Grade Calc - Community Project

So you're trying to do everything that the edge is already doing? How do you "calibrate" the elevation? Just accepting what Garmin gives you at the start of the DF? Or it doesn't matter because this POC is for grade calculation, not for continous elevation calculation based on some algorithm that is maybe better than Garmin's?
Also wouldn't it be interesting to display both Garmin's elevation and grade and your calculation?

jim_m_58 said:

remember, the pressure can change if you stand in the same place long enough.  Say there is a front moving in

Not going to affect grade much (not really at all).

jim_m_58 said:

Counting floors is also done with the baro, and I've seen "floors" if I'm in front of a vent when the AC kicks on.

How do you manage to get anything done? This is a fairly pathological example. It's a barometer. It has limitations. Nothing is perfect. You know that GPS doesn’t really work indoors, right?

People would be happy to get better values for regular, normal situations. They aren't going to care about odd/rare situations.

Toybox.Activity.Info.altitude 

The altitude above mean sea level in meters (m).

Elevation is derived from the most accurate source: Barometer or GPS

So if you have good gps you should have good altitude, planes based on gps data so I think it's enough to bicycle ride.

perfect is the enemy of good. it’s easy to shut down any criticism of the status quo, any attempts to improve things, any feature requests and/or any debatable bug reports by saying the alternative isn’t perfect. it’s also a fun debating tool if you like being a contrarian for its own sake.

I see this a lot when ppl criticize garmin products in general. “you know apple/suunto/polar aren’t perfect, right!!!1!”

"Elevation is derived from the most accurate source: Barometer or GPS"

I think this is a key part of the problem.  How does garmin determine which source IS the most accurate at any point in time?  Is it just if there is no GPS signal that it uses Barometric?  

I'm keen to see the results from Dave's test DF to compare baro readings (and derived alt calculation) to the garmin "altitude" reading…  (I'm not riding at the moment otherwise I'd test it too).

In my own tests of a device with a baro altimeter and one without at the same time, what I see on watches themselves, GPS is far less accurate,  When you upload an activity to Garmin Connect, if there isn't a baro, GC by default applies elevation correction making the data look much better. 

GPS was designed more for horizontal measurements. Also, being off by 1000 ft horizontally is no big deal while being off by 1000 ft is something people start caring about.

Barometers are much more sensitive and better-suited for grade (which is being measured over short distances).

For elevation, Garmin considers barometric data as higher quality than GPS data.

GPS is used when a barometer isn’t available. 

Garmin Connect turns off the elevation correction if the device has a barometer.

_psx_ said:

So if you have good gps you should have good altitude, planes based on gps data so I think it's enough to bicycle ride.

Not the same thing. 

Airplanes are traveling much, much faster than bicycles (which could be as slow as 3 mph going uphill). Being faster means getting better data (the signal to noise ratio is much better).

Pilots only care about absolute elevation. Cyclists care about changes in elevation. Also, airplanes will generally have better access to more satellites (giving better accuracy).

I don't think planes use GPS for altitude measurement. Maybe if both the other sensors are dead.

Firstly they use barometer + frequent reports from nearest airport's weather station to calibrate the airports's known altitude above the see with the measured ambient pressure => from the difference they know their altitude over the airport.

Secondly, mainly during landing, they use radio altimeter, where a device emits radio signals downward and measure the reflection. This is very precise over land, not so much over water.

While airplanes do care about grade, mostly to fit the appropriate glide slope on landing, they measure it in feet per minute, and though they usually descend with a rate of a few hundred feet per minute to 1500 ft/min (hardly more) they also travel lot horizontally in a minute, even when landing (slowest).

This thread in the forum is not about elevation, it's about grade. I don't think the Garmin grades are off by value (I have no way to know, but never heard about that problem), the problem is that when grade is changing it takes lot of time to adjust. Often I reached the top of a hill, started a downhill, I look at it and it's still showing some positive grade.