Just read that the post that the garmin 255 has been approved for use, which is great news , my question is on other models . from the web I found that the fenix 7 uses the same chipset

off the web ..
The Fenix 7 offers different modes for using GNSS (global navigation satellite system). Since Airoha's AG3335M chipset is used in all models, the watch supports GPS, Galileo, GLONASS, BeiDou, NavIC and QZSS.If the GPS mode is activated,

I am no expert but wouldn't it be easier to approve all models with this chipset as long as files are uploaded in the correct format .

possibly easier. But we need to be sure the different models work the same, it's not just the chipset, it's the way the different watches implement them.

The chances are, there will be no problems, we just don't want to create a problem for us down the track.
So a bit of patience is called for.

If you have both Fenix 7 with the Airoha's AG3335M chipset and a Motion or ESP, we would love for you to send us the files from the same session for test verification.

We have gathered a couple of sessions for Fenix7 Pro Solar, but need a few more.

Same with any other models that use the Airoha's AG3335M chipset.

Oh, and please note, the watches apps must be set to All Systems and Multiband.

Select to expand quote

bc said..
Just read that the post that the garmin 255 has been approved for use, which is great news , my question is on other models . from the web I found that the fenix 7 uses the same chipset

There are two related chipsets used in Garmin watches.

- AG3335M is the multi-band chipset
- AG3335MN is the single-band variant, although technically it supports dual-band for the Indian system (NavIC)

The chipsets of every single Garmin watch are listed at logiqx.github.io/gps-details/devices/garmin/watches/

Earlier this year, I also used the API at GPS-Speedsurfing to ascertain the popularity of different Garmin watches. There are 10 times more fenix sessions than Forerunner sessions on GP3S, and a similar number of Instinct 2 sessions (which is non-Airoha) as the various Forerunners.

The fenix 7 models are great for their ruggedness, but the different models will all need to be assessed individually. This will obviously take time and require lots of data - Garmin vs Motion / ESP. Some of the fenix 7 models use an AG3335MN, so don't support multi-band.

The Forerunner 255 has been tested pretty relentlessly during 2024 and is excellent value for money. I wrote up the findings of my ongoing testing and investigations at logiqx.github.io/gps-details/devices/garmin/review/

It is worth noting that the Airoha-based watches aren't quite so precise as a Motion or ESP, but they are certainly impressive. They are way better than any of the older Garmins that used MediaTek or Sony chipsets.

The performance of the Airoha and the availabilty of APPro Windsurf + GPSTC V4 are just what we need right now!

Select to expand quote

sailquik said..
If you have both Fenix 7 with the Airoha's AG3335M chipset and a Motion or ESP, we would love for you to send us the files from the same session for test verification.

We have gathered a couple of sessions for Fenix7 Pro Solar, but need a few more.

Same with any other models that use the Airoha's AG3335M chipset.

Oh, and please note, the watches apps must be set to All Systems and Multiband.

I have 2 years of fenix 7 pro solar and 3 months fenix 8 pro can try get some files for you.

Select to expand quote

sailquik said..
If you have both Fenix 7 with the Airoha's AG3335M chipset and a Motion or ESP, we would love for you to send us the files from the same session for test verification.

We have gathered a couple of sessions for Fenix7 Pro Solar, but need a few more.

Same with any other models that use the Airoha's AG3335M chipset.

Oh, and please note, the watches apps must be set to All Systems and Multiband.

Still wonder...

"The bottom line is that those watches are great toys for on the water feedback, but don't ever rely on them being accurate all the time for competitions or comparisons with your mates. "

Magic words by Sailquick ...

www.seabreeze.com.au/forums/Windsurfing/Gps/Garmin-Fenix-7?page=2

Select to expand quote

CarlosSainz said..

sailquik said..
If you have both Fenix 7 with the Airoha's AG3335M chipset and a Motion or ESP, we would love for you to send us the files from the same session for test verification.

We have gathered a couple of sessions for Fenix7 Pro Solar, but need a few more.

Same with any other models that use the Airoha's AG3335M chipset.

Oh, and please note, the watches apps must be set to All Systems and Multiband.

Still wonder...

"The bottom line is that those watches are great toys for on the water feedback, but don't ever rely on them being accurate all the time for competitions or comparisons with your mates. "

Magic words by Sailquick ...

www.seabreeze.com.au/forums/Windsurfing/Gps/Garmin-Fenix-7?page=2

Bear in mind that was over 2 years ago, prior to Garmin fixing an issue in their firmware.

The performance of the Airoha-based watches was transformed by an update later in 2022.

Select to expand quote

K888 said..

CarlosSainz said..

sailquik said..
If you have both Fenix 7 with the Airoha's AG3335M chipset and a Motion or ESP, we would love for you to send us the files from the same session for test verification.

We have gathered a couple of sessions for Fenix7 Pro Solar, but need a few more.

Same with any other models that use the Airoha's AG3335M chipset.

Oh, and please note, the watches apps must be set to All Systems and Multiband.

Still wonder...

"The bottom line is that those watches are great toys for on the water feedback, but don't ever rely on them being accurate all the time for competitions or comparisons with your mates. "

Magic words by Sailquick ...

www.seabreeze.com.au/forums/Windsurfing/Gps/Garmin-Fenix-7?page=2

Bear in mind that was over 2 years ago, prior to Garmin fixing an issue in their firmware.

The performance of the Airoha-based watches was transformed by an update later in 2022.

Agree.... never say never

Select to expand quote

TRIMMER said..

sailquik said..
If you have both Fenix 7 with the Airoha's AG3335M chipset and a Motion or ESP, we would love for you to send us the files from the same session for test verification.

We have gathered a couple of sessions for Fenix7 Pro Solar, but need a few more.

Same with any other models that use the Airoha's AG3335M chipset.

Oh, and please note, the watches apps must be set to All Systems and Multiband.

I have 2 years of fenix 7 pro solar and 3 months fenix 8 pro can try get some files for you.

Great Trimmer send the comparison files to info@gpsteamchallenge.com.au

A big thanks to Daffy, Decrepit and a few others for doing the huge amount of work and validation for these devices. Many complained about the ban on watches but very few wanted to do the work. And there was a lot data to go through. Not having the tech knowledge on how , but understanding the 'why it's important 'gave me the confidence to say yes to these watches. Over the next few years it shouldn't be an issue as the old watches die and are replaced by better products. I hope the GPS community understands and appreciates the work that that's been done to improve but maintain the quality of the GPS Team Challenge. I do.

And the big thanks must go to Andrew Haigh for all the work he has done with the Garmin watcher's ,just about every windsurfer and wind foiler is wearing a garmin watch, thanks mate.

Have just sent you guys some Garmin Forerunner 965 / Motions files for comparison hope this helps

Select to expand quote

kato said..
A big thanks to Daffy, Decrepit and a few others for doing the huge amount of work and validation for these devices. Many complained about the ban on watches but very few wanted to do the work. And there was a lot data to go through. Not having the tech knowledge on how , but understanding the 'why it's important 'gave me the confidence to say yes to these watches. Over the next few years it shouldn't be an issue as the old watches die and are replaced by better products. I hope the GPS community understands and appreciates the work that that's been done to improve but maintain the quality of the GPS Team Challenge. I do.

Funny, in recent years the TC almost crucified anyone talking about watches and now presenting it as a major accomplishment... anyway happy with this direction, just received my ESP-GPS and more than willing to share some data from a Fenix 8 and ESP-GPS session ??

Select to expand quote

YP1 said..

TRIMMER said..

sailquik said..
If you have both Fenix 7 with the Airoha's AG3335M chipset and a Motion or ESP, we would love for you to send us the files from the same session for test verification.

We have gathered a couple of sessions for Fenix7 Pro Solar, but need a few more.

Same with any other models that use the Airoha's AG3335M chipset.

Oh, and please note, the watches apps must be set to All Systems and Multiband.

I have 2 years of fenix 7 pro solar and 3 months fenix 8 pro can try get some files for you.

Great Trimmer send the comparison files to info@gpsteamchallenge.com.au

Done

I'd like to ascertain which firmware releases in 2022 fixed the problems with the fenix 7 and Forerunner 255 / 955.

If anyone is willing to share some data for this investigation, please can you contact me via PM?

There is no need for any reference data, just some FIT files from your watch for a few months throughout 2022.

Thanks in advance!

Maybe a silly question, any chance of the Apple Watch getting the same sign off?
on the fence about the Garmin FR 255, a little worried about compatibility with Apple stuff.
cheers

Rob

GPSTC have not as yet considered any Apple watches.
Here is a short report by Michael George who has done a little research.

logiqx.github.io/gps-details/devices/apple/ultra/

It is very unlikely that GPSTC will look at Apple Watches in the near future.

Select to expand quote

merlincat said..
on the fence about the Garmin FR 255, a little worried about compatibility with Apple stuff.
cheers

Rob

I just checked and there are garmin connect apps in the playstore.

I'm running Linux on this computer and can download my files, in common usb mode.
I don't think you'd have any problems.

Select to expand quote

merlincat said..
Maybe a silly question, any chance of the Apple Watch getting the same sign off?
on the fence about the Garmin FR 255, a little worried about compatibility with Apple stuff.
cheers

Rob

I mostly use Apple Mac and Iphone. Both seem to work fine with the Garmin apps. :-)

Select to expand quote

CarlosSainz said..
Agree.... never say never
Still wonder...

"The bottom line is that those watches are great toys for on the water feedback, but don't ever rely on them being accurate all the time for competitions or comparisons with your mates. "

Magic words by Sailquick ...

www.seabreeze.com.au/forums/Windsurfing/Gps/Garmin-Fenix-7?page=2

The statement about the older watches still stands 100%.

And the new ones are deemed 'good enough', particularly because there is a severe shortage of the Motions, but no one who has studied the data is under any illusions that they are anywhere near as accurate/precise and the Motions/ ESP's .

Have you studied the data doing large numbers of side by side comparisons?

Select to expand quote

merlincat said..
Maybe a silly question, any chance of the Apple Watch getting the same sign off?
on the fence about the Garmin FR 255, a little worried about compatibility with Apple stuff.
cheers

Rob

I think you have nothing to worry about with apple. You install the Garmin Connect app to you iphone and pair it to your watch, and then it will just auto upload activites to Garmin Connect (cloud on the internet). You can then use any computer with a browser to export the FIT file and then upload to KA72 or whatever you normal workflow is for getting it onto GPSTC. Garmin works very well with Apple phones/Macs etc.

In regard to other smart watch functionality... there are some limitations compared to an Apple watch. Garmin being more fitness focussed and Apple more smart watch focussed. Pros and Cons of each. But the battery life of your typical Garmin is WAY better than an apple watch which sold me.

Hi folks,

Thanks for the replies, so integration is looking good then. What I like about the Garmin is the price, ?200 and the battery life.

Does anyone use the Corus pace 3? A few of the guys have them here.
although probably opt for the FR 255, as it's been qualified as being ok.
Merry Christmas

rob

Select to expand quote

sailquik said..
The statement about the older watches still stands 100%.

And the new ones are deemed 'good enough', particularly because there is a severe shortage of the Motions, but no one who has studied the data is under any illusions that they are anywhere near as accurate/precise and the Motions/ ESP's .

Agreed. The older watches weren't remotely close to being suitable, unless anyone considers plus or minus a couple of knots to be acceptable.

The newer watches are significantly better and on a good day will only be 1 or 2 tenths of a knot out on the results, but up to half a knot is not uncommon for 2s and even 500m can be off by a few tenths. In an ideal world we'd have something better, but these watches are readily available and when set up right, they produce pretty decent results. The phrase "good enough" just about sums it up perfectly.

Anyone who has access to a Motion or ESP will still get more accurate results. The general variability and errors in those results are an order of magnitude smaller than the best watches available right now. These are observations that apply to pretty much every single test that has been undertaken throughout 2024, and recent collaborative efforts.

Select to expand quote

merlincat said..
Does anyone use the Corus pace 3? A few of the guys have them here.
although probably opt for the FR 255, as it's been qualified as being ok.

The speedsurfing activity on the various APEX and VERTIX models is the only mode that provides Doppler-derived speeds. Every other mode on COROS is non-Doppler and implements some kind of filtering / smoothing which is undesirable when speed sailing.

The PACE 3 doesn't have the speedsurfing activity, so you are left without a suitable mode. Owners of the PACE watches are best advised to use the GPS Cardio mode, but it's still non-Doppler and the FR 255 will perform a lot better.

Select to expand quote

merlincat said..
Maybe a silly question, any chance of the Apple Watch getting the same sign off?

Apple have used a variety of GPS / GNSS chipsets from different manufacturers over time and the performance has varied. I've yet to see any data from an Apple watch that is as encouraging as the Garmins, but that's not to say the situation won't change in the future.

Select to expand quote

K888 said..

sailquik said..
The statement about the older watches still stands 100%.

And the new ones are deemed 'good enough', particularly because there is a severe shortage of the Motions, but no one who has studied the data is under any illusions that they are anywhere near as accurate/precise and the Motions/ ESP's .

Agreed. The older watches weren't remotely close to being suitable, unless anyone considers plus or minus a couple of knots to be acceptable.

The newer watches are significantly better and on a good day will only be 1 or 2 tenths of a knot out on the results, but up to half a knot is not uncommon for 2s and even 500m can off by a few tenths. In an ideal world we'd have something better, but these watches are readily available and when set up right, they produce pretty decent results. The phrase "good enough" just about sums it up perfectly.

Anyone who has access to a Motion or ESP will still get more accurate results. The general variability and errors in those results are an order of magnitude smaller than the best watches available right now. These are observations that apply to pretty much every single test that has been undertaken throughout 2024, and recent collaborative efforts.

merlincat said..
Does anyone use the Corus pace 3? A few of the guys have them here.
although probably opt for the FR 255, as it's been qualified as being ok.

The speedsurfing activity on the various APEX and VERTIX models is the only mode that provides Doppler-derived speeds. Every other mode on COROS is non-Doppler and implements some kind of filtering / smoothing which is undesirable when speed sailing.

The PACE 3 doesn't have the speedsurfing activity, so you are left without a suitable mode. Owners of the PACE watches are best advised to use the GPS Cardio mode, but it's still non-Doppler and the FR 255 will perform a lot better.

merlincat said..
Maybe a silly question, any chance of the Apple Watch getting the same sign off?

Apple have used a variety of GPS / GNSS chipsets from different manufacturers over time and the performance has varied. I've yet to see any data from an Apple watch that is as encouraging as the Garmins, but that's not to say the situation won't change in the future.

Thanks for the reply K888, glad of the extra information. I am hoping with the purchase of the FR255 my speed improves dramatically! ??

just need to see what Father Christmas brings!
Have a great Christmas everyone.
rob

I have a GW60, Motion Mini and two DIY GPSs. I'm happy to compare my Samsung Galaxy Watch 6 at the same time if that helps.

Select to expand quote

sailquik said..
And the new ones are deemed 'good enough', particularly because there is a severe shortage of the Motions, but no one who has studied the data is under any illusions that they are anywhere near as accurate/precise and the Motions/ ESP's .

I disagree. The idea that the Garmin 255 delivers very accurate data arose when tbwonder posted the result of many comparisons to u-blox based GPS units, which showed very small differences. All the Garmin 255 data I have seen are indeed very accurate.

Don't confuse the "+/-" numbers that the analysis programs give with actual accuracy. Especially for the longer categories (e.g. nautical mile), it is quite common that the observed differences between 2 u-blox units is greater that "statistically expected". That's because the statistics used assume "white" (completely random) noise, while GPS data have multiple sources of "colored" (non-random) noise.

One category that's prone to marginally higher differences are alphas. That, however, is largely a calculation artifact. Basically, higher-rate data can often get closer to 500 meters than 1-hz data, which can add 0.1 knots or so. This particular issue is something that can easily be fixed by interpolation, for example splines. The original software packages like GPSAR gave that as an option, although probably just for top speed categories, not alphas. The authors should know the details.

Select to expand quote

boardsurfr said..
I disagree. The idea that the Garmin 255 delivers very accurate data arose when tbwonder posted the result of many comparisons to u-blox based GPS units, which showed very small differences. All the Garmin 255 data I have seen are indeed very accurate.

Yes, "very small differences" that we have agreed are 'Good enough' or our fun competition team challenge. ie: mostly in the order of 0.1 to 0.2 knots sometimes a bit more.

But 'very small' is relative. I suppose you could call 0.1 to 0.2 knots 'very small', but in the context of individual rankings, where many individuals can be within that range of 0.2 knots, it suddenly does not seem so 'very small'. If you have just spent all day battling with your mates on a speed course in the best conditions of the year and at the end a couple of them beat you by 0.1 or 0.2 knots, it does not seem like such a small difference.

Controlled and real sailing side by side testing has repeatedly shown them not nearly as accurate as the Ublox Multi Hz devices, as K888 has stated above, confirmed through his testing. In the same sort of tests we are seeing almost an order of magnitude greater with the ublox Multi Hz devices (In the order of 0.02 knots) . And that does not rely on comparing +/- numbers, but comparing actual results.

I agree with your comments about NM/ 1 hour results.

And with Alphas you are correct that the higher Hz data can find a better ideal selection of the track, but I think there are other factors involved in the greater differences as well, related to the sudden 180 degree turn and possible changes in the constellation view of the device in the turn.

But the biggest factor with these watches that convinced us they are 'good enough' is that they do not seem to be prone to the very large errors we saw quite often in previous generation implementations, well illustrated by tbwonders large body of data.

And example of controlled side by side testing off the water:

Select to expand quote

sailquik said..

Yes, "very small differences" that we have agreed are 'Good enough' or our fun competition team challenge. ie: mostly in the order of 0.1 to 0.2 knots sometimes a bit more.

But 'very small' is relative. I suppose you could call 0.1 to 0.2 knots 'very small', but in the context of individual rankings, where many individuals can be within that range of 0.2 knots, it suddenly does not seem so 'very small'. If you have just spent all day battling with your mates on a speed course in the best conditions of the year and at the end a couple of them beat you by 0.1 or 0.2 knots, it does not seem like such a small difference.

Controlled and real sailing side by side testing has repeatedly shown them not nearly as accurate as the Ublox Multi Hz devices, as K888 has stated above, confirmed through his testing. In the same sort of tests we are seeing almost an order of magnitude greater with the ublox Multi Hz devices (In the order of 0.02 knots) . And that does not rely on comparing +/- numbers, but comparing actual results.

I agree with your comments about NM/ 1 hour results.

And with Alphas you are correct that the higher Hz data can find a better ideal selection of the track, but I think there are other factors involved in the greater differences as well, related to the sudden 180 degree turn and possible changes in the constellation view of the device in the turn.

But the biggest factor with these watches that convinced us they are 'good enough' is that they do not seem to be prone to the very large errors we saw quite often in previous generation implementations, well illustrated by tbwonders large body of data.

And example of controlled side by side testing off the water:

More importantly, what bike are you riding?

Controlled side by side testing is one thing. But in the real world people stick their GPS devices in all sorts of places. Sometimes on their wrist, perhaps on their arm, they are prone to slipping out of position. Some stick them in their back packs or even under the wet suits. In my experience even Motion devices can report up to 0.3 kt differences when worn in different places. Usually, they are closer but differences of 0.1 are not uncommon between other approved devices.

Select to expand quote

tbwonder said..
Controlled side by side testing is one thing. But in the real world people stick their GPS devices in all sorts of places. Sometimes on their wrist, perhaps on their arm, they are prone to slipping out of position. Some stick them in their back packs or even under the wet suits. In my experience even Motion devices can report up to 0.3 kt differences when worn in different places. Usually, they are closer but differences of 0.1 are not uncommon between other approved devices.

Agreed, I wear a GW60 on the top of my wrist (not underneath which could be better), a Motion on my upper foream (which can slip below my forearm) and one or two DIY GPSs on my boom. I can't remember seeing a difference greater than 0.1 kn.

Select to expand quote

tbwonder said..
Controlled side by side testing is one thing. But in the real world people stick their GPS devices in all sorts of places. Sometimes on their wrist, perhaps on their arm, they are prone to slipping out of position. Some stick them in their back packs or even under the wet suits. In my experience even Motion devices can report up to 0.3 kt differences when worn in different places. Usually, they are closer but differences of 0.1 are not uncommon between other approved devices.

Of course this is true about real world use. We have all seen some very poor results with poorly worn devices. Even the best devices will produce poor data if they are worn in a way that severely compromises their function, like in chest or back pockets of wet vests, or under the wetsuit sleeve, or what all too commonly happens when they can slip around under your arm.

For well worn and positioned Motions, errors of 0.1 Knots and above are very uncommon in all of the huge numbers of comparisons I have analysed. In my testing, with properly wrist worn identical watches on each arm, differences of up to 0.1 knots and sometimes slightly over 0.2 knots have been more common than I would have ideally liked, but I think they are still acceptable, given the availability bind the GPSTC find it's in, and the huge step forward from previous Garmin Watches.

But the point is, that controlled side by side testing establishes the precision of the devices in good conditions. With the G255 watches it is demonstrably in the order of 0.1 knots or slightly better. With Motions it is the order of 0.02 knots or slightly better. So when they get used in 'real world' less than ideal conditions, the degradation of accuracy isn't starting from the same base, and so far I have seen nothing that suggests otherwise.

The GW-60 watch is a pertinent example. It tested at a very high level of side by side precision when tested in ideal orientation by myself and others. As did the GW-52 which used the same 'engine' (but different antenna). I have a lot of data on this! But of course, the GW-60 was primarily as wrist worn device (and is GPS only, and not multi band). It demonstrably suffered from wearing in its intended wrist position when underhand facing down to the water, and whipping around suddenly during gybes, and bouncing around with the wrist more than an upper arm mounted device did. And yes, I did see much greater differences between that and the upper arm and helmet mounted devices I wore at the same time. Yes, they could often be in the order of 0.1 knots, and even higher in extreme cases. Of all the inherently accurate multi Hz devices I have tested, those watches performed the worst in real world use when worn as intended on the wrist.

So I was very pleasantly surprised when your initial data showed quite good correlation with normally arm worn Ublox based devices and the Garmin 255 wrist worn watches. I agreed with you that this was something well worth looking into. And it seems that the advancement of the technology of the Airoha multi GNSS and Multiband chipset, and the way Garmin have implemented it in their Sports watch, has been a big step forward, largely mitigating the worst of the effects we saw with the GW-60 watches and other, previous generation gps watches.

But all my testing and analysis has clearly shown two things:

1. The Airoha Multi GNSS/Mutiband Garmin 255 watch is 'good enough' for the GPS-Teamchallenge competition.
2. It is NOT as accurate as the Motions and ESP's if you want a higher level of confidence and precision in your speed results.

And the third really important thing to emphasise: Any GNSS device will give much better and higher confidence accuracy, when it is worn with care in the best position that is practical and avoiding as much as possible those things that we know can compromise it's performance. One thing that this Garmin watch was got going for it, is that if it is worn on the outside of your wetsuit sleeve, it should give consistent results, as it is hard for it to suffer some of the other more serious compromises mentioned.