[aravis] Re: Debugging latency & long pause issues with Blackfly camera...
- From: Patrick Doyle <wpdster@xxxxxxxxx>
- To: aravis@xxxxxxxxxxxxx
- Date: Tue, 15 Apr 2014 10:17:48 -0400
On Tue, Apr 15, 2014 at 9:52 AM, Emmanuel Pacaud <emmanuel@xxxxxxxxx> wrote:
> Hi,
> The problem with the blackfly camera I own (a BFLY-PGE-14S2C-CS) is the
> camera timestamp is wrong. The camera send in the image packet a
> timestamp expressed as timing ticks. Aravis then use the content of the
> register TimingTickFrequency in order to compute the timestamp in
> nanosecond. Unfortunately, TimingTickFrequency register returns a wrong
> value, and then the timestamp in nanosecond is also wrong.
>
I notice in arvgvdevice.c that you read the timestamp tick frequency
as 2 32-bit numbers (a high & a low).
But I notice that arv-tool-0.4 reports a GevTimestampTickFrequency as
a single 32-bit integer for my Blackfly camera (a BFLY-PGE-03S2M-CS).
I was going to go investigate this further -- what is it that you do
when you call arv_device_read_register() for those two 32-bit
registers, and how does that differ from reading the 32-bit integer
through whatever interface arv-tool-0.4 uses.
I'll also go play with arv-camera-test once I get back in front of the hardware.
--wpd
Other related posts:
