Hello!
I’ve noticed that the iTrack Precision analysis in parts of the track shows my line offset from the track… like sometimes going through the dirt.

anyone else? Know a fix?

thanks!

 

For Aim Matt has said the GPS data is accurate but the Google map itself isn't accurate to the same level so you get those sorts of offsets. If you overlay the data on various different companies' maps the line from the device bounces around as each map is a little different. Probably same issue here. Just a disconnect between the GPS data and what the online sat maps show.

 

Matt’s correct when he says AIM GPS is correct RELATIVE to OTHER AIM GPS plots, but not correct when he says Google (or MS or other) online maps are not accurate.

Sometimes (more often than not), AIM does line up. Other systems line up even better.

I would NEVER use PPTA or HLT without an additional GPS source like a VBOX Sport as the arbiter of where you are on an online map. Just relative to each other and as a heat map.

And yes, I’ve been working with GPS loggers since 2001. Every day.

 

Thanks guys, appreciated. I’m coming to the same conclusion Peter.

I find the Porsche Track Precision app to be fantastic for (1) tons of car data and (2) great video and (3) simple to use, and the CarPlay integration makes a great display on the PCM screen for seeing if I’m net positive vs negative time-wise. I really like it. And I can AirDrop the laps to my iPad for review in the paddock after the session.
BUT… the line data is always messed up, and rarely but every so often the CarPlay integration doesn’t work right (that’s ok).

So I’m pretty sure I need to have a secondary setup. Trying to decide between Apex Pro and Catalyst. I’m concerned I won’t adequately SEE the lights of the Apex Pro while in a turn, and I’m weary of expanding my workflow to a 2nd device for the Catalyst (and given that I do post session analysis, and generally can find opportunities myself, is the Catalyst worth the extra $$)….

 

So when I pull up the Bing map and the Google map why does the data point move? If the maps are "right" why don't they agree with each other?

 

The maps are not always accurate. A quick review of Google and peer reviewed literature reveal this if you don't believe me https://www.agsgis.com/Advanced-Mobi...Up_b_1062.html A nice one from Harvard where they breakdown the accuracy regions and found the accuracy is about 4.38 meters. https://ui.adsabs.harvard.edu/abs/20...1053W/abstract

Dig into the u-blox info and it gets a little more murky in how the coordinates all really work. I've dug pretty far into this with AiM and on the GNSS systems and have seen the good and bad of how they work. There are a few ways you can check out systems and see which performs better and under which conditions. It can be pretty revealing on how they use the GPS data nd interpolate between points.

 

Someone commented in another post that I made on this topic that I’d have better luck w/ a 10Hz GPS… I completely forgot about that option being supported by the app. I wonder how much better it’ll be w/ that.

Still, my 2 goals are:
- Improved location data for review
- real time feedback of some sort.

So I think I’m still going to need a secondary device.

 

Substantially better, for sure.

I know a number of folks very happy with the VBOX Sport.

It’s supported in the PPTA app and has been for awhile.

 

Thanks guys. I'm going to get a VBox Sport and see how that changes things. At the end of the day it seems the VBox also logs the data, so I can view it in Circuit Tools too.

 

Absolutely right. Best of both worlds. Helps PPTA, plus the added benefit of getting better info for viewing in Circuit Tools 2.

 

I think you'll have more success this route. Plus, as you say, you can use Circuit tools for your analysis.

Imho I consider using GPS for line analysis as a bonus. The consumer level GPS systems are getting better all the time but there's always a question mark, in my mind, about where the car really was on track ... especially if you go under a bridge or have your logger slightly offset to one side of the car or with poor "line of sight" to the sky.

Like you mention, lining up your driven line with references like Google maps should be amazingly useful - and often is - but often it is not. When it's not it can at best cause you to waste time trying to work out what's gone on, at worst miss lead you into showing you something you think you're doing that you're not - say missing an apex.

The VBox Sport is a good upgrade - I ran one in the past - and you should get better data than you are, but also be aware you might still get these strange misalignments - both on Google maps and even between your sessions during a track event.

I wrote a bit more about why this happens with GPS here... https://www.yourdatadriven.com/is-ra...-gps-any-good/ ...

... but in summary, the satellites your data system uses are moving relative to your position during the day.

You get better positional accuracy the more:
- satellites your device can see,
- evenly spaced the satellites are in the sky,
- above you the satellites are, verses just peaking over the horizon,
- clear and sunny a day it is.

There is not much you can do about any of that (!) so my approach has been to accept the GPS line isn't something I can completely rely on, and then think of other ways of getting the same - or better - insights on what line the car took and what opportunities you have to improve. You can use other methods, for example, using your video, the speed trace and gSum.

Couple of other thoughts:
1) Is a higher frequency GPS logger more accurate?
So I was confused by this at first so just in case this applies to you, the frequency of the GPS system doesn't imply better accuracy. The difference between a 1Hz or 10Hz GPS system is the number of times your data logger tries to work out where you are. A 1Hz logger will record once per second the current reading it has for your position. A 10Hz logger will record 10 times per second where it thinks you are. The benefit is that with more data points, you can hopefully get more confidence in the numbers - because there are more of them - plus, given the speed you'll be going, you will travel less distance between readings, and as such getting smoother - and more useful - looking data... particularly useful for looking at how you are coming off the throttle on to the brakes for example.

2) The Garmin Catalyst.
You mentioned the Garmin. Whilst this device does have some frustrations - we run one and the locked data is extremely annoying, for me anyway! - it is worth noting that it does something a bit clever and different regarding line analysis.

Instead of using just the GPS, the Garmin actually uses the video too, plus some clever software, to work out the edges of the track and then where you are in relation to those edges. It then shows you and overlay on what it things you did. It doesn't always work perfectly but - as an engineer - I'm still super impressed with the approach, and that they are even tackling this issue in this way.

Regarding your workflow comment, totally appreciate this as a concern. You don't want to be adding in more stuff to do. However, I have found the Garmin a bit of a game changer, in a good way, for our workflow - particularly during a busy test day. This is because it processes your session on the device, straight away, so no lap top required. One experience you might find interesting. We were testing recently. I was driving and came in for a tyre temp and pressure check. Whilst my buddy was doing that I sat in the car, reviewed what I'd just done, found a few areas I could improve - braking and line into T2 - and then, when my buddy finished with the tyres, I went straight out and knew what I was trying to do to improve. What got me was I was doing data and video analysis, still fully strapped into the car! - I didn't even have to take my helmet and gloves off!

My car is open so I've not used the coaching function but even without it I've been impressed by how easy the Garmin makes doing basic-during-the-day analysis i.e. helping me in the moment answer is one line faster than another, how well am I braking, what lap time is possible. For more in depth stuff, then, yes, I still need to get the laptop out, spend the time and do it properly but I just thought it worth mentioning that the Catalyst approaches helping drivers in a slightly different way. Appreciate it is not cheap though...

Hopefully this helps you clarify what you can expect a bit? Let us know how you get on!