Improve InputLoopThrough latency

Hi everyone,

I'm currently using the Decklink SDI 4K card and running the InputLoopThrough example based on Playback Scheduling in the Windows SDK. With this approach, I'm able to get a 50 ms latency (3 frames * 60Hz), I was wondering how can I reduce this latency?

By searching on Google, I found some flags such as bmdDeckLinkConfigLowLatencyVideoOutput but it doesn't seem to be supported anymore.

Thanks in advance!

[Cameron Nichols]

Hi Timothee,

Two considerations that will impact latency:

• Your preroll size for your scheduled output
• Whether you are locking to a common reference?

Regards
Cameron

Hi Cameron,

thanks to your quick reply.

Is it possible to tweak the preroll size? There's

Code: Select all

SetPreroll (uint32_t prerollSeconds);

in the SDK but I don't think it's the right function.

Whether you are locking to a common reference?

I'm sorry, what do you mean?

Thanks!

[Cameron Nichols]

Hi Timothee,

SetPreroll is a method in the IDeckLinkDeckControl interface. The preroll is the number of video frames you have scheduled with IDeckLinkOutput::ScheduleVideoFrame prior to starting playback with IDeckLinkOutput::StartScheduledPlayback. The larger the preroll, the less likely for late or dropped frames, particularly if your video pipeline has time variance in providing output frames, but it will increase the overall latency.

By common reference, are you connecting a genlock signal to both the DeckLink SDK 4K reference input, and also your input source?

Regards
Cameron