This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

DS90UB960-Q1: DS90UB960 two channels receive video at different frame rates

Part Number: DS90UB960-Q1

Hi Team,

Our Customer have some questions about the DS90UB960-Q1.

1.Two CAM and different frame rates send the video to the DS90UB960.

As shown below, is this method of application possible? 

2.As shown below, Two Cam video input to Deserilizer, and the Deserilizer ouput RAW and YUV on a lane.

As shown below, is this method of application possible? 

Thanks,

Xiaoxiang

  • Hello Xiaoxiang,

    1. The first configuration is possible. However, customer will not be able to use any of the Synchronized Forwarding methods in the 960. They will need to program the 960 to use Round Robin Forwarding.
    2. Yes, the second scenario is possible. The 954/936 does not care what data type is in the CSI-2 packets it is receiving. FPD-Link devices will simply act as a bridge that forwards the CSI-2 packets it receives at an RX Port to a specific TX Port. 
      1. The only exception is if you connect the 954/936 deserializer to a DVP serializer, such as the 913A/933. In this case, the 954/936 deserializer needs to be programmed to package the received parallel video data into CSI-2 packets with a defined data type.
      2. But if you are connected to a serializer with a CSI-2 interface, then the deserializer does not need to do anything with the data type of the received CSI-2 packets.

    Another thing to be careful of is the supported video throughput rate. The serializer supports up to a certain amount of video throughput, which can be calculated below:

    Video Throughput = HActive x VActive x FPS x Bits-Per-Pixel x Overhead (Assume 25% if unknown)

    For example, the 953 supports up to 3.328Gbps of video throughput but the 935 supports up to 2.528Gbps of video throughput. Make sure the selected serializer can support the video throughput that is being inputted to it by the connected camera sensor.

    After that, make sure the deserializer is capable of outputting all of the video data it receives faster than the rate at which video is being inputted. Here is a short video that demonstrates the concept:

    https://training.ti.com/csi-2-aggregation-breakdown?context=1134310-1139225-1148220

    I also have an internal calculator tool that can help with calculations, so please let me know if you need support with calculations.

    Best,

    Justin Phan