[SOLVED] v15 keeps crashing

Help on issues you run into with LapTimer; in case you have a question on how to use LapTimer, use the forum "Using LapTimer" instead
Saturnalian
10 or more Posts ★
10 or more Posts ★
Posts: 13
Joined: Tue Jan 17, 2012 6:31 am

Re: Tested V15 all weekend and some issues

Post by Saturnalian »

Harry wrote:
lmcgrew79 wrote:Harry, a few times this weekend the v15 seemed to have issues. Some sessions would not recorded data. I either go to assistant then racer or timer, then select video, at that point it says will start recording once the start finish line is passed. So everything seem to be working fine sometimes it would record the session with data, others i found later just seemed to record the whole video, which i found by looking thru the select video area. I didnt think it recorded anything but apparently it did, just no data.
I have chosen the quotation above, but there are several others that relate to the same issue.
It seems I have found the problem that created the the issues named:

Problem is that LapTimer crashes at the end of the second video recorded lap if the initial lap has been triggered by a stop/go (instead of go) trigger. Oddly, I did not have this case in my - or my beta tester's - test cases... And so we missed the bug... As this operation is probably a very frequently used one, this creates quite some issues currently. I have a fixed version available now and will submit it to Apple asap (after having run some additional tests). I will post this to the announcement section in parallel.

Can't say how sorry I am this made it to production. I will try to be as fast as possible.

- Harry
Hi Harry,

I spent two days at the track this week, gave up on using the Emprum ultimate GPS sometime ago but am finding great success with the Dual XGPS150 as far as staying connected to Laptimer goes.
I have posted before about the fact that i do motorcycle track days and can't be looking at HLT while it is running as it is forbidden to use lap timing devices at my track.

I started by trying to test out a revised Eastern Creek track set that have since sent you for certification.
Out of 12 x 20m min. sessions over the 2 days i managed to get Laptimer to trigger and record for 3 sessions only.
I don't use any video, there are no signal problems with the Dual XGPS.
it is quite disheartening to come back in from each session to find that Laptimer has shut down and not recorded anything.

I did manage to get the track tested on the first day and send it through to you for certification.
On the second day it was only the last 2 sessions that i managed to get it to trigger and record successfully, including the new lap splits and cancel lap trigger that i added into the track set.

Is it correct to say that if you have selected the driver and vehicle once in a day, if you know you have only 1 correct track set loaded already that the easiest thing to do is press start driving and go for it.
I don't feel that the startup protocol is totally clear in the documentation or processes of Laptimer, and when things don't work a few times the frustration can make you do silly things.

I have a few crash logs from the 2 days that i shall email to you
I apologise if this post is a little indirect but i believe in your product, have spent considerable $$$ trying different GPS dongle's to get one that worked for me, (XGPS is my third), and would really like to make sure that when i head out on track each time and press go that i will get some data at the end of it, particularly with the incredibly high cost of track days here.

Thanks very much,

Jamie
User avatar
Harry
Site Admin
Site Admin
Posts: 10518
Joined: Sun Sep 12, 2010 10:32 am
Location: Siegum, Germany
Contact:

Re: Tested V15 all weekend and some issues

Post by Harry »

Is it correct to say that if you have selected the driver and vehicle once in a day, if you know you have only 1 correct track set loaded already that the easiest thing to do is press start driving and go for it.
Yes, that's how it should be.
I have a few crash logs from the 2 days that i shall email to you
Yes please. I hope to have fixed the issues that occurred for you but certainly would like to make sure.

- Harry
Image Image Image Image
User avatar
Harry
Site Admin
Site Admin
Posts: 10518
Joined: Sun Sep 12, 2010 10:32 am
Location: Siegum, Germany
Contact:

Re: Tested V15 all weekend and some issues

Post by Harry »

I have posted a summary of issues with v15 here.
Good news is, all issues should be closed in the upcoming v15.0.3 bug fix release I have submitted to Apple already.

Thanks for your patience.

- Harry
Image Image Image Image
gplracerx
20 or more Posts ★★★
20 or more Posts ★★★
Posts: 836
Joined: Thu May 03, 2012 5:26 am
Location: Kingsport, TN USA

Database errors

Post by gplracerx »

It looks like the program is creating a new database error about every second. Every time I check the database for errors, some are found. The longer between checks, the more errors.

I've also had problems with the accuracy of the location of the start and finish in autocross mode. I can do the set start and finish, but if I then go to map mode for the course through the lap list, the position of the starting line is not correct.
User avatar
Harry
Site Admin
Site Admin
Posts: 10518
Joined: Sun Sep 12, 2010 10:32 am
Location: Siegum, Germany
Contact:

Re: Database errors

Post by Harry »

I'd assume your database is still corrupted leading to these follow on errors. Although LapTimer v15.0.2 is not able to fix all inconsistencies introduced in v15.0 and v15.0.1, it does not create new problems to my knowledge. Depending on the amount of data you have I'd suggest one of two approaches: 1) in case you have no valuable data, delete the app and install it again. 2) in case you have data you want to keep, send you databases to me and I will try to recover as much as possible. Please send and email in this case (and state your forum name so I get the connection :) ).

The second point is most probably due to a bad GPS position while setting the positions. Please check you have accuracy of 10m or better when setting the triggers. Furthermore, the map view is often shifted, do not expect absolute accuracy here.

- Harry
Image Image Image Image
gplracerx
20 or more Posts ★★★
20 or more Posts ★★★
Posts: 836
Joined: Thu May 03, 2012 5:26 am
Location: Kingsport, TN USA

Re: Database errors

Post by gplracerx »

I rebooted the iPod and the database corruption problem seems to have gone away. There were no new errors after the first check. I'll let you know if the problem happens again and try to define the conditions better.

The accuracy was reported to be 10 feet (3m) when the start line was first defined. And it wasn't just offset on the map. I had to edit the latitude and longitude to bring it back to the same physical location. I have the Emprum Ultimate GPS dongle attached to the iPod and it seemed to be working correctly. In Google Maps, the blue dot matches my physical location. I'll try it a few more times to see if the movement is consistent or random.
User avatar
Harry
Site Admin
Site Admin
Posts: 10518
Joined: Sun Sep 12, 2010 10:32 am
Location: Siegum, Germany
Contact:

Re: Database errors

Post by Harry »

Yes, in case you can reproduce this, let me know.

- Harry
Image Image Image Image
gplracerx
20 or more Posts ★★★
20 or more Posts ★★★
Posts: 836
Joined: Thu May 03, 2012 5:26 am
Location: Kingsport, TN USA

Re: Tested V15 all weekend and some issues

Post by gplracerx »

Harry,

I tried defining a start and finish line multiple times using the Autocross wizard on the Assistant page as well as defining POI's from the GPS page. I'll just list the seconds from the latitude and longitude as the degrees and minutes don't change: Before starting, the GPS page was reporting an accuracy of less than 10 m.
POI Lat Long.
start 19.0N 37.2W !!!
start 17.8N 41.2W !!!
start 17.0N 41.3W
start 17.0N 41.6W
start 17.0N 41.5W
start 17.0N 41.5W

finish 17.7N 38.5W !!!
finish 16.9N 41.7W !!!
finish 17.6N 41.1W
finish 17.8N 40.9W
finish 17.7N 41.1W
finish 17.6N 41.1W

Then using the button on the GPS page

start 17.0N 41.5W
start 16.9N 41.4W
start 17.0N 41.3W
start 16.9N 41.5W
start 17.0N 41.5W

finish 17.9N 41.2W
finish 17.9N 41.2W
finish 17.8N 40.9W
finish 17.9N 41.1W
finish 17.7N 41.0W

Then I cold rebooted the iPod, waited for the GPS dongle to be recognized and the accuracy to get below 10m:

start 16.8N 41.2W
start 17.1N 41.5W
start 17.1N 41.6W
start 17.0N 41.5W
start 16.8N 41.5W

finish 17.1N 40.7W
finish 17.8N 41.0W
finish 18.1N 40.9W
finish 17.9N 41.0W
finish 17.9N 41.0W

Even using the GPS page it looks like you should throw out the first reading and keep measuring until it becomes stable. But shouldn't that be done in the program? Since you're not moving when you define the start and finish, there's no reason not to take multiple readings and average them, preferrably testing for stability in the process. Is the autocross wizard somehow still relying on the iPod internal GPS?
User avatar
Harry
Site Admin
Site Admin
Posts: 10518
Joined: Sun Sep 12, 2010 10:32 am
Location: Siegum, Germany
Contact:

Re: Tested V15 all weekend and some issues

Post by Harry »

As long as an external sensor is connected, no function in LT accesses the iOS API. You are right, it would be well possible to run a kind of sanity check until the coordinate drift is gone - especially hat 5 Hz (at 1 Hz this would probably take to long to have an acceptable user experience). BTW, you should see the drift in the GPS view too - it displays the polar coordinates as you know.

- Harry
Image Image Image Image
gplracerx
20 or more Posts ★★★
20 or more Posts ★★★
Posts: 836
Joined: Thu May 03, 2012 5:26 am
Location: Kingsport, TN USA

Re: Tested V15 all weekend and some issues

Post by gplracerx »

I was just looking at the accuracy number and the speed. I would have thought that drift would factor into the accuracy as well, or is it just based on the measurement conditions like number of satellites, ground stations and signal strength? When fixing a point like the start and finish line, why not take readings for 5 seconds and calculate the mean and standard deviation and report that as the accuracy? Either that or show the distance to the point immediately after the measurement with an option to save or measure again rather than just a green check mark.
Post Reply