Tegra X2 4K x 2K 60Hz Video Input


As Tegra X2 is capable of encoding video resolution of 4K x 2K 60Hz, we are keen to understand about the video front end (interface) used for video capture of 4K 60 fps on Tegra X2 SOM/Devkit during testing of 4K 60Hz encode.
Please share the details.


Hi Shareef,

The capture capabilities continues being the same as in Tegra X1, the main difference is that you have more master clocks coming out from the Tegra to your camera sensor in case of being needed, obviously several camera sensors use their own crystal oscillator to generate the clock, in those cases you don’t need the Tegra master clock.

Anyway, as summary you have 12 MIPI CSI channels, so you can use three 4 MIPI CSI inputs to grab 4Kx2K 60Hz. The interface is still the same in terms of software. You can read more about the differences here:


Are you using your own board? What is generating the image?


Hi David,

Thanks for posting your response.
We plan to order and use Tegra-X2 SOM/Devkit for our testing and would like to understand as to how Nvidia have actually tested 4K 60Hz video capture on TX2-SOM/Devkit?


Thanks David for the response …

Here is additional comments for your reference,

it states,
“Tegra TX2 can encode 4K60fps H265 while Tegra TX1 did only 30fps. Now the encoder is fully accelerated by hardware, on TX1 the encoding was hybrid.”

To test that, you could use 4K@60hz video clip as source for video encoder and see the output.

Following gstreamer pipeline commands will showcase 60FPS video encode using videotestsrc as input,
(details under https://developer.nvidia.com/embedded/dlc/l4t-accerated-gstreamer-guide-27-1)

gst-launch-1.0 videotestsrc num-buffers=300 ! “video/x-raw, width=(int)3840, height=(int)2160, format=(string)I420, framerate=(fraction)60/1” ! omxh264enc ! qtmux ! filesink location=test.mp4 -v -e

Hope this helps …