235 Memory Leack

I can imagine some of these garmin apps/widgets installing interrupt/device handlers, which take away time from the HRM measuring routines.

I have to say I somewhat agree with the original post.

I have numerous issues with sluggish responsiveness flicking between screens, my workout countdown beeps are erratic and I often get cadence lock. Interestingly when I get a cadence lock I remove the device from my wrist compeltely and run for half a mile and it still reports a heart rate of 170+!

Anyway, my point is, I've accepted its an issue and I have an RMA for a replacement as it's clear myself and the OP are in the minority. It's most likely a faulty unit rather than a bug.

I'd suggest going through the support channels and doing the same.

I can imagine some of these garmin apps/widgets installing interrupt/device handlers, which take away time from the HRM measuring routines.

You can imagine a lot of things, but that isn't even something that is possible through the SDK.

You can imagine a lot of things, but that isn't even something that is possible through the SDK.

i can't understand this kind of "engineer-security" (sorry if i'm using some wrong terms)
If someone tells there are some problems and these problems are not related on the device because these problems were not seen with previous firmwares, and moreover these problems seems to disappear just uninstalling apps.... why you can't imagine it is possible come software or piece of code not so well defined?

You talk about installing/uninstalling apps and activating/deactivating widget pages.
It does not appear that you were actually using these - e.g. did you pair the VIRB and were running with it?
Otherwise installing an app you are not using is basically just putting a file onto your device.

I also note that the example activities you mention appear to be interval training. It is a common issue that optical HR devices generally perform less well in such situations.

The only time a CIQ app/widget/wf is running is bretty much when you can see it running on the watch. The DF's will continue to run if you switch to a different data screen while the run app (for example) is running.

Other than that, they are just siting in your mass storage space and using some space there. (with limits on how many there can be (16) and how much total space they take (varies by device))

CIQ itself is VERY restricted in what it can do. Basically it can only "look at" data the FW provides, with the only real exception I can think of, is when a CIQ app causes a .fit file to be recorded (and there it really can only say "start", "stop", "save" and "discard" that recording).

... why you can't imagine it is possible...?

Easy, because it's not possible.

As am I.

A plain memory leak would simply run the watch out of memory to do anything. The HR wouldn't become spotty, it would become non-existent and the whole system would likely crash. For the other scenario that some people refer to as a "leak", assuming that it could leak bad data into the HR section, your HR wouldn't be reading 220s instead of 170s... it would be reading random garbage numbers with little to no correlation to each other - the HR chart would either be a straight line on a random bad number from one recurring overwrite, or it would jump all over the place from repeated overwrites.

You could also search the ConnectIQ Programmer section of the forums and see that the only issues relating to memory leaks are in the GarminConnect Android app, not the watch's ConnectIQ app/widget/watchface system.

I have had zero performance difference from any # of apps, watchfaces, or widgets installed. You would need many tests containing repeated totally trashed HR data as described above - a single bad run or cadence lock or whatever else doesn't mean squat given that no OHRM 100% accurate to begin with.

To answer your original complaints -
Delay between screens is because the watch is not constantly sitting there waiting for your input. It is conserving battery 99.9% of the time, and has to wake itself up from so-called "low power mode" to respond to your button press the first time. Even after that, every screen you switch to requires it to swap in a whole new application - and like GRZLDVT says, this isn't running a quadcore exynos processor with 1GB RAM to hold all the apps constantly in memory for instant results, it's a low-battery-usage watch designed to maximize battery life.

The transfer data warning is delayed intentionally to conserve battery.

One bad run with 3 apps vs One good run with 1 app means just about nothing. Come back when you've repeated multiple test runs of the same intensity on the same path with the same HR Firmware installed with 3 apps vs 1 app (because lots of things can affect the HR accuracy a hell of a lot more than how many apps are installed). I have multiple apps installed and there has been no effect on the HR accuracy - what has had an effect is the Firmware updates over the last month in which Garmin actually changes the function on the OHRM.

Great answer