pace target data screen showing not actual pace

I found this in the 945 forum (same issue):

forums.garmin.com/.../bug-in-the-interval-workout-screen-with-pace-as-target-the-watch-doesn-t-use-instant-pace

I'm on a HM training plan from a coach and noticed the same. i guess it's intened to be used when you are able to keep that pace and when you don't deviate to much from it. 

On the longer paced workouts i needed a pace between 5.45 and 6.00 but as i was running with others (as always) we went to fast (well at least for the coach ).
After ignoring the warnings at the start the pace indicator indeed showed 5.30 but after slowing down it still showed the avg (fast) pace instead of the 6.00 pace i was running. Checked with current pace in different data screen. i like the gauge/meter approach and ease of use and it will work when you keep to the wanted pace from the start but if you don't it would be nice to have a rolling pace there indicating the current vs. avg pace.

Garmin coaching will calculate your average over a longer time than the data field showing current pace (which calculates the average over a handful of seconds).

The idea of having the coach calculate the average over a longer time is to accommodate and smooth out brief changes in pace which can come from say, hilly terrain. You don't want runners entering Z5 hr when desperately trying to belt up a short 20% incline on a 5:30m/km easy run because the average has suddenly dropped to 8:00m/km.

The only downside to this is if you get your pacing wrong in the beginning, it sticks with you for a good part of the training session. I'll often go out at a pace 30 seconds faster than goal pace, only to drop back down to being 30 slower than the pace in at attempt to bring it back. To counter this, I've started taking note of the coach's suggested pace and I'll then switch over to watching instantaneous pace if I've really messed up the pacing in the session. This avoids the massive compensation needed to bring the overall average back.

I get what Garmin are trying to do and it's certainly not a bug. We're just having issues as we're lacking the discipline in the first part of the training session :)

I also don't think that it is a bug.

But it is not a good approach either. For me, it was totally wrong not only in the beginning, but also at the end of a 30min step. The Gauge/meter was showing me that I would be going 5:38, while the lower boundary was 5:17. As I wanted to push a bit more, I tried to speed up, but the gauge was still showing sth too slow, while I was actually going around 5:00. They could keep the step pace as a small number (e.g. replace the step distance for a time-based interval), but the big number/what is triggering the alarm needs to be the instant pace in my opinion. Otherwise you need to switch to another screen to see if you should be going slower or faster, which doesn't make sense.

Of course, if we would all be able to go perfectly at a constant pace, we wouldn't need that.  But then, honestly, we wouldn't need a watch at all. ;-) It's those things like the nice gauge that (should) make the training more fun and easy. But of course they need to show things that make sense.

JanH said:

I also don't think that it is a bug.

It is a design bug. Which is the absolutely worst kind of bugs, because the software behaves as the designers intended. Consequently, they will defend that behaviour to their death, no matter how wrong the behaviour is.

Implementation bugs are much easier. There you just need to make the developer understand that the software behaves differently than than the developer intended it to behave. Then he will want to correct it.

So why is it a bug? Because:
By only giving the user information of his average pace, you are asking his mental "pace control loop" to act like a PID control loop with the P and D part almost completely disabled. This is a sure way to make a control loop go into huge oscillations.

 yes, I agree, you can call it a design bug. I believe they had good intentions but didn't think until the end or used the function themselves... This seems common at garmin / to have happend e.g. also for the UltraTrack feature.

I'm not enough expert to approve your PID analogy though  but for sure my mental pace control loop got into huge oscillations looking only at the gauge.

AllanOlesen67 said:

By only giving the user information of his average pace, you are asking his mental "pace control loop" to act like a PID control loop with the P and D part almost completely disabled. This is a sure way to make a control loop go into huge oscillations.

I haven't done an interval workout in a bit (need to get back to that eventually), but i don't remember the reported pace and arrow on the gauge reporting the same value. Are they?

nmyeti said:

I haven't done an interval workout in a bit (need to get back to that eventually), but i don't remember the reported pace and arrow on the gauge reporting the same value. Are they?

As I remember it, there is some difference between pace and HR targets:
For one of them (and I don't remember if it is HR or pace), both the value and the arrow will be based on a lap average.
For the other, the value will be based on instant reading, and the arrow will be based on a lap average.

I have more or less given up on that screen because of these inconsistencies and the resulting difficulty of using the screen for maintaining constant intensity. And because of the lack of support for running power as a target. So instead I use my own screen layout with the data I want, which means that the targets for each workout step is only available as a notification at the start of the step. I can live with that, but it surely feels like a waste.

AllanOlesen67 said:

As I remember it, there is some difference between pace and HR targets:
For one of them (and I don't remember if it is HR or pace), both the value and the arrow will be based on a lap average.
For the other, the value will be based on instant reading, and the arrow will be based on a lap average.

I'm going to have to go out and test this tomorrow on my run. I *think* i remember HR being instant for both and i thought for pace it was lap for one and instant for the other. 

On the 5x plus frankly instant pace is pretty useless when the watch is the source and that makes this type of workout screen really difficult because it is really hard to hit targets (especially for short intervals), but on the Sony chip devices i have, instant pace is much more stable (maybe more averaged, but certainly filtered) so if it works the way i *think* it does, it might be more useful. I just need to get off my butt and decide to go get some anaerobic training load (which frankly is hard to find motivation for when i'm not actually training for anything).

nmyeti said:

On the 5x plus frankly instant pace is pretty useless when the watch is the source

Agree. But I don't use the watch as a pace source. I use a Stryd.