I’ve just finished scoring today’s test run with the Polar Ignite and the Suunto 5. The Suunto 5 briefly, just briefly, took the accuracy crown away from the Ambit 3.
HEADS UP: Suunto 5 Review
I’ll cover the IGNITE first…it’s easier: I like the Polar Ignite as an overall package aimed at being a “fitness offering” but today’s GPS test results were not good. I’ll have to add a positive and say that its oHR DOES seems good. I’ll say no more and return to the more interesting news.
>> Test Methodology + spreadsheet results for all watches HERE <<

Back in August 2017, I had the AMBIT 3Run turn in a very impressive 85% in my test. Since then, apart from the Coros Apex 46mm, it’s only really been some Suunto devices that have come close to taking the accuracy crown away by delivering scores of over 80%. Putting it into perspective, the highest ever Garmins are the Fenix 3, Fenix 5s Plus and Forerunner 935 all at 79%.
Yet today the Suunto 5 DID score 85%. However, its GPS track is not ‘pretty’. I don’t take prettiness as an objective criterion but the track just doesn’t look right and jumps up and down every second; albeit jumping up and down very close to where I really was at the time. I’m guessing that the jumpiness then led to a relatively bad over-reporting of distance by more than 200m (or >1% over 10 miles). In itself, that over-reporting of distance is unusual and normally means that I will see a bad score.
So I looked closer.
Even when reviewing the track I think I was ‘balanced’ the first time around. However, I think the lack of prettiness means that I need to be harsh and stern in a grumpy-kind-of-way. So a few marks were docked and a final result of 81% ensued. That’s still good and it’s based on GPS-only.

Instant Pace: I was running with STRYD, so the pace was accurate and stable because of STRYD.
Final Thoughts: This is a tricky one. I think Suunto need to be a bit cleverer here and take a leaf out of Apple’s book. If they cheat and add in a bit of smoothing probably based on the simple GPS points rather than taking into account accelerometer movements then the track will look super-sweet. I’m not sure that will help the over distance inaccuracy though.
I’ve also revisited my results from last month and I can’t see the same level of jumpiness there (there had been SOME jumpiness). I also double-checked at GPS-only WAS used. So this is also a little confusing.
I guess I could re-do this with GLONASS and then GALILEO. But it all takes a lot of time and my life is only so long. In any case, GALILEO will probably need a few more months (as we have seen with Garmin)