The DCR Analyzer Is Now Available For You To Use!

DSC_7355

And by now…I mean like 5 months ago. But no matter, I’m finally getting around to telling you about it in some manner that might be noticeable.  Or even semi-formal.

But wait – what’s the DCR Analyzer?

You know in my in-depth reviews (you do read those, right?) where I’ve got all these fancy charts that compare one power meter to another (or 4 others)?  Or one heart rate strap to another or three?  Or possibly one GPS track to 7 others?  Well, that’s the DCR Analyzer.

It’s a purpose built tool that I’ve been using for almost two years now, designed specifically for comparing two theoretically like things.  So how does one power meter track against another on the same ride.  Or how does one optical HR sensor compare to a heart rate strap.  That’s the goal.

It originated from my frustration with spending 15-45 minutes per ride/run/activity trying to normalize and compare the data using off the shelf tools.  I used to use a combination of Golden Cheetah, Excel, and a large hammer to make everything work.  The problem was that no tool exactly fit what I needed.  And with a typical in-depth review having anywhere from 10-30 sets of data contained within it, it became a nightmare to quickly generate the core data that the reviews are often based around.

Hence…the DCR Analyzer was born!

I worked closely with a developer (of FitFileTools.com) to inch our way towards a pretty sweet toolset for comparing data from the same activity.  And early last year (2016) a smattering of you started beta testing it as well.  Then earlier this past year (January 2017) I opened the doors to anyone who wanted to use it.

Since then, far more of you than I ever expected have been making comparative data sets and seeing how your devices measure up.  It’s been really cool to see!  Especially cooler to see it being used throughout the industry, and even some of the UCI World Tour Pro Cycling teams are now using it for analyzing and validating data.

FullCapture_thumb

Now there are free options out there to do similar things.  But there are a bunch of nuanced reasons why I feel the DCR Analyzer is better suited for data comparisons:

– Automatic Alignment: It automatically aligns files based on GPS timestamps (or any timestamps).  So you don’t have to worry about pressing start at the same time, or about pauses
– Data file support: It doesn’t barf on files with the latest data types.  Many 3rd party tools have trouble with file format specs that are updated with new data types.  Since I use the tool on usually unreleased products, I get these supported in the Analyzer well before you even know the products exist. :)
– Sensor support: The DCR Analyzer allows you to save your sensors and then automatically recognize them for .FIT files.  This is great if you record lots of things at once and accidentally mix up which unit recorded which.
– Proper mapping: The ability to get true 1-second recording data tracks on the map is really important to me.  A lot of sites/tools won’t properly display all the data points on the GPS map, which sorta defeats the purpose for analysis reasons.
– Constant improvements: We’re super open to adding features that make comparative analysis quicker and easier.  I’ve got lots of ideas for how to keep improving the suite of tools for this purpose – without losing sight of the fundamental purpose: Comparing data files
– Data Types: It supports comparing a slew of data types: Heart rate data, Power meter data, Cadence data, Elevation data, Distance data, GPS Tracks

Now, to be clear – this is NOT a training log site.  It’s not designed to replace sites like Training Peaks, Strava, SportTracks…or apps like WKO+ or SportTracks.  Rather, it’s simply a tool to compare multiple sensors/tracks/whatever’s on the same activity.  I still use all those tools myself as my training log.  The DCR Analyzer is more like a graphing calculator than logbook.

image_thumb24

So how do you get started?  Well, there’s two options.  First is a day pass, and then an annual pass.  The day pass gives you 24 hours to create up to 3 sets of data (each set contains multiple files), and then you’ve got indefinite read-access after that.  This is great if you’ve got a quick one-time use for it.  This is $5.

The second option is the all you can eat annual pass.  This allows you to create unlimited sets for an entire year.  All 365 days of it!  This costs $29 for the year, and like the day pass, your sets remain active indefinitely, even beyond your membership.

You can hit the account creation page here!

No matter which option you choose you can share the sets with public links, just like I do in my reviews (and you can also keep things totally private too).  With public links you can post links to forums or e-mail, Snapchat or on a giant sign by the side of the road.  Whatever floats your boat.

If you want to check out more about all the technical capabilities of it, you can read-up on the manual of sorts that I’ve put together here.  This covers how to use the Analyzer, and some of my pro tips for analyzing hundreds of data sets with it.

With that – thanks for reading- and enjoy!

Post a Comment

Your email address will not be published. Required fields are marked.
If you would like a profile picture, simply register at Gravatar, which works here on DCR and across the web.

*

 Notify me of followup comments via e-mail.

You can click here to Subscribe without commenting

Add a picture

46 Comments

  1. Peter

    I find you comparisons in the in-depth reviews really helpful. Making it available for us to use at a pretty reasonable fee takes it to a whole new level. As an owner of several power meters, head units, HR monitors etc I can really get get stuck into the data now. A new level of geekdom! Thanks Ray.

  2. Julien

    Thanks – I have been looking for an easy way to compare my interval workouts!
    Specially 30 / 30 repeats
    This will be perfect.
    I really appreciate the sharing.

    • Logan

      From the manual:
      The tool is not designed for comparing workouts done at different times. Meaning, it’s not designed for comparing a Monday workout to a Tuesday workout on the same course. For that there are better tools as part of training logs.

  3. Dörte

    Is there a free trial version? Or is there at least a price reduction von umeployed?

    • No free trial, mostly because you can check out the public sets (like the one linked to in the post). Though, there is the day pass – so that’s kinda like a reduced price trial. :)

  4. Ryan Nelson

    This seems like a tool that could be used for data geeks that ride together, too — if it will align multiple files based on GPS timestamp, that tells me that one could grab a few buddies’ ride files, throw them together, and get a comparison of individual outputs.

    That seems like a freaking awesome option to me for certain situations where nerding out benefits the team. =)

    • Indeed it would. Would be especially interesting in a team dynamic (like a team time trial even), to see how power drops off one one rider and then increases on another. Nerding out indeed!

    • Steven

      Okay let’s see some old team data. SKY, BMC, who else will post, amateur team? It would be cool to see the aero effect all the way back.

  5. Ryan M.

    Stumbled on the tool a few months back and have had a bunch of fun with it. Super easy to use.

    link to analyze.dcrainmaker.com

    My fiance laughs at me whenever I get ready for a run when I feel like testing out some data.

    • Awesome! And not too bad of numbers in that comparison either.

    • Ryan M.

      The 935 does a very good job on tracking. I find it still struggles the first 5-7 minutes, so try to start the watch up well in advance for it to hopefully find a good lock. I find the H10 strap comfortable so tend to stick with that as the accuracy is so good.

  6. Craig

    Registration asks for FTP. Clearly utilizing a tiered architecture where higher FTP users are given more processing power. ;-)

    Nicely done. Fairly priced. You continue to impress. Keep it up.

  7. Logan

    Awesome. Now if only I had multiple devices to use on the same run…

  8. Theo Katz Battaglia

    What about other types of analysis? Like comparing one metric for one same workout you repeat periodically. I am not aware of any tool where I can get advanced analysis (like fitting matematical models to sections of heart rate curves or comparing heart rate for multiple intervals in one same workout), other then cumbersome work on Excel.

    • It’s not so good for comparing the same workout over and over (I know Sport Tracks has something like that, as does WKO4 and Golden Cheetah). Only because everything is based on timestamps, so if those don’t match, then you’re kinda up a creek.

      Technically speaking you could skew the timestamps to make them align, but you’d have to do a lot of math since the offset adjuster is based on seconds. :-/

    • Or you could use the Time Adjuster tool on FIT File Tools to align the file start times before loading them into the Analyzer.

    • Bruno Moraes

      Theo, you can try Today’s Plan. They have tools to easily compare intervals within the same activity or even across activities.

    • Karl

      Not sure that would work. I think what Ray was alluding too above is that 1s resolution is pretty low (a lot can happen in a second). You’d probably have to resample the data unless you just wanted to compare longer averages. And resampling comes with it’s own (aliasing) issues.

    • Karl

      (Probably fine for HR, maybe not power depending…)
      Another option for FIT files, is the Perl tool “fitsed”, which is a stream editor for FIT files and would let you shift times with a one-liner. (Please ignore if “stream editor” and “one-liner” mean nothing to you…)

  9. Jonathan Burchmore

    It’d be nice if the tool displayed pace for running activities. Also, either I’m doing something wrong or it doesn’t support bluetooth devices paired to a fenix 5X.

    • Hmm, interesting thought on pace. Never come up before – let me look at that.

      For Bluetooth sensors, it definitely supports them (I’ve done some BLE power meters on the 5X). However, it won’t allow you to save sensor ID’s for BLE devices, since the spec itself doesn’t record the sensor ID’s for BLE devices. :(

    • Jonathan Burchmore

      I uploaded a file that should have data from a Bluetooth footpod, but it’s not listed under sensors. My Garmin footpod does show up in the sensor list.

  10. Bob Floyd

    Sorry, embarrassed, but on the registration page I don’t know what to put down for “FTP (watts)”.

  11. Artur

    Thanks Ray!

    Just in time. Got myself new P1 pedals. The first ride showed wattage much higher than I expected from my last FTP test on KICKR, and i’m definitely less fit since then. Would love to know by how much these two are apart, given they are the extreme ends of the power measurements. The DCR analyzer will be perfect for doing such a comparison, and for me readjusting my known power KICKR levels to the new P1s.

    Thanks!

  12. Joe

    Thanks Ray! Been using it for the last few weeks to compare heart rate info from the FR935 vs. HR strap and power data for the P1 pedals vs. KICKR1 or G3 hub. Takes an extra few seconds to rename the Garmin files, but super easy to use.

    • Artur

      HI Joe,

      Please, do tell about the P1 and KICKR1 comparison, as well as the FR935 vs. HR strap comparisons. Would love to hear!

    • Joe

      Artur,

      KICKR1 tends to consistently run 10w higher than P1. Usually use ERG mode in Zwift. Seen the same pattern on a few short free rides on Zwift, so not just ERG mode. I only have one outside ride so far this spring with P1 vs. G3 but they were pretty spot on (even though I forgot to calibrate), so I’d say the KICKR1 is the outlier. Done spindowns, calibrations, crank arm setting adjustment, firmware updates, etc. First link is P1 vs. KICKR1. Second link is P1 vs. G3.

      link to analyze.dcrainmaker.com
      link to analyze.dcrainmaker.com

      Still getting used to the FR935- got it a few weeks ago. Overall I like it. Seems good for 24/7 stuff and GPS, mixed bag for OHR on activities. I’ve been recharging on Sundays with 35-40% left – takes about a minute per 1% to recharge. Also good for data across running/cycling (like recovery time) because it’s on one device instead of two. Seems to not like getting moved around/fidgeted with. Might be mostly my fault though as I tend to wear it *on* the wrist. Tried *off* the wrist, but tends to slip down to wrist since my wrist is narrower than further up. Probably will still use a HR strap for now. Might try OHR again here and there to see if things change.

      Outside running (laps around a cemetery) seemed pretty good on the running laps, but lost the plot once I started walking. Fidgeting with it seemed to help it get the plot when it was off:
      link to analyze.dcrainmaker.com

      On the indoor bike, off the wrist was pretty good. Again, when it was off, fidgeting with it seemed to help *reset* it, particularly in the first few minutes of the workout:
      link to analyze.dcrainmaker.com

      Treadmill run/walk workout, on the wrist – mostly ok:
      link to analyze.dcrainmaker.com

      Weight lifting/resistance bands – off the wrist – pretty bad:
      link to analyze.dcrainmaker.com

      Also tried with mowing the lawn (no link). Similar to weight lifting pretty bad – I’m guessing because of all the jostling.

    • Joe

      New data for the KICKR1: Wahoo updated the KICKR1 firmware about a week ago. I updated it on my KICKR1 yesterday. It’s just one workout, but the P1 vs. KICKR1 power readings are much closer now (average power was only off by 1w). The KICKR1 power data is definitely smoothed (at least in ERG mode – has been for the last 6 months I’ve been using ERG mode in Zwift). Anytime I slightly changed cadence, I could feel the KICKR1 adjust the resistance but on screen it showed the same power (it felt more like the P1 lines show):
      link to analyze.dcrainmaker.com

      Nice to see Wahoo still making updates to the KICKR1, even though they have the KICKR2.

    • I had noticed the 10W difference, that tends to reduce with usage. Have’t updated the fw yet.
      Will see what happens

  13. Pieter

    Thanks – esp for this!!!

    Automatic Alignment

    Will give it a try :)

  14. Tim Grose

    Hello Ray
    This is very useful but there are a few things that would be very useful (at least for me) to consider for additions.

    1. Support for Garmin CIQ “developer data”. I thinking, for instance, of comparing Stryd power. I have got two of them and may want to compare different firmware releases…

    2. Cumulative difference in distance accumulation. Over on the Garmin forums you get a lot of comments like this run on my new watch “lost” distance compared to my “trusty” device. The longer one is always “right” yeah – or is it… It would be nice to compare 2 or more files on the same run to see if the difference was “systematic” or caused by GPS spikes somewhere along the way. I am thinking very similar to the Strava segment compare tool.

    3. Running pace (as already mentioned) would be good as kind of ties in with the previous one. You could then compare GPS derived current pace with foot pod derived pace to see how they shape up. In fact this might also be better done as a difference rather than an absolute value as would then be easier to see where it is “close” or not.

    • Thanks Tim.

      Great suggestions, and ones we’ve been tossing around on how to implement. CIQ developer fields is definitely highest on the radar. And I like the idea of distance accumulation.

      Pace is tricky actually. See pace isn’t recorded in the .FIT file. Rather the GPS locations and time stamps are, and pace is calculated. One of the core tenants we’ve taken is to always ensure whatever data we have exactly matches what’s in the files – and not to guesstimate. So i we did something in this area we’d want to ensure that it’s matching what is shown on the vendors site for pace (i.e. Garmin Connect). Definitely not saying no, just explaining some hesitancy there.

      Thanks again!

    • Tim Grose

      Hello Ray
      Puzzled now with pace and speed? When I open up an activity FIT file using FIT File Repair Tool then Speed (in km/h) is one of the “native” displayed fields (although you can display it as pace) and it does not appear that the program has computed it. You can sort of see that as one data point (with every second recording) has zero speed but a different lon/lat to the adjacent data points.

      Also if I open up a treadmill run or a trainer ride, it again has speed but can’t be computing it from GPS as that was off!

      I am fairly sure the FIT file contains the distance accumulation data (hence my suggestion) recorded by the device as this is how GC matches your device as opposed to some 3rd party apps that may take the GPS data given in the FIT and recompute distance from that.

  15. Steve

    >The DCR Analyzer Is Now Available For You To Use!

    I wait, until it provides a kitchen sink.

  16. Philip

    I have noticed your’e not able to choose Power2Max power meters (Original, Type S or NG). Also you have 4iiii power meters listed but not the heart rate monitor.

    • No problem – added!

      I’ve been adding them as I do comparisons in the calculator – so oddly enough the P2M one likely would have been solved this afternoon in due course. :)

      But good call on the others and the Viiiiva. Thanks!

  17. SpandexBoyWonder

    This looks a really useful tool for a lot of people. Great work developing this!

    Will you have access to the data uploaded by users, even in anonymous form? It strikes me that with a large enough data set of comparisons between pairs of sensors you should be able to rank their accuracy. For example (off the top of my head) you could calculate the RMS difference between the sensors in each comparison, then for each sensor, calculate the mean of the RMS differences across all comparisons against all other sensors. The more accurate sensors should have lower numbers, although it may be more complicated to derive an actual number for the accuracy of each sensor.

    This should work for any measurement I think, including GPS, HR, Power. I think it would be a great complement to your reviews where you can typically only test a single device of each type.

    Just a thought…

    • It’s something I’ve thought about – but not something I’m doing today. Part of the challenge is the unknown on how accurate different folks are when it comes to calibration and such. With a much larger sample size it becomes less important though.

      If things were to change I’d make it opt-in of course, and folks could decide if they wanted their data anonymously included. But for now we’re just working to get in other requests that focus on making analysis better/etc… I think medium term there’s more room for graphs/charts that better tell the story of accuracy for certain data types (power/HR especially) that require less human analysis.

  18. I wish there was a paypal option for paying rather than a credit card …

  19. Changren Y.

    No free trial use for DCR supporters? :)

    • It’s actually something I looked at doing, but logistically we were having some issues getting it done in a timely manner. And by we, I mean me. The DCR Supporter piece is handled in WordPress as ‘Users’, while the Analyzer runs on a separate platform and Stripe handles the backend there.

      So doing a one-time thing while manually cumbersome is possible (a messy mail merge basically), but trying to make it such that if someone signed up as a DCR Supporter that they’d get (instantly) the coupon codes for the Analyzer is where things fell apart a bit. It’s on my radar though – as I definitely want to provide more value in the DCR Supporter piece.

      Thanks for being a supporter!

  20. Brad Davis

    This is super cool. I’ve been playing with a Moov and I want to compare between the Moov data, my Garmin 520 and my FR 235 but I’m too lazy to do it. Now i have one less reason to be lazy about it!

  21. Why not merging the premium subscription to the blog with the usage of the data analyzer?

    • Ok, didn’t read the previous comment :)

    • Sorry, yeah. Definitely was something I was trying to do (and still want to do). It’s actually a core reason I delayed some 4-5 months on announcing the Analyzer. Obviously, not all that time was spent on figuring it out, but one of those things that I just kicked the can down the road every time I tried to figure out how to do it in a non-clunky and sustainable manner. :-/