very high percentage and even on flat and clear route also (pic #2, maybe did that few trees cause it? The Fenix3 usually within 1-2m comparing to my real position, with clear skyview, 100% sure, even if i pass the same spot 3 times during the ride , the Edge 830 is not really ) 

I'm afraid you are right that most likely is a hardware design issue (PCB/antenna/chipset), otherwise the would have fixed it via software in 1 year of attempts. 

I did similar tests with my Fenix 3 and it always had was more stable and clean track than 830. 

Garmin-Matthew

It's not that we need perfect GPS track to be recorded BUT this GPS deviation is greatly affecting the navigation in urban environment, when Edge "thinks" you are on the other side of a 4 lane boulevard and it will route and re-route accordingly.

In my opinion and use case, this (navigation) is one of the core features of the device and reasons for buying it. 

Having recalculation on is generally a bad idea with these units if you are following a pre planned route as when they recalculate they will create a new route to your destination point which is highly unlikely to be the same as one you planned.

I'm not referring to the scenario where I follow a planned route but when creating ad-hoc routes on the device while I'm moving around the city or finding my way to the start of a planned route. 

That's where bad routing happens due to GPS position being reported wrong by 30+ meters. 

My experience is very different. When I'm following a pre-planned route and I go off course my 830 prompts me to make a U-turn and re-join the route where I deviated. I have never had my device recalculate a new route.

I'm curious how your set-up your navigation options in your 830.