iPhone Apps for the track
04-22-2012, 09:50 AM
#41
GPS is *barely adequate* for basic road racing car position and speed, but is totally inadequate for driver analysis or instruction (I can't state this strongly enough!). The level of error per sample is so huge (*documented* as 10 meters per sample when in motion -- that's 32.8 feet per sample!) that there's *no way* to do *any* detailed analysis. Even if the error was far less than that, the limitations of physics mean that the GPS signal traveling at the speed of light (299,792,458 meters per second) means that if you sample every thousandth of a second means that the signal travels 299,792 meters (that is your accuracy). Millisecond sampling (0.000001 time between samples) means 299.8 meters traveled. Nanosecond sampling (0.000000001 time between samples, which is *far* faster than most *desktop* computers are capable of today) means that you could get to .3 meters per second -- or only about 1 foot positional accuracy!
GPS does have some benefits -- speed can be calculated fairly accurately for our uses, and if you don't look *too* closely at the position (looking at the whole track instead of a single corner) you can get a fairly accurate picture of the track.
GPS was originally designed to hit a specific city block. The newer GPS is designed to hit a specific room within a building. That is still FAR from accurate enough to determine whether you hit the apex or were 10 feet from the apex, much less whether you were close to a cone or not.
04-22-2012, 01:37 PM
#42
Autocross doesn't have the same track configuration each time you're at the event (something that most people -- especially hardware manufacturers -- fail to realize). So, even if you're at the same autocross site, you'll still have to put in the current track configuration *each time you're at the site!*
GPS is *barely adequate* for basic road racing car position and speed, but is totally inadequate for driver analysis or instruction (I can't state this strongly enough!). The level of error per sample is so huge (*documented* as 10 meters per sample when in motion -- that's 32.8 feet per sample!) that there's *no way* to do *any* detailed analysis. Even if the error was far less than that, the limitations of physics mean that the GPS signal traveling at the speed of light (299,792,458 meters per second) means that if you sample every thousandth of a second means that the signal travels 299,792 meters (that is your accuracy). Millisecond sampling (0.000001 time between samples) means 299.8 meters traveled. Nanosecond sampling (0.000000001 time between samples, which is *far* faster than most *desktop* computers are capable of today) means that you could get to .3 meters per second -- or only about 1 foot positional accuracy!
GPS does have some benefits -- speed can be calculated fairly accurately for our uses, and if you don't look *too* closely at the position (looking at the whole track instead of a single corner) you can get a fairly accurate picture of the track.
GPS was originally designed to hit a specific city block. The newer GPS is designed to hit a specific room within a building. That is still FAR from accurate enough to determine whether you hit the apex or were 10 feet from the apex, much less whether you were close to a cone or not.
GPS is *barely adequate* for basic road racing car position and speed, but is totally inadequate for driver analysis or instruction (I can't state this strongly enough!). The level of error per sample is so huge (*documented* as 10 meters per sample when in motion -- that's 32.8 feet per sample!) that there's *no way* to do *any* detailed analysis. Even if the error was far less than that, the limitations of physics mean that the GPS signal traveling at the speed of light (299,792,458 meters per second) means that if you sample every thousandth of a second means that the signal travels 299,792 meters (that is your accuracy). Millisecond sampling (0.000001 time between samples) means 299.8 meters traveled. Nanosecond sampling (0.000000001 time between samples, which is *far* faster than most *desktop* computers are capable of today) means that you could get to .3 meters per second -- or only about 1 foot positional accuracy!
GPS does have some benefits -- speed can be calculated fairly accurately for our uses, and if you don't look *too* closely at the position (looking at the whole track instead of a single corner) you can get a fairly accurate picture of the track.
GPS was originally designed to hit a specific city block. The newer GPS is designed to hit a specific room within a building. That is still FAR from accurate enough to determine whether you hit the apex or were 10 feet from the apex, much less whether you were close to a cone or not.
04-22-2012, 04:14 PM
#43
I just purchased Harry's. Will take it to the track for the first time on Friday just using the iPhone's GPS. I understand it won't be all the accurate but I don't need that as a beginner. I just want to have a rough idea what my speeds/times are and how consistent I am.
The video overlay feature is nice but the iPhone has some limitations for video, primarily memory if you have the 16GB version. It also takes twice as long as your recording time to render the overlay. So 20 minutes on track needs 40 minutes to render....as I understand it.
You can use the data from Harry's to overlay on video gathered from another camera, like a GoPro. However, to do so you have to import the video into the iPhone and then execute an overlay. Again, slow processing and not much storage capacity. The second way is to purchase software to merge your data from Harry's and your video. Some of these are pricey and as far as I know none of them works in OS X.
These are all limitations of the iPhone, not Harry's Laptimer.
There is a list on Harry's site that documents GPS add ons that work with his program. Not all will. The ones he recommends are not expensive starting at around $90.
My only complaint with Harry's Laptimer......prior to ever using it on the track....is that the documentation is not very clear. He needs someone to write an manual for him. It is a very full featured, complicated app that can do so many things for so little money but a concise, complete, well organized manual would ease the pain and decrease the amount of time it takes to figure it all out.
He has a forum for his products and answer many if not most of the questions himself. So the support is excellent.
The video overlay feature is nice but the iPhone has some limitations for video, primarily memory if you have the 16GB version. It also takes twice as long as your recording time to render the overlay. So 20 minutes on track needs 40 minutes to render....as I understand it.
You can use the data from Harry's to overlay on video gathered from another camera, like a GoPro. However, to do so you have to import the video into the iPhone and then execute an overlay. Again, slow processing and not much storage capacity. The second way is to purchase software to merge your data from Harry's and your video. Some of these are pricey and as far as I know none of them works in OS X.
These are all limitations of the iPhone, not Harry's Laptimer.
There is a list on Harry's site that documents GPS add ons that work with his program. Not all will. The ones he recommends are not expensive starting at around $90.
My only complaint with Harry's Laptimer......prior to ever using it on the track....is that the documentation is not very clear. He needs someone to write an manual for him. It is a very full featured, complicated app that can do so many things for so little money but a concise, complete, well organized manual would ease the pain and decrease the amount of time it takes to figure it all out.
He has a forum for his products and answer many if not most of the questions himself. So the support is excellent.
04-23-2012, 07:52 AM
#44
I have a the Race Logic performance box for data: http://www.performancebox.co.uk/
I can use that for predictive lap times and data analysis; I believe it samples at 10Hz.
I'm just looking for a clear/cheap/easy way to share video with the speed/track overlay/etc., so I can spend time on crunching data from the Performance Box instead of fussing around with video processing.
I can use that for predictive lap times and data analysis; I believe it samples at 10Hz.
I'm just looking for a clear/cheap/easy way to share video with the speed/track overlay/etc., so I can spend time on crunching data from the Performance Box instead of fussing around with video processing.
Thread
Thread Starter
Forum
Replies
Last Post
Lt_Ballzacki
Brakes & Suspension
39
08-06-2021 06:19 AM
