This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Calories Estimate Different on 1040

9756031
over 2 years ago

The calories calculated on the 1040 is significantly different from that on the 1030 (not Plus). Same cycle, same sensors (ported across), same ride area similar av. speeds, climbs, etc. Generally on the 1030 the calories per mile would be close of 42. Same everything (given no 2 rides can be absolutely identical) but on 1040 so far calories per mile is always around 38. Which is actually 10%'ish change!

Only sensors are speed and cadence (cadence sensor being used does seem to have an impact on the calculated values). I appreciate it's not an absolute measure and just a number indicator more for comparison that explicit use but it suggests a different algorithm or weighting or something.

Anybody else noticed this or know why? (I have no idea about what happened on 1030+.

  • 0 over 2 years ago

    Without getting to deep in the science and math, if you're not using a power meter, any calorie estimate you are receiving is an estimate that uses fixed values for certain inputs to the math. Bottom line, even the estimate on the 1030+ was probably wrong. Because it is at best a scientific guess without a PM, if knowing the calories burned is important to you, getting. the lower estimated number is better than an overestimate.

  • 0 over 2 years ago in reply to der_bierpuncheur

    It's not important as an absolute number e.g. I don't ride 300 kcal so I can eat a 300 kcal bar of chocolate not anything like that. More for comparison between rides - a bit like an index of speed & ascent, an indicator of much effort you put in. So difficult to compare rides which were recorded on 1030 with rides now recorded on 1040.

    Just as an index, I also treat ascent as an index for comparing rides rather than thinking 300m, more yesterday was 300, today 450 so today was more than yesterday.

  • 0 over 2 years ago

    I've noticed that the 1040 reports about 10% less total ascent than the 1030 (speed, cadence and HRM used). This could be one factor in the lower calorie estimate by the 1040. I had compared the 1030's estimate with http://bikecalculator.com/ and thought it reasonable.

  • 0 over 2 years ago in reply to JRat1

    Without a power meter, which records work performed in kiloJoules (kJ), which can be reliably converted to calories, the Garmin has to take into account several factors to estimate calories burned.

    This is where it gets tricky. The potential inputs are (a) rider+bicycle weight, (b) HR data (if you are using an HRM), and (c) elapsed time.  The problem with each of these inputs is that the Garmin has to assign a value to these things, and in each case there are ways that the data inputs will give wildly different values, day to day, rider to rider, or even as the ride progresses and exhaustion and dehydration set in. Even bikecaluclator is a guess. Total ascent is not a factor for calories burned...other than climbing raises the HR. 

    Bottom line, calories burned between rides is a neat comparison with no scientific basis if you are not using a PM. Without a PM the calories burned is subjective. Comparing total ascent and total distance, or average speed, is a better way to compare rides because the numbers are verifiable and objectively comparable.

  • 0 over 2 years ago in reply to der_bierpuncheur

    I agree that there are numerous assumptions in the calculation but we could benefit from an explanation from Garmin why the result, given the same input data, is lower than the 1040 than the 1030. As I noted above, the 1040 appears to be recording a smaller total ascent than the 1030 which could partly explain the difference but what else has changed. Has Garmin done more research and analysis?

  • 0 over 2 years ago in reply to JRat1

    Like I said, recorded ascent isn’t the issue, because the meters gained or lost is not directly a factor. If you are using a HRM, climbing/descending will generate different HRF and thus different inputs to the calculation. If no HRM, it’s just a factor of time, and then the ascent has no bearing at all. I’d guess that Garmin (by way of FirstBeat which they purchased a few years ago) has improved the algorithm they use, so the lower number may be more “accurate,” but as I said, without a Power Meter, it is at best a rough estimation with some science in the form of data extrapolation involved. I can’t really explain it better because I am not a data scientist and I am not a Garmin employee.

  • 0 over 2 years ago in reply to der_bierpuncheur
    der_bierpuncheur said:
    Total ascent is not a factor for calories burned...other than climbing raises the HR.

    Ascent does definitely impact the calorie calculation with no HR and no power meter (at least it did on the 1030). I rum with just speed and cadence sensors and a hilly ride will add calories far faster than a flat ride (more noticeable than speed). So without the extra sensors old algorithm considered ascent as a fairly important in the calculation.

  • 0 over 2 years ago in reply to 9756031

    Interesting. So you’re saying that with just a GPS plot a hillier route showed more calories burned than a flatter route? Maybe because it took longer? Or maybe it was on a different device with a different algorithm that uses different estimates. Like I tried to make clear above, if you want accurate calorie counts, get a power Meter, otherwise it's all guesswork, though using the same device to do the guesswork will make it generally relevant. Comparing one device not another is not really relevant, though.

  • 0 over 2 years ago in reply to der_bierpuncheur

    Different routes so no subjective "proof". My average speed is pretty constant between hilly and flatter routes (you go faster and get a rest on the downhill). I tend to work on calories per mile in my head (doing simple math in my head whilst cycling is something I seem to do quite a lot).

    I'm also confident that on 1030 the presence of a cadence sensor gives more calories per mile (I've had problems with the sensor on occasions so have done the same rides with and without cadence sensor),

    I used speed and cadence sensors (the Garmin ones). I've only used Garmin 1030 for several years and the 1040 for a short time and the calorie estimate is different on the 1040 to 1030 so I can only assume 1040 has a different algorithm.

    I regard the calorie estimate as a type of index, ignore the units, just as an indicator of how hard you've worked.