rjdmoore

Hey, ok so camera id is probably a poor name. It's actually the USB id (nth device on bus). If you only have one usb device connected, src_fn should be...

Hmm, ok. Basler cameras are a little poorly tested unfortunately, so there could be an API change in the newer versions that's causing an issue. Basler code was originally added...

Hmm, ok good debugging. Device 5 with SDK looks the way to go. It also looks unfortunately like it's an SDK problem now. It's not something I'm going to be...

Hmm sounds odd. The Basler code was contributed by another user but has been working before. I will have to look into it. I don't have a Basler cam right...

The problem is that opencv get() returns 0 if the device doesn't support the property (POS_MSECS). See [here](https://docs.opencv.org/3.4/d8/dfe/classcv_1_1VideoCapture.html#aa6480e6972ef4c00d74814ec841a2939). So fictrac's _timestamp would just be zero for ever in that case....

> Thanks for your quick response and thanks for all your work in > My two cents is that it would be easier to just have two columns in the...

> Btw, why is `ms_since_midnight` defined for `PGR_USB3` > > https://github.com/rjdmoore/fictrac/blob/9ac055e52d89f49f492a8eb4e1f7c5b8cd6df40a/src/PGRSource.cpp#L236-L244 > > but not for `PGR_USB2` > > https://github.com/rjdmoore/fictrac/blob/9ac055e52d89f49f492a8eb4e1f7c5b8cd6df40a/src/PGRSource.cpp#L292-L300 > > ? This looks suspiciously like a bug. Will...

Yeah, I have changed nameserver for the website - it should be back up again soon. cheers, Richard On Thu, 11 Apr 2024, 19:12 Heberto Mayorquin, ***@***.***> wrote: > As...

Hey, thanks for the report - I've changed slightly the way timestamps are read from the camera ([diff](https://github.com/rjdmoore/fictrac/compare/5b8d022..016d92c)) Hopefully this resolves this issue.

@sgherbst are you able to pull the latest develop branch and test whether this fixes the issue?