BYB Tech Releases Chrono Timing System

 Hi Enrico,
This is a nice looking product, and I'm very much interested. Anyone who has used Strava can confirm that it is not suitable for accurate timing. GoPro is good for accurate timing when you pass certain parts of the trail but this is a manual process of going through frame by frame.
I totally understand wanting to overlay the speed profile for two laps. I have a few questions:
1) Do you blend together wheel speed and GPS data into a single channel?
2) Can you export as .csv to overlay in external software?
3) Is the accelerometer data included in the data set? e.g. can you use this to align the data (in motorsport they use track bumps to align overlays to a specific bump/kerb on the track)

 @rojo-1: Hello rojo-1, thanks for your questions.
1) The GPS is always present (since it's coming from the GPS), but you can decide to exploit the built-in speed sensor (thanks to a magnet placed on the wheel) to get the real distance traveled. This is very important to get accurate comparisons.
2) We're considering letting you load the Chrono files into our BYB Telemetry PC software. In this case, it will be possible to export the raw data.
3) Yes, the IMU data (3-axis gyro and 3-axis accelerometer) are included in the data set. You can consult the charts or see the shocks directly on the route (the route is changing colors according to the speed carried or the shocks perceived).

We're always available to reply to your questions also here: info@bybtech.it

 @rojo-1: if you have a decent jump on the track, that would be pretty easy to align with as probably the only low variance -1g z, 0g x and y section

 @RodEnry:
Would the system also work with more than two triggers for precise split times?

(Though GPS should already give al ot of detailed insights)

 Considering the system has accelerometers, does this imply you can also read out the acceleration data? I'm not even that interested in lap times perse, but more in where I'm accelerating or decelerating (for instance to judge whether I'm gaining speed in my corners or whether I'm stalling). Obviously the acceleration in corners is different depending on where on the bike you attach the sensor. But if I keep it attached in one place, I should be able to compare my efforts when sessioning a corner. Does it work like that?

 Or you could just use Freelap

 @Janne88: at the moment with two, but it's possible to use more. Just an app update to add this...

 @Life1: Well, it's missing some things... and it's even more expensive

 @vinay: yes, you can compare runs and see the IMU lines. I'd recommend using the built-in comparison, which provides the delta time chart. This chart shows you in which turn/part/section you are going slow/fast. For example, you can pick a faster friend and overlay the two runs. You'll be able to spot where your friend is going faster.

 @RodEnry: would you have some good screenshots/video showing how it would look in the app comparing the accelerometer data between 2 runs in the same section (eg comparing your run to a faster friend run)?

 Can it be used with apps like Litpro to compare lap times with others using Litpro?

 @RodEnry: How short of a open lap can you create? For instance if you want to work on one technical section, and figure out the line etc your lap may only be 5 seconds. Is this accurate enough for such a short track? Assuming using the dual triggers.

 At 300 euros a pop, thats a helluva race cost outlay. Zone4 rents for $3 per chip and $125 for each timing point.

 Yeah - I think that's fair. I imagine it'll eventually be integrated with their data acquisition app for an all-in-one resource for racing.

 Anyone who pushed strava seriously know how innacurate is.. depending on the brand model and many other factors there is a 5sec innacuracy

 @PauRexs: So that's not a "strava" inaccuracy, it's a consumer-level fitness GPS tracker inaccuracy, right?

 @pmhobson: is strava available on any devices with higher levels of accuracy? seems like kind of a moot point

 @KolaPanda: Suunto has the most accurate gps watches I’ve found and paired with Strava I don’t notice any difference. Then I don’t actually run the Strava app I just upload from my watch. Not saying you’re right or wrong but I am saying that Suunto is better than the rest lol

 @KolaPanda: Yeah -- the recording device absolutely matters. There are different satellite constellations available, but older devices don't have the hardware to access them.

Strava doesn't record any data. The device does. You could upload the GPX (or equivalent files) for the device in the article to Strava if you wanted.

 I have worked with a lot of different types of GPS units over the years. I’ve used traditional RTK and non-RTK (LIDAR positioning systems and theodolites/total stations) for various government contracts. The kind of accuracy needed for Strava (3 dimensions is the problem with triangulation - X,Y horizontal accuracy is not the problem) is not possible with smart watches etc. You need better equipment and that combined with trails often being in heavy canopy cover and the rider moving relatively quickly is going to produce quite large errors for traditional RTK devices.

 @pmhobson:
Strava's interpretation of your ride data is only as good as their algorithms AND the quality of the raw data provided by your phone or GPS tracker. So it's typically gross assumptions based on minimal data of dubious value. After more than a decade on Strava, I'm finally seeing a couple of trends - I'm getting slower and buying less bikes.

 @a-prince: RTK = Real Time Kinematic Positioning (pro tip: you are explaining technical concepts to an uneducated audience - do not use unfamiliar acronyms)

 @ShawMac: As someone who has had the misfortune to work at races where the governing body mandated using zone4 I can safely say you are very wrong.

 @blueH2Oj: I second that. I use a Suunto GPS watch too. The accuracy of their watches is scary. Not sure why more riders don't use them.

 @rpb10276: how do you know they’re accurate?

 @Super7: I won't argue. I was just going off the cost estimates on the Zone4 website for the basic rental. I gather there are some serious hidden costs then.

 isnt the 5s delay to do with satelites?

 I as an average rider use my bike computer. It has timing, I start at a certain time, usually the full minute, and check at the end of the timed segment. It's simple, accurate and there's no point in comparing to other people's times anyway as there are to many variables. Strava sucks for timing, I used it and was amazed time over time at how inaccurate it is oftentimes.

 @naptime: radio waves are traveling at the speed of light. A 5-sec delay would come from the ping rate of the device (i.e., phone, head unit, watch)

 Strava isn't accurate. I found the best timed training device is a just a gopro. Drop the clips into DaVinci Resolve, put the timer on with the start and end on the same frame and you have accuracy down to .001 of a second.

 Yeah, the main thing here is that you can pair with two remote triggers (transponders), one at the start and one at the finish. The trigger time is real and accurate.

By the way, the GPS of Chrono is not comparable to a smartphone's or ciclo-computer's. It's very expensive, 25Hz refresh rate, and multi-constellation. You can clearly see that classic ciclo-computer GPS lines are usually "broken and very segmented". Chrono, even at high speed provides you smooth lines.

 @a-prince: you could argue Strava and smart watches give data in real time, but technically RTK is a specific method using a rover and a base station. But yes, the tree canopy is the biggest weakness in Strava and gps watch accuracy. Most people seem to just think cos they paid $500 for a watch, the answers it gives is always right.

 @Joe89: Yes a rover and a base is what I’m referring to as “traditional RTK” systems. The correction stream can be broadcast with LTE/IP address or radio waves depending on your needs.

 @pmhobson: yes. Use an iphone to drop a few seconds off a Strava time. They always give faster times than a Garmin.

 @lelandjt: also, if you want to cover more vertical distance use measurement via GPS instead of barometric (which is flawed, too).

 Stava is about as accurate as counting in your head

 @ShawMac: You're right about the hardware it's s pretty solid but the software is a joke. I would love for my them to get bought or partner with another company to get that sorted.

 So according to Strava I used to be the KOM on lower a-line for years, just ahead of this Danny Hart fella. Good enough for me! Also, strava is just a bad bang for your buck for mountain biking.

 @Super7: Maybe shit software so you have to pay for the premium support staff to keep it running

 @blueH2Oj: I also use Suunto watches (Ambit 2S, Ambit 3 Peak, Spartan and currently the regular 5, not the peak version). I like them for their durability (even Casio G-Shock falls apart on my wrists) but I'm not too sure about the GPS accuracy or whether I should even expect perfect accuracy. It is nice for training but the autolap function I now have on the 5 definitely doesn't click in the very exact same location. And as said, I don't even think I should expect that. If I would need that, a beacon can't be beat. So, the product being mentioned here is great for doing just that. Then of course a Suunto watch does a lot more that this device doesn't so both have their place.

 @KolaPanda: Just out of curiosity, I've ridden laps with my phone hooked up to the $300 "GPS booster" the local trail nonprofit has for trail work and laying geotracks in Avenza. If you do multiple laps, the lines are still not completely on top of each other, and sometimes it's shocking how much variance you still get. I'm not sure there is anything that makes sense from a consumer perspective that would make Strava accurate enough for it to do anything besides telling you whether you are in top 5%, top 10%, etc. (and tempting you to act like a douche bag).

 @dancingwithmyself: I have no experience with Strava but it seems to me like it is a piece of software that just has to do with whatever you feed it. Garbage in, garbage out. I think I can connect it to the Suunto software (which obviously gets the input from my Suunto watch) and the Strava log should be as accurate as my Suunto log. Or the other way around, Suunto had their "3" watch which used the GPS hardware from the phone so the Suunto GPS log would be as accurate as whatever the phone could log.

I think there is a difference how watches and phones construct the path you traveled. My Suunto watch uses GPS data in combination with acceleration data to fill in the gaps. Phones also use GPS data but in combination with known networks in the area. Not sure how much that would help with training data but at least this is something I gathered from the phone manual. It does have acceleration sensors on board but I'm not sure whether it also uses these to contruct where and how fast you're going. A car navigation system (I'm using TomTom) tries to lock onto the nearest or most obvious road.

Either way, I'm no expert on the subject but what I'm trying to say is that along with the logged GPS data many (or all) devices use other technologies to fill the gaps. These could be accurate but it could also be that the end result appears accurate (that is, smooth) but in reality it isn't.

 @ShawMac: Yep. You can do it yourself but they are happy to take your money to provide support over the phone or better yet onsite staff.

 The more data you have, the more inaccuracies you see. It feels like "the more you know, the less you know".

I use strava mostly to track maintenance intervals.

 @RonSauce: Feeds into my ProBikeGarage app for the same reasons (tracking maintenance on five bikes is brutal).

 @canadaka May be a relatively easy feature for trailforks?
Or do I have to dust off my old app project

 @MrDuck: I don't use Trailforks,but will check it out...
But tell me more about that project of yours!

 But do you know how slow you’re getting???

 @lotsofpasta: I'm the one with the truck, so the homies are forced to wait before getting the next lap.

 This is what I've been searching for! Do you have plans/description of how to do this?

 @phillyforester: I don't have anything written or photos because it was many years ago. Anyway, I'll summarize what I did. I had taken a normal bike speedometer, I opened it, on the two contacts of the button I soldered a two-wire cable to the end of which I soldered a button. As a support I used a v-brake lever to which I made a hole and fixed the button.

 @phillyforester: look for finger stopwatch on amazon- its a little simple one button watch that can strap to a bar pretty simply- like $30 or so

 @jhilly: Definitely an option and in fact, I've got one of those. It works reasonably well, but it's a bummer do do a run where you're going 10/10ths, and all the risk that entails, and find that you didn't hit the start button right.

 @blacktea: Thank you - sounds easy enough and if I screw it up, the parts aren't too expensive. Should be a perfect weekend project!

 @phillyforester: Yes it is very simple if you are familiar with a soldering iron. Unfortunately I moved house and it is almost impossible for me to find this chronometer that I had modified

 Riding with someone who cares about strava sucks alot of the fun out of a ride.

 @RonSauce: I'll never forget the jackass behind me yelling "STRAVA STRAVA STRAVA" trying to get a group out of his way. It made me wanna drag my brakes on a climb. I honestly don't mind people competing with themselves or even each other, but doing it in the middle of the day on a crowded trail and expecting every trail user to get out of your way on a multiuse trail is just dumb.

 @shinook: I always happen to fall down when that happens! And right across the trail blocking the way, it's so odd... But no one offers to help, and perhaps that's even odder. They just get mad, even though I'm the one on the floor.

It also happens if they yell "training", or almost anything that implies that whatever they are doing is more important than everyone else on the trail.

 There’s no rule. Depends on what gps receiver your using, what satellites can it use for tracking, where your using it. Sky view to satellites. If you put the gps on top of your helmet and you ride in the desert. It’s going to give you pretty good results as opposed to a 15 second portion of a-line with heaps of tree cover.

 Well thats not a simple yes or no question: GPS actually has an embeded signal for time keeping called 1PPS (1 pulse per second) which depends on atomic clocks installed on the satellites that send a signal 1 second appart with great accuracy counting seconds from epoch (chosen as 1980) from which you can calculate exactly what time it is. But that does not really convey to GPS accuracy for a timing system where you have to measure time elapsed between two points in space and time. GPS positioning is in principle extremely accurate as it actually uses Einsteins theory of special relativity to account for the relative motion of all the bodies involved resulting in a theoretical accuracy of centimeters with satellites being positioned about 20.000 km above the earth which shows how insane piece of engineering we actually take for granted. But the real life measurements are not nearly that accurate due to many factors like very small consumer grade antennas not being that great, atmosphere/clouds/trees/... obscuring the signal or mountains completely blocking it and if you mount a receiver on a moving object (like human, bike, car,..) it will constantly change direction losing the lock on the satellites firther decreasing accuracy (lock on more satellites means better accuracy). Years ago in early days of smartphones I was always riding faster than my friend but he would get all of KOMs thanks to his crapy Sony and the differences were even bigger than 5s on one of our twisty trails. So its really hard to accurately estimate gps error and its use as timing depends on the case, it is comonly used for extra long races where its not really likely two athlets would end up in the same second, but not so in fast racing like DH or XCO. How do I know all this? Studied some back in my university and I'm actually building my own timing system which uses previously mentioned 1pps to sync the start/end clocks

 @winko: why do you want to incorporate gps? What’s wrong with just a the chip system? This so you can run it though Strava?

 @Joe89: The GPS can be useful for both seeing the track and basic, less-precise, timing without putting out the triggers. Start doing laps with just the GPS, make big obviously different line choices, narrow down the selection, then set up the triggers for the next session to tweak the chosen lines. Combined with the wheel sensor, you can also get a pretty decent look at where you lost or gained speed, and can use that to pick and choose lines per section to get the lowest overall lap time. Just knowing the overall time doesn't help you figure out which permutations of line choices actually give the best potential time, unless you ride every permutation, which you just won't have time for.

 @Joe89: System runs on raspberry pi and they use software clock, which is not really great as it only counts time well when processor is idle and is not really the case if it also reads data from the chip reader and saves the results somewhere. I started the project on pi zero which is single core, switched to pi zero 2 since which has 4 cores and I'm not sure if this would suffice, it just might, but it is also an amazing opportunity to learn and dig a bit into the GPS and I find it really interesting, plus 1PPS GPS receiver costs like $10. It would for sure be good enough to just use external temperature adjusted oscillator clock but where's the fun in that... I also get hints from a friend who works at the company that build high precision instruments that NASA and companies alike around the world use so it could be made simpler but so far I find it interesting and that what really counts, first such project that I've started by myself and the actual cost to get nano second measurements is surprisingly cheap nowadays so why not.

 @winko: You need a crystal (with its pF capacitors) to keep accurate timing, plus a good GPS module.
Actually, 4 cores are not needed.
You need a proper antenna, proper RF matching, robust code, robust hardware, and a lot of other things...


Chrono is working very well in the woods, but for the best timing experience, you can always use the remote transponders. You can still compare routes and draw custom waypoints (based on GPS this time).

 @RodEnry: again that depends on your system but linux for example already has tools available which sync its clock to the 1pps signal. I use cheap receiver and it gives me sub 100ns accuracy inside the house. What I meant with multiple cores is that one would be reserved for software clock but I dont know if it works like that and you can actually set a dedicated core as a clock, never investigated that but my friend would probably kill me just by thinking about it...

 Obvious pass I already know I’m slow

 Strava as such is not enough, you also need a strong believe that GPS timing actually works accurately.

 @Muckal: then this system is no better

 @Muckal: why?

 @grizor: It depends, I'd be willing to bet it has a better gps receiver compared to sports watch or mobile phone so it should work better in open space, but if you use it in hilly terrain with limited sky coverage (as we mountainbikers often do) it will be flawed by fundamental gps principe of accuracy depending on number of receiving satellites.

 @RonSauce: because it doesn't. It obviously depends on the quality of the device, but generally available GPS tracking devices like watches or phones are not accurate enough to measure seconds in mountainous, forestal terrain. Back when I used Strava I could have had a KOM by a long way without even riding that particular trail that day. I rode down the one next to it, about 10m further down the hillside which ends about 20m to the right of the one i supposedly crushed everybody on.

 @Muckal: I agree - on one of our local trails the KOM are held by XC skiers, on skis (and apparently doing 65 kmph on 5% gradient technical single track), in winter and somehow Strava cannot differentiate that data set from the mountain bike data set? Strava has some uses but precision timing and ride comparison is not accurate.

 @andrewbikeguide: that’s the users using the wrong activity type

 Sorry everyone, just to be clear I was joking. Strava is fun but no way to time a run if you actually need some accuracy. Leave it to Pinkbike to turn everything into a life or death debate.

 No external triggers on the Garmin. Not accurate enough to get the start and end points of a track correct.