DTM works with this strange addition

[Thomas]

I used the sample code from nordic with nRF51822. DTM works fine. sensibility test gives me 0% on the 3 channels. Perfect.

Then I copied and paste into my code, calling it directly after main. The test gives me other results !! I don't understand.

But if I have this code

// Same as nordic except for the 4 lines I add at the end
for (;;)
{
    // Will return every timeout, 625 us.
    current_time = dtm_wait();  

    if (NRF_UART0->EVENTS_RXDRDY == 0)
    {
        // Nothing read from the UART.
        continue;
    }
    NRF_UART0->EVENTS_RXDRDY = 0;
    rx_byte                  = (uint8_t)NRF_UART0->RXD;

    // ////////////////////////////////////////////////////////////
    // ////////////////////////////////////////////////////////////
    // THESE 4 LINES ARE THE LINES I ADD
    // ////////////////////////////////////////////////////////////
    // ////////////////////////////////////////////////////////////
    if((rx_byte == 0x61) && (!is_msb_read))
    {
        uart_put_char16(0x4F4B);
        continue;
    }
  ...

So if I do this, the test works again (0%, 0%, 0%). If instead of 0x61 I do 0x62, it works too but (0.5%, 0%, 0%). If instead of 0x4F4B I do 0x1234, similar result : almost 0% but not.

So I am confused. First I don't know why my copy and paste code doesn't give the same result as nordic, and second I don't know why adding my uart_put_char16 makes the test work better than nordic sample code.

Any ideas ? Thanks

[Thomas]

I don't know what it is, it is one of the few tests done by the anritsu machine. 0% 0% 0% means it is perfect