What about Gradient Lag?

Have 1040 running 14.20 SW.

lag delay have not been improved since I bought the Edge in Juni were SW was 12.x 

Delay is about 15-20 seconds. 
...my old 520 is spot on! 

Garmin: please focus on this! 

Did you sow any information that it was fixed?

Garmin just jump in the office and start fixing lag, with lag

good that winter is coming, till next summer 1050 will be released 

ps : I am sarcastic

Like many peoples, i found this lag is not acceptable for the latest and expensive garmin 1040. My old edge 1000 is better on this point.

I am quite disappointed by this decline of quality and the lack of reaction of garmin. I cannot recommand garmin product at this time

Garmin: please focus on this! 

So this debate is still ongoing. This is definitely a Garmin “Design Feature” and the debate has been ongoing for years with previous Edge models.

https://forums.garmin.com/sports-fitness/cycling/f/edge-1030/212208/grade-digits

This above post shows the removing of the decimal point was a conscious design decision. I think probably at this time they also flattened out the gradient as reported on page 6 by the Garmin consultant:

The current algorithm appears to use a 15 second rolling average to smooth the value, which leads to the lag that has been reported.

I do not think Garmin will change this as they have taken no notice for all these years of complaining is this and other posts.

it s hard to not notice, but very easy to ignore it 

No, that is not correct, it's NOT using an average, certainly not explaining up to 15 seconds. I have tested a lot this week and I'm sure about that. And in that you quoted it's "appears to use a 15 second rolling average to smooth". I think it's just a guess.

Garmin's official explanation is "This lag is a result of the amount of data inputs being processed"
https://support.garmin.com/sv-SE/?faq=kZjqTdE5ks374UVhZNHmAA
not some average.

I have a short tour I have tested some times now, one short dale(english?), down 3m and up again in 150m total, and one up and down where the ascent is appr. 100m long.
2 times when I have tested this the altitude is lagging appr. 15 secs (and the gradient some second(s) more) - on the way out. When going back again there is NO lag! The "funny" thing is that in the short "dale" the last time I did the test the saved data is almost 1 second ahead of the real altitude on the way back. I noticed that also when riding, it was quicker than my Edge 800 which I had to compare with, both altitude and gradient. I repeat no lag when going back on the same way. That is no lag for altitude, and some second(s) for the gradient, as it must be.

1. How can the altitude be lagging with up to 20 seconds? (both saved and displayed, I wrote something else incorrectly in some post before)
   You have also noticed the altitude lag in page 3.
    
2. How can altitude sometimes be lagging and sometimes not, in the same place depending on which direction?
   It was the same result in two tests on different days using the same way.
    
3. It's clear to me that it's the altitude that causes the problem, the gradient was in every test lagging a few second(s) more than the altitude.
    
4. If it have been some average it should not show the maximum gradient for one short time (only some second), as I have seen in a quick descent less than 10 seconds (showing it some 15 secs later because of the lag).

BTW, today in a different route I had at least 20 seconds lag of altitude (and gradient some more) in both directions.

Perhaps it's some heavy calculating comparing with GPS and/or saved data somewhere. One thing I want to know, was the gradient lag introduced at the same time as Climb PRO or something comparable with that?

i would rather have a usable gradient than climb pro. This is too much form over function. 

Exactly

244,

If you look back through the current 27 pages of this post, you will see that I have performed numerous tests.

The 1000 was pretty Ok in the gradient respect. No sensor can be 100% accurate, but it gave you a reasonable idea of where you are on a climb. The processor in the 1040 is supposed to be an upgrade (one would hope so as it’s 8 years later).

I worked in computers and your comment “heavy calculating” shows that you did not. A 20 year old chip could process the data required to calculate the Gradient in milliseconds. One would hope the 1040 chip could do this in nanoseconds. 

There is very little data in the .FIT as regards position, entries are in the seconds, there are not 1000’s of entries for a 50M section of climb.

This is a design feature as per the Garmin Consultant, whether it’s 15sec, a few less or a few more is irrelevant, this is the way they are displaying the Gradient and almost no one agrees with them.

Sorry, but the use of the term “a design feature” is an OXYMORON, regarding this NEW, super duper, upgraded, nano second-reacting chip.

That is - “design featured” by Garmin engineers and marketers, to react in 15 seconds, to give an approximate gradient reading that is more or less as reliable, as the 8-10 yr old Edge 1000, which reacted in 2-3 seconds!

Sheesh…