what evidence have you seen that the 220/620/fenix2 etc are less accurate to any substantive degree?

I've had a 220 for a while now. It definitely stops me at the same spot each time in my 2 and 3 mile runs.

Of course it could be consistently wrong =)

---

---

Kat Hunter reports on the San Dimas Stage Race from inside the GC winning team
Aeroweenie.com -Compendium of Aero Data and Knowledge
Freelance sports & outdoors writer Kathryn Hunter

Same 2.5 mile OWS that was measured independently.

• My Fenix2 measured it as 3, 3.5, and 4 miles on three different occasions.
• My 910XT meaured it exactly 2.5 miles twice.

The Fenix2 has many problems, just one of which is poor accuracy in OWS measurement.

Would be interesting to plot the data from each, might lend a clue as to the source of the error

---

---

Kat Hunter reports on the San Dimas Stage Race from inside the GC winning team
Aeroweenie.com -Compendium of Aero Data and Knowledge
Freelance sports & outdoors writer Kathryn Hunter

Precise but not accurate . . .weird the crap you remember.

---

Ironman Certified Coach

Currently accepting limited number of new athletes

Of course it would be odd to have a consistent but inaccurate GPS meter in the same way it would be odd for a power meter =)

---

---

Kat Hunter reports on the San Dimas Stage Race from inside the GC winning team
Aeroweenie.com -Compendium of Aero Data and Knowledge
Freelance sports & outdoors writer Kathryn Hunter

All swims were done with the watches on my wrist. Based on vibration feedback that notifies when the GPS signal is lost, the Fenix loses the GPS more frequently. I assume the algorythm for filling in the missing blanks is better on the 910. But, the 910 loses the signal less frequently as well.

The Fenix is so bad, it showed me swimming deep on land at times.

It also made a mess of my Challenge AC swim, missing an entire section (maybe a quarter to half a mile) of the course.

This, on top of it getting water inside on two different watches, and I sent it back twice and have been using my wife's 910 while deciding what to replace it with.

Depends on the source of the error - a precise but not accurate unit would likely be getting the correct input from the satellite and muck up the conversion in the software, whereas a unit that's all over the place is likely calculating it's relative position incorrectly.

Same here, my 220 alerts me for splits so consistently on my regular routes that I know when it's going to vibrate within a 5-10 foot window. KM 1 is as I run by a speed limit sign, #2 is as I go by a certain tree, #3 is in the last 10 feet of a parking lot. Before that I had a 305 and it alerted me in the same exact spots. I've not used a 310/920 to compare.

I always wonder about this with these GPS units.

For instance, I use my 910 for every run I ever do, easy, all out, temp, etc. I have it set to give me a 0.25 mile split constantly. There are a few odd ball things that very consistently happen.

First, there is a place in my neighborhood, where no matter how fast I am going, how I am feeling, that quarter mile is always way slower than the one before it and the one after it. It is like the signal is blocked and it thinks I run a shorter distance over the same time period in that one spot.

Also, when the thing is first getting going (even if it has 20 feet accuracy noted or less), it will always predict that my first quarter heading north is longer than the same distance and same location heading back south.

I only point all this out to say, there are some oddly consistent inconsistencies (:-) inside that 910. All in all, I wouldn't ever trade it. Love the watch for swim/bike/run for sure.

---

Ironman Certified Coach

Currently accepting limited number of new athletes

It looks like the problem isn't the accuracy of the different units, but the quality and this really only presents itself in OWS. when running, the units should always have a decent GPS signal and should all have the same accuracy. But when OWS or perhaps trial running or urban canyon running the units will lose the signal. the ones that maintain the signal longer and pick it back earlier will give much better results. For those conditions, the algorithms are also critical in generating accurate tracks and weeding out the bad data.

Yeah I can see how water would present extra challenges, and maybe also the movement of the arms.

---

---

Kat Hunter reports on the San Dimas Stage Race from inside the GC winning team
Aeroweenie.com -Compendium of Aero Data and Knowledge
Freelance sports & outdoors writer Kathryn Hunter

Also could vary depending on stroke rate? Higher rate = less time to get a signal fix? Then again it could be the reverse that the faster turn over of arms leads to less loss of a signal. Surely there is a TTL (Time To Live) on the signal for these watches before it considered the signal lost?

---

------
"Train so you have no regrets @ the finish line"

Funny that you mention this. As I've been doing more speedwork, I've gotten the feeling that the 220 just isn't as consistent as my 610 and I'm thinking about switching back. It's too bad, because the 220 is a MUCH more comfortable watch. The issue I'm having is whether or not the 220 is "good enough." I mean, when I run in hilly canyons, my pace is not that much of a concern to me. It's on the open road with a clear view of the sky - most of my races and speedwork runs - that I value accuracy the most, and these are also the situations where it's most easy to be accurate.

But yesterday, doing a long tempo run on tired legs, when my RPE:speed is least accurate, I felt like I wasn't getting an accurate picture of my run. Like my RPE:speed is whacked, but the watch was even more wacked.

In general, I've felt like the 220 is "good enough," and the comfort is worth a lot over the 610. But on some of my runs, it's becoming more of a concern as I feel like I'm really not getting an honest picture.

Honestly, I thought I was crazy. I had seen some grumblings about the 220/620 accuracy, but that was before a lot of the firmware updates. Had no idea that had stopped working with SiRF. That is a massive disappointment.

But I definitely felt like I was "that guy." I mean, who blames their GPS? But maybe I actually wasn't so crazy...

I've got another tempo tomorrow. Maybe time to throw on both watches and see what I get.

---

"Non est ad astra mollis e terris via." - Seneca | rappstar.com | FB - Rappstar Racing | IG - @jordanrapp

There are other head ot head comparisons on the garmin fenix2 forum on their own site.

It's a well known phenomenon.

My own experience is that the fenix2 and 910 give very different results on the same swim. The fenix2 jumps so much that it gives extra distance to every swim. The reading jumping forward and back along a path won't change much since it will average out. But the side to side jumps perpendicular to the true path add distance.

I've put nearly 500 miles on my FR620 this year & I haven't had any of the GPS issues that some have claimed. All of my tracks have been spot on. The only issues I have had are the widespread ones when new software was released and occasionally the watch will take longer to lock onto the satellites, but never an issue with accuracy. I hit my mile marks on consistent tracks in the same spot every time.

I believe that there are two possibilities; either there is a QA issue at the point of manufacture & they are letting bad units out in massive numbers alongside the good units & the customer is getting the luck of the draw, or people's negative opinions are being fueled by posts like this one and the review most often quoted by fellrnr that calls Garmin's GPS out.

Garmin recently reduced the price of the fenix2, REI is simply reflecting that change in price. The question to ask is why Garmin reduced the price, is it to move product at the end of the season? Move a product that is perceived to have a lot of issues? Prepare for a 910 replacement?

I have both the 910xt and Fenix2. For running they seem to agree on the distance within a small variance (under 0.1% difference). Swimming is a different matter, the 910xt seems more consistent than the fenix2 with the fenix more likely to over state the distance or report odd swimming behavior (doubling back for 50 yards or swimming over land at 5mph), I think this is more of a software issue but that's just a guess on my part. There are times with both units when it records crazy data, running over buildings, moving at 30+ mph, swimming over land, etc..

I think DC rainmaker reports accurately based on his experiences, but he is a small sample size which means he won't see every problem. Some of his reviews include problems he experienced. Garmin ships a lot of units, some of them are going to be bad even if the defect rate was 1 in 1000 there would still be many posts on the internet complaining about them. My n=1, I'm on my third 910xt (in two years) having had a variety of issues a quick search shows many people have also had to have repairs or replacements with the 910xt, I've had the fenix2 since march and its running strong.

The team that handles the fenix2 is much better about software updates to resolve issues than the team handling the 910xt.

Well he's going to miss a few things.

The biggest mistake I think I have ever seen him make was missing that the speeds recorded by the timex global trainer were off by about 25%.

You can actually see it here, http://home.trainingpeaks.com/...JQGGJJAG2HVRC6FSI3FY. That's the file of a ride he did when he was trying out the watch.

The total time and distance make sense, an average of 24 miles per hour. But look at the speeds in the graph, the green line, according to that line he hardly ever got over 24 mph and spent most of his time at 18 mph. If you were able to highlight a section, like trainingpeaks allows premium users to do, the speeds for that section would be off as well.

As far as I know, that was never ever fixed on that watch incidentally.

Even for an extensive test, there are things he is going to miss, Ray seems like a good dude and does the best reviews out there; but he's not infallible :-)

The issues with the fenix are subtle enough that I can easily see where they would go unnoticed for the two or three weeks he used it.

GPS Accuracy tests

Also gives me another reason to switch more fully over to the V800. Just need to get a dual band HR strap...

---

"Non est ad astra mollis e terris via." - Seneca | rappstar.com | FB - Rappstar Racing | IG - @jordanrapp

My 220 is MORE consistent on runs than the people that I run with that have 910 watches. Theirs will be at least .05 mile +/- over a mile. My 220 is within feet of accurate on a measured course. We all use footpods.

I think the footpod is an important factor.

That FellRnr.com test was done before the 220/620 had footpod compatibility. Now that you can use the footpod, I'd expect it'd add accuracy, just as it seems to with every other watch.

---

"Non est ad astra mollis e terris via." - Seneca | rappstar.com | FB - Rappstar Racing | IG - @jordanrapp

A few random things.

1) Hi.
2) I publish the vast majority of my bikes and runs to Strava, you're welcome to poke around.
3) I have to base the review on things I see and reproduce. If I forum surfed for every possible issue and then posted them all in a post, it'd be silly. It's 'my review', not 'the entire Internet's review'. Think if a reviewer did that for the iPhone?
4) Oh, I do forum surf though. I read virtually every Garmin Forum (and ST) thread on recent products, and answer in many of them. I often push to have issues that I can repro fixed.
5) As noted above, you're going to see issues on any unit when you apply enough volume. 1% of 100K units is still 1,000 people. 1% would be a very remarkable consumer electronics device failure rate.
6) Try typing "GPS accuracy" and putting the GPS product name after it.
7) While I usually have units supplied for companies early on in my reviews (simply because they aren't available), I actually often switch to retailer shipped products before the review is out.
8) I don't believe the plot seen above is as useful as folks are making it out to be. It's on a trail run, where the data isn't being provided, and the thousands of 'attempts' are really just runs divided up into tons of tiny little sections. Comparing foot pod to non-foot pod is sorta cheating. After all, that's the point of foot pods on trail runs.
9) As for the sales you see, I explained the reasons pretty clearly a week or so ago in a post when the sale came out.
10) That's probably about it.

---

-
My tiny little slice of the internets: dcrainmaker.com

You're the man.

Enjoy EB!

x2

---

2024: Bevoman, Galveston, Alcatraz, Marble Falls, Santa Cruz