I've been happy with it. I use it on the road and on trails, and it has been as accurate as anything (as far as I can tell). I often use it in conjunction with my Wahoo Bolt and the numbers are very similar. I will admit, I don't compare the two side-by-side but I also haven't had reason to.

---

---

Works 90% of the time, every time

That’s a ringing endorsement if I’ve ever heard one.

No problems.

---

Let food be thy medicine...

It’s much better than it used to be. My open water swims are now close to what everyone else in my group is getting on their watches.

What does 'dying' look like for a 735? IOW, what is the problem?

Battery life has declined significantly, pacing and GPS distance has gone from rock solid to inconsistent. Biggest problem is that it freezes up mid-run a couple times a month.

I would hope you can find a better deal than 600 for a 945. You looking at classifieds and other places? Get a fenix 6 for 150 off or a 5x for 300ish on amazon etc.

It is a great watch. I would not hesitate to make the move.

Have you tried a factory reset? That's unlikely to help with battery life, but I wonder if it could resolve some other problems. The GPS could be an antenna failure, but that would not impact the freezing (probably). Sounds more as if something is corrupted that might get corrected with a factory reset. That will lose all of your stored activities, courses, customizations to activity profiles, CIQ stuff... So, backup 1st, factory reboot and see if some of the problems resolve. If so, you could restore from the backup one piece at a time (b/c something stored in the backup might be the culprit and you don't want to restore the problem!).

I have a 920XT and it seems to 'accumulate' problems associated with uploading. A factory reset makes the WiFi and BT uploads quick and reliable, but over many months, uploads are sluggish and require multiple tries, sometimes the only successful uploads are through USB. Then another factory reset and everything works great again. Weird and also a PITA. Also, you might try Garmin. Your device is almost certainly out of warranty. Even if so, Garmin's customer service seems responsive in these sorts of cases, sometimes offering a refurb at a discount (or free). I see that the outlet store has the 735XT for $250 if you can live with teal. If it must be black, then $295. These are new, so I imagine that Garmin must still have some refurbs for warranty replacements...

EDIT: I just realized that you could buy the teal 735 and swap the band from your current 735 - the teal version has a black case with teal/blue strap. I also noted that the outlet store has the 935 as a refurb for $295. Maybe that's the way to go?

Similar to mine after 2.5 years. Battery life poor and takes longer and longer to find GPS. Mid-run, my tracking isn't too bad, but throws up some very slow splits, so seems to be missing a beat or two. I went for the Fenix 6, arriving this week.

---

My race site: https://racesandplaces.wixsite.com/racesandplaces

My only complaint is a garmin thing and maybe not a 945 thing. Running pace seems to be "smoothed" too much. There is no way I can run at a pace that only changes by 5 second intervals but when the over all run pace is calculated it to the second....not based on a factor of 5 like is displayed while running. Maybe there is a weird setting I'm missing but I haven't found it yet.

My 2₵: GPS does a terrible job of calculating real-time pace. The very best method is a foot pod. I do not think any of the Garmin's have an option to use the onboard accelerometer to calculate real-time pace instead of GPS. So, if you want real-time pace, get an external foot pod. The best of the foot pods for pace is the Stryd.

Edit: I kind of misread your post a little... I have not studied it, but my Garmins seemed to update more frequently than 5 seconds. I do not know the frequency, but it is well-aligned with my needs when I am doing intervals. Maybe that is a benefit of the foot pod thing as well.

Gotcha....thanks

I don't think that is *must* be so - laggy GPS pace is a consequence of low power GPS chipsets and firmware designed to boast 24 hr runtimes rather than 'instant' pace updates. I have a 920XT - the GPS pace is too slow to change to be useful in anything but endurance training. I have the Garmin footpod and it's much quicker to respond to pace changes than the GPS. BUT, I also have a FR305 (yes, I take it out of the museum case occasionally). Works great and the GPS pace is *way* more responsive than the 920XT. Of course, the 305 is a brick bracelet with a primitive mono screen and shorter runtime than the sleeker, color screen, long running 920XT. There is no reason that a modern sports watch can't have responsive GPS pace. But there is a price to pay for that and the marketing favors battery life and other features over responsive pace.

It has nothing to do with battery. It is purely the technical limitations of GPS absolute position precision, frame update frequency, and 1 second sampling timeframe. Now, the Apple is taking occasional GPS readings to conserve battery, but I have never heard of the Garmin devices being accused of similar offenses.

I had a 305XT too. I do not remember it being anything special and any better than any other devices with instantaneous pace. But, it died a natural death years ago, so I cannot drag it out to test. I do know that my Stryd is freakin’ great compared to GPS on my last three Garmins (910XT, 735XT, and 945).

Not sure how the 305 pulled it off, but I've worn it and the 920XT on opposite wrists. Night and day difference in pace responsiveness following a change in pace. I understand that GPS can be limited by sampling frequency. Perhaps the 305 sampled more frequently than 1 Hz and only reported data once per sec? That costs energy, reduces run time (all else equal). I still think (but don't know - not trained in this field) that a GPS could sample more frequently / process data from more satellites and get a more precise location more quicky than a design that sampled less often and, say, only computed location from the strongest 4 satellites (instead of all that you can fix). With the 920XT, it can take 30 sec or more to get a reliable pace update (that could just be *my* 920XT - it's had some troubles - but this is the same as the day it was unboxed). That seems too long if it were operating close to 'optimal' wrt GPS precision.

While thinking about this, I found a reference that says: The government provides the GPS signal in space with a global average user range rate error (URRE) of ≤0.006 m/sec over any 3-second interval, with 95% probability. While I'm sure that this is simplied and, maybe, implies some steady state velocity, I'm not getting anything like this level of responsiveness withthe 920XT...

Folks with deeper and more articulate GPS understanding should chime in. The challenges are delivering reasonable accuracy around 10 feet/second velocity with absolute position accuracy around 15 feet while subframes only update every 30 seconds. At low speeds, the absolute accuracy injects much noise.

Agreed. And, we've drifted away from the OP's question...

I have problems with Garmin watches near my home. Observations:
1) It is NOT better with Suunto- I was told they use the same satellites.
2) My watches take forever to say they are ready to start recording- I assume they are not getting a full range of good satellite reception.
3) The shape of the course that I run effects the accuracy of the watch- with half mile loops being most accurate, and straight lines being the least accurate. They don't seem to handle elevation properly either.
4) The routes do not map correctly- it shows me running over rooftops, up the center of freeways, in rivers.

Explanation:
1) The watch does not have sufficient reception for successful satellite triangulation- causing shapes to warp and hills to be smoothed away.

Comment:
It seems like strava or trainingpeaks could correct the data. But they treat incorrect Garmin data like the word of God.

GPS can also use the doppler shift of the signal to determine speed, which I believe is a more accurate measure. Of course it's all coming from the same signal, so reflections or obstructions I presume would affect both the frequency shift and position over time. http://gpsinformation.net/main/gpsspeed.htm. This reference says Garmin specifies 0.1 mph accuracy in automotive settings, but in testing shows 0.5mph. My own 910xt is usually jumping around between +/- 30 seconds over my actual pace per mile. Very annoying when I have a target pace I want to hold.