any news on altimeter bugs, Garmin?

Il colmo... dopo qualche giorno tranquillo con v. 16.0, e mai connesso l'orologio al PC, mi si é appena aggiornato alla 17.50! É in incubo!!

Ciao Luca

Anche a te la 16 migliorava la situazione?

Sí, anche se io non ho mai avuto sbalzi giganti come altri utenti indicano. Quello che a me dava fastidio é che sempre esco con mia moglie in bici, e il suo fourunner é molto piú stabile e costante nei reisultati. Peró la 17.50, litteralmente fa piangere, la barreria dura una quarta parte del normale.

The barometer problem is a physical one with the barometer sensor port. They keep trying to provide software workarounds but ultimately the best fix is to buy or 3d print your own barometer cover. Search for shapeways barometer cover for an example of this.

If it were physical it would never have worked. In versions pre-16.00 the elevation gain accuracy was very good for me, post it was not. There may have incidents of clogged up sensor ports but it is not a universal problem. 

The barometer sensor is always functional, that's not the issue. The issue is with it's inconsistent accuracy. Try this experiment yourself. Run up and down a mountain on the same trail both ways. What you'll notice is some elevation errors. What we expect is a nice mirror image of the mountain's elevation. And we expect the height of the peak to be consistent across multiple days.

I tried everything and nothing really worked until I 3d printed a physical fix. I'm a gen 1 owner so this is across multiple software updates since initial release. After installing the physical fix, my elevation graphs have been perfectly identical with crisp detail uphill and downhill, across multiple days. And this is during mountain biking when there's wind, dirt and sweat.

I don't doubt that this worked for you, but there is no reason there can't be multiple issues. Myself and many others have had fine results with good accuracy before 16.00. Those results deteriorated with more recent updates. You can't just dismiss that. Also, I'm a trail runner, so velocity isn't really a factor anyway. bandit_74 has done a pretty exhaustive analysis, about as much as you can expect without having access to the source code; did you follow that?

I agree. There are multiple issues. 

1. The design of the barometer port is not good, which leads to inaccurate barometer readings. Fixing the physical design is expensive for them.
2. They decided to compensate for the inaccurate barometer readings with software fixes. 

Without any physical fixes, it relies 100% on the software fixes, which means tweaks to the software in the updates can influence the accuracy (even if the tweaks were designed to help).

I am not trying to change your mind, I just want to share my datapoint that with a physical fix using a cheap 3d-printed part, the barometer accuracy problem is solved completely. My elevation graphs have been rock solid for years since installation. The shape of the mountain's elevation going up and down is exactly mirrored. The height of the mountain peak is identical across multiple days. Neither of these was the case when I relied purely on their software fixes. Also, I do not see any deterioration with the more recent updates. 

I noticed the spikes in elevation this week. There was high humidity this week (comparing to previous week) and spikes have gotten worse (20X). At the beginning of activity there were no spikes and they appeared later. I think it's because the wrist was wet with sweat. A simple solution was to wipe the wrist. Waiting for a firmware fix (guessing it is doable because it was mitigated in older versions)

same for me, 18.01 did not fix the issue & agree an all of your comment.

if no solution soon i will give back the watch still i warranty

if no solution soon..i'l go for an "hardware" option...purchase a  sunto