Really dumb question. When you own an Element, what do you download the rides to in terms of a program? I am so used to Garmin Connect when directly transfers to Training Peaks

On their app or online you link your accounts, for example i am linked to training peaks and trainerroad. Then whenever you finish a ride it will auto upload if you connect to wifi or through your phone. It is really supper fast and my favorite part about Wahoo computers. They also have the ability to view the workout in their app but it is very basic. If you are concerned about cost there is always golden cheetah which is free that it can be sent to as well.

Comprehensive article on things I haven't run into...Although I think you may be onto something in general with it.

SUUNTO, I had a SUUNTO at the same time I had a Foretrex 401. Granted this was 6 years ago and the GPS world is much different. But SUUNTO's GPS was awful.

---

Washed up footy player turned Triathlete.

But when something has a feature that can’t be used on release. And when working it doesn’t work well that should be commented on. Which Ray did well. And because those things detract from the unit the price seemed to high.

For me it’s an interesting one. I’ve owned:

910
510
920
520
Varia
F5
F5X+

And for the vast majority of time my devices just worked. Did they need resetting on the very odd occasion? Yep. They did. Same
Can be said about my cell phone so for me I don’t really worry. My F5X+ is great. Having Spotify on my runs is just awesome. Love it. Will be getting a 530 shortly.

A good mate of mine however, has the opposite. He seems to be able to kill pretty much every Garmin device he has. He hates it. But is kinda locked in to their ecosystem at the moment.

I have no idea why our experiences are so different but they are. So for now, whilst I read about these issues, to date I’ve had good experiences with Garmin and will continue to use them.

have you ever manufactured anything? if the answer is no, not a problem. most people don't. there's something else you do well, that i do poorly.

having been a manufacturer, i have a little more sympathy for this business class. being a manufacturer is rewarding in that it's like being a novelist: you're creating something out of virtually nothing. you're taking amorphous, anodyne stuff and assembling, via reverse entropy, a complex, useful tool.

because i was a manufacturer, i recognize something and i write it over and over in my product reviews: you have to do the important stuff well. first. then we can talk about the sexy stuff. your thing has to go when you want it to go; stop when you say stop. it shouldn't hurt you. it should fit. then we can talk about your mobius speed ridges or the hyperenzyme powerfastener.

the theme of this thread is that garmin fails, too often, to major in the majors. my theme is that it's not fair to, when writing a review, major in the minors, flogging a brand for something less consequential, virtually ignoring its competence in the major stuff.

look, i hear you. but were you going to buy a ROAM the day it came out? why not say something like: this product does all these things well. it is worth the asking price, just, only commensurate with the features promised. the ROAM is a $379 product if and when the promised features arrive. the discriminating consumer may want to wait until those features arrive and then invest in a ROAM. thankfully, wahoo is in an industry where running changes arrive over your wifi.

how would you feel if wahoo, ironman, hoka, cervelo (fill in the blank) were gone tomorrow? how would i feel? i ask myself these questions before every single product review or brand critique i write. i don't give them a pass. i'm not afraid of these brands or the people who run them. i'm probably in my last 5 years of active work in this industry. i don't need anyone's permission or money or approval. i'm happy to tell anyone, to his face, or i'll tell it to his audience, how i feel. but i've also seen the entire arc of this sport. i try to write in a way that places a discrete product inside of the bigger picture, capturing what might be a more comprehensive truth. i don't mean to tell anyone else how to write. that's how i write.

---

Dan Empfield
aka Slowman

Dunno about baydad...but, yes I have and still do. I've been designing embedded hardware/software products since the early 90s for vet med, aerospace avionics, industrial process control, and consumer electronics...at 5 different companies, and countless leadership regimes (and styles). I've designed, built, manufactured, and provided support for products with unit sales of 6 to 6 million (give or take a couple hundred thousand), and support lives from 5 years to 50. I've started, led, and participated in software teams from 5 to 100 engineers designing software from 500,000 lines of code to 5 million....from simple digital thermometers, to entire military aircraft.

So, I have more than a passing knowlege for what goes wrong (or right)...specifically, in the embedded software product arena....which is quite a bit different than what happens in the mechanical systems arena. I've seen things go pear-shaped in so many different ways that I really hesitate to comment.

Generally, software quality issues (which I think these more "rightly" are vs. reliability), are NOT a result of lowly software engineers. Don't get me wrong....they CAN be. But, it is more typical that the teams are handcuffed in some way by process, culture, resources, staffing or some combination of these. It is more typically a Senior Leadership issue that prevents the engineering teams from being able to do what they "know" is right. Often this is from a heavy handed, micro-management based culture.

Again, that's just speculation based on my range of experience. I used to know a couple of engineers at Garmin (I went to school is Kansas), but I haven't spoken with them in ages, and don't know if they even still work in Olathe.

They both should be flogged. The entire industry should be flogged. The trend is in the wrong direction, across the board. QUALITY is getting worse, reliability is getting worse. The "underhanded" used to release beta, now they release alpha (or pre-alpha). Defect reletivism shouldn't be a selling point (our defects aren't as bad as theirs).



I don't think its worth the asking price...in any "value based pricing" consideration. I don't think Garmin's are worth the price either.

I'm also not sure about my final highlight being a "selling point". It WAS, once-upon-a-time. But, it has become an industry crutch. "We don't have to get it right, because we can just publish field upgrades later." I've lived all this from the inside. I started when almost none of that even existed conceptually. And, I implemented some of the first systems to have those capabilities. And I've seen the management view that "we can upgrade later", be a driving force to "release as-is, and keep working on the fix".

I read your post several times and I 'm not sure what side of the issue you are on, whether you think Garmin (and others) are a problem, or you are suggesting that we give them a break.

I am in manufacturing - I'm the Plant Manager for one of a Fortune 1000 company's defense contractor divisions. We make complex mechanical devices that have to work every time, whether out of the box or after years of service, in very demanding environments where the failure of one of our components and can kill its users. We make no excuses for the quality of our product - they 100% meet every spec (or better) every time. They are manufactured under tight quality controls with internal independent quality control checking the processes and external government reps checking the product. We throw away annually what some middle size company's would consider a great gross sales year. It is part of doing business in our world.

I don't expect Garmin to operate at our standards because I won't pay the price they would have to charge to do what we do. That doesn't mean that they get to be lackadaisical in their customer support or manufacturing control. I understand that problems will occur, they will slip through to the shipped product, and they will not always get fixed in a timely manner. That does not excuse Live Track from not working for several years and Garmin not acknowledging the problem, let alone fixing it.

---

"...the street finds its own uses for things"

I'm in general agreement with your sentiment. Specifically with regard to support, acknowledging issues (even if not understood), and timeliness of repair for serious flaws. Simply acknowleging issues can go a long way towards retaining customer trust. THAT only goes so far, though...as you have to ultimately follow through on the promise to fix the issue. And that window is finite and rather short.

However, manufacturing quality is a fundamentally different beast from software quality. Software quality issues are VERY hard to fix in a released product. In any suitably large/complex software product, the number of undiscovered bugs is essentially "fixed" at the time of design and construction. From that point forward all you can really do, "move them around".

Generally, we are lucky to only create one bug (to be discovered at a later date, perhaps years into the future) for every bug we fix. I can decide what parts of the software I want to be "bug-free". But, all that really means is that I will create bugs in other places duing the debugging process. You can't test quality into a software product.

That seems to mirror what I saw in the DCR comments and some of this thread. I figure it has to be working for most/many people. Of course the angry ones are going to grumble online.

I sound more like your mate. While my early, early Garmin stuff worked well, I have had my last 4 devices replaced due to bugs. Vivoactive 1 stopped telling the time. Replaced and the battery one that went to crap awfully quick. Good deal on the Vivoactive HR, but the battery Fenix 5 battery life went to <1 day on standby after 5 months. It took a solid 6 weeks of tech support telling me reset and/or try new software before they replaced it with a refurb. The refurb has been solid the ~6 months I have had it though. I bought a 510 the month they came out. It would lose the occasional ride and a factory reset would get it to behave for a few more months. Occasionally the ant+ antenna would stop receiving, so a factory reset would get it working again. I just sort of lived with it for 4 years because it wasn't bad enough to jump ship.

The 520 Plus I got for Christmas has bluetooth that has never stayed connected beyond initial pairing (so I have to pair it before every ride) and the battery died on me at 7.5 hours with backlight off, bluetooth off, and all of their recommended battery saving suggestions except 1-second recording. Half of advertised battery life on a unit with maybe 10 rides on it. I'm in the middle of the RMA process for that right now, but not convinced a replacement 520 Plus will be any better. The 520 Plus re-routing is absolutely awful. All it does is put a pin on the spot you went off course and do its best to get you back to that pin. "U turn, U turn" is all the rerouting it does unless you accidentally get back on course. No audio notices for turns even though the previous non-navigation units would beep at you before a turn with a loaded course. I can hardly believe what a piece of garbage that 520 Plus is.

Part of me feels that the ant+ ecosystem is somewhat part of Garmin's marketshare. There really weren't any ant+ bike computer outside of Garmin until Wahoo mildly disrupted things. If you had a power meter, you had a Garmin bike computer. It's not hard to have a few thousand dollars wrapped up in power meters. Wahoo was the first chance to jump ship and still use your power meter a few years ago. Now, most power meters transmit bluetooth as well and it's less of an issue. It's kind of weird to have a different watch than your bike computer, so it's easy to get a Garmin. Wahoo was supposed to have their watch out May of 2018. Oops. Nobody else is using ant+. Suunto really shot themselves in the foot by not including ant+. Same with Polar. They make good products, but it feels like they sort of alienated people with foot pods, heart rate straps, or bike related accessories that ran on ant+.

I think I'm the only one that liked the old 410 and the goofy touch bezel. It was also the last Garmin I owned the 'just worked.'

I think you have to have followed the discussion about wahoo’s ROAM launch 6 weeks ago to understand what I’m writing. My comments largely refer back to the toilet flushed on wahoo’s head. The comments about Garmin here only make sense in the context of its competitors. Is wahoo an implicit winner in software and ecosystem? Is that the takeaway? If so, I noticed an absence of credit for that during the ROAM launch.

---

Dan Empfield
aka Slowman

I did then. I just went back and read it again. I still agree with the toilet flush on Wahoo's head, as I do with THIS one on Garmin. I can say the same thing about Apple...because I literally HATE my iPhone, there are so many stupid things about that device I don't even know where to start. I can say the same thing about Google...my Android P phone is only now starting to have a reasonably decent UX, and I STILL have to reboot the thing, or cycle airplane mode off/on to fix one stupid issue or another.

The consumer software industry as a whole has somewhat gone off the rails. The entire "iterative software design process" could use a toilet flush to some degree. Not because it is fundamentally flawed and incapable of creating quality products efficiently (it's not). But, rather because it creates an illusion of continual readiness in the minds of largely hardware oriented leadership minds. Further, it can easily shortchange "big design decisions" that can have long-lasting quality implications, for more simple designs that may get extended beyond their useful life...because no one wants to incur the redesign/refactor cost/schedule "right at this moment" (so called technical debt repayment).

you are a hard man to please.

---

Dan Empfield
aka Slowman

I dint quite follow that rationale.
In software development, when you follow the 'V' diagram, you precusely DO test to verify that the software does what it is supposed to do, with the inputs defined.
And when you change the code you should re-test.

So if someone is introducing new faults somewhere else, then the scope of the testing is not correct / wide enough.
Or the product is being release without the appropriate testing.
Or the product manager is shoving it out the door anyway irrespective of the finding of the tests.

I don't really think so... but, that has been said before. I'm a Product Manager, though. There is an aspect of that job that is setting the right expectation. My best teams have always seen where I set the bar and then raised it. Then *I* have to talk THEM down off the ledge as launch day approaches.

You married? :-)

---

Dan Empfield
aka Slowman

That's a question that is really hard to respond to in the context of a forum post. Today's software is not a linear set of operations that always occur in the same exact sequence. Designing for robustness, designing for test, test methodology are entire courses, and curriculums today. There is nothing simple about it.

In any non-trivial piece of software there are (for all practical purposes) and infinite set of test conditions....hundreds of quintillions of possibilities is a small number in this context (we would typically state them in exponent notation 10**xxx, or 2**yyyyy). If I have a really good automated test setup, I might be able to test a few hundred thousand. That's nothing, a tiny fraction of 1% of all cases. Further, I can only test those in a very specific and prescribed sequence. Inevitably, I will get exactly the same result every time. That's sort of exactly the point of these systems. However, in the real-world of real users, I'll get lots of different results because its timing, location, sequence of events, atmospheric condition, etc driven. All things I can't replicate in a controlled test configuration. In many cases I can't even anticipate them.

I can't tell you how many times I've heard a bug report that involved the most bizarre set of conditions and steps...that I was just left going "You did what? And what happened? Nuh-uh. How is that even possible?" Then I've spent months with 100s of machines setup to atuomatically reproduce the alleged sequence of events only to reproduce the "bug" approximately once a week, or once every 100,000 attempts. The bug was caused by a timing of events, that required 3 randomly occuring events to occur in the right sequence, with a very precise set of timing. When THAT happened, the software crashed. This particular software had been written to run on two processors. It was reused to run on 4, 6, 8, 16, or 32 processors. 99.999% of the time it worked without issue. ie, it made it through all of our testing pre-launch. But, when you ship that to 6,000,000 customers...guess what, it happens several times? Yes, that's a real story from ~2003.

Yes, we have test programs such as the V-model. But, the V-model (in particular) isn't very good at testing out-of-bounds conditions. its generally only good at verifying that the software does what it is supposed to. It says very little about if it does anything its NOT supposed to do.

To an angel for 28 years. yes, she probably questions that decision weekly, if not daily. So far, she hasn't changed her mind. That's her fault, not mine. :-)

I guess that I would also add that the V-model, when its used, is inherently a developmental testing model. In other words, its the model that gets you TO the initial release. However, it is generally NOT followed after that initial release. And THAT is the point where things begin to get trickier.

Again, the V-model isn't a great model for ensuring the negative doesn't happen. There are other approaches that are better. But, the laws of statistics still rule, and you can't prove the unanticipated consqeunce doesn't happen. Why? Because you didn't anticipate it.

God bless DC Rainmaker.

Never seen such a dominant brand with a product as wonky as Garmin's all tend to be. Still...love/hate them.

Using the menus on ANY garmin device is like going back to 1980. If you think like a Garmin engineer, you'll find that sub menu you are looking for buried in 5 different menus...but if you are a triathlete...good luck to you.

The thing I don't understand is Garmin music and trumpeting you don't need your phone to make you Garmin play music. Instead, why don't they make it possible to intuitively setup the infinite data points and features they keep adding to devices in an easy way using my smartphone and then just send that setting to the device and presto? Yes complicated I'm sure. But that is actually a new feature that would have made me go out and buy a 945 instead of keeping my 935.

not just their software, their hardware reliability is terrible. My 18 month old Varia UT800 light just died (1 year warranty of course). A few months ago it was the Edge 1000 that had a flimsy power button that many people had issues with when it finally came off, had to spend $$ for a refurb. My 735xt no longer is detected by Garmin connect, it is also a refurb after the previous incarnation died , again just beyond the warranty.

They have lots of great features, products that look good, however they should focus on packaging and quality versus more features.

i'm not so sure... i've had a few issues with my garmins over the years so i'm not defending garmin.
however fitbit are absolutely appalling on both hardware and software as well as support, i'm saying this based on experience with simple activity trackers compared to complex functionality garmin devices. it seems to be industry standard to focus on sales features rather than making the basics actually work.
i'm thinking of ditching my fitbit in favour of going all garmin

I've gotta say, I've got a Garmin 520 thats going on 4 or 5 years that has been faultless and also happy with a Fenix 5 (only complaint being occasional bluetooth connectivity issues with iPhone- although I just upgraded from a 6 to 8 and it seems much better now (so maybe it was the phone).