FAQs

We are often asked for a recommendation for a smart trainer.
And as is so often the case, the answer is: “It depends.”
– Should it be a smart trainer or a smart bike? -> If you don’t need your bike outside every day in winter or if several people want to train on the device, a smart trainer is more advantageous (= geometry and saddle of your road bike).
– How powerful should the device be? -> Entry-level appliances like to advertise with e.g. 500 W (which sounds like a lot), but they overheat if you use more than e.g. 175 W continuously. That’s why maximum values of 1500/2200 W make a lot of sense in our eyes.
– Are special functions important to you? -> For example, if you want to use the device away from home without a power connection. Or you ride longer units and need a flexible base frame to relieve your buttocks. Are you preparing for the Ötztaler and want the incline to adjust? And much more.
– What budget would you like to spend? -> From €300 to €3,000, anything is possible. But you can get really good smart trainers from € 500.

Our only clear recommendation is:
If you want to buy a smart trainer, buy one with a direct drive (i.e. one where the rear wheel is removed and the chain runs directly over the roller trainer). Smart trainers where only one roller is pressed against the rear wheel are too loud and simply don’t feel good during intervals.

Depending on which device or which software you use to look at the same route, the altitude readings are very different? There is a mathematical reason for this. Of course, the barometer on one device can be more accurate than on another device. But this doesn’t really matter that much. Neither is the reason whether the device is currently calibrated or not. The altitude meters are counted relatively, without calibration “only” the start altitude and also the end altitude are not correct. The decisive factor is when an altitude meter counts!
If a route is only ascending, i.e. each measuring point is higher than the previous measuring point, it is very simple. The altitude meters are the highest minus the lowest measuring point. But what happens if a route is undulating and a measuring point is sometimes a few centimeters below and sometimes above the last measuring point? Depending on the device/software, there is a limit as to when an altitude meter counts at all. This is often, but not always, around 3 meters. In other words, if you ride 2.99m “uphill” and then 2.99m “downhill”, this is not counted as an altitude meter, but as a flat route with 0 altitude meters, the same applies to 10cm uphill, 10 cm downhill, 10cm uphill, 10cm downhill… This is also often the logic in the procedures where Strava and GarminConnect “recalculate” the altitude meters and arrive at the values that we know.
And this is precisely the reason why the values are very close together on real climbs, but far apart on undulating or flat routes. On a relatively flat course, the difference between the centimeter-accurate addition of the altitude meters and the smoothing is in the range of a factor of 3 and here the difference is all the smaller the better the barometer is.
Here are a few comparisons based on a Challenge Roth course ridden with a Garmin bike computer (first lap plus descent, i.e. 93.6 km, the second lap is only just under 85 km long), which illustrate this very well:
The resulting GPX file has 11,236 individual points. If you count the meters in altitude to the nearest centimetre, you get 1,397 meters in altitude (a). If you calculate out all changes in altitude less than 3m, you get 701hm (b).
The original route comes to 803hm with GarminConnect with altitude correction (c) and without altimeter correction (ie barometric) to f=709hm (d). We see that thus (b) and (d) only 8 meters apart. This should be a realistic value.
icTrainer smoothes the route in GPX mode with the average value of 15 seconds each, so you still get 751 points with 804m to the centimeter (e) and again all changes in altitude less than 3m calculated 699hm (d). Thus the value corresponds to (d) from icTrainer again almost exactly the above values (b) and (d).
If you now upload the route traveled in icTrainer to GarminConnect, you get 901hm there with altitude correction (g) and without altitude correction to 702hm (H). Again, we see the value (h) is very close to (d), (b) and (d).
You can see the wide range of results here. Incidentally, Challenge Roth specifies 1,200 meters in altitude for the entire route, which would mean 625 meters in altitude on the first lap plus descent and 575 meters in altitude for the second lap. This figure is therefore significantly lower.
Similar information can also be found in the following article on the same topic.