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.

TMDS171: Color shift

Part Number: TMDS171

Hello.

We are using TMDS171.
I have a question.
If a DVI signal is input to the TMDS 171, It causes color  shift.

Q1 Have you ever experienced such a color shift phenomenon?
  Please let me know if there is any countermeasure.
Q2 Is there anything easy to do, such as register setting with a solution?

please see the attached document.

Thank you.

TMDS171.xlsx

  • Hello Masazumi

    The TMDS171 is only a redriver/retimer, it doesn't deal with the information transmitted.
    Can you confirm if the swap/polarity of lanes are correct? This option only works in retimer mode. Can you share the status of DEV_FUNC_MODE and your data rate?
    Please, perform a double check in the retimer configuration.

    Regards
    Francisco
  • Hello.
    Francisco-san.

    thank you for your answer.

    Can you confirm if the swap/polarity of lanes are correct? This option only works in retimer mode
    →Lane polarity is Disable.

    Can you share the status of DEV_FUNC_MODE and your data rate?
    →EVM is used as initial value.
    Register setting is not changed.
    Attach register setting value read with EyeScan Tool
    TMDS Clock:134.25MHz

    thank you.

    TMDS171.pdf

  • Hello Ishii-san,

    Does HPD toggle low to high when the resolution is changed?

    Regards,
    JMMN
  • Hello.
    Francisco-san.

    The situation of our board is as follows.

    HPD_SNK is not used (not controlled). It is fixed to High.
    (EVM is also the same.)

    Also, we are using the board which is being used with HPD_SRC = Open.
    EVM, J1 2-3 is short.

    Please refer to attachment

    Do you understand something about cause and measures of color shift?

    Thanking you.

    TMDS171-2.pdfbbbbb

  • Hello Masazumi

    Sorry for my delay in it.
    When you say "when specific signal is input, color misregistration occurs rarely" what do you exactly mean? it means that with other kinds of input you don't have any problem with the same configuration?
    I would like to confirm if you see the same behavior in both cases (our EVM and your app)? If not, can you share the schematic?
    Can you share the status of EVM's jumpers?
    when you say "Writing "1" to the 0x0A[2]APPLY_RXTX_CHANGES bit works normally" what do you exactly mean? the effect disappears?
    this one only happens in retimer mode?
    feel free to share all the information that you have.

    Regards
    Francisco
  • Hello. Francisco-san

    The environment in which TMDS 171 performs color shift is as shown in the attached document.
    It is happening using TMDS 171 EVM.

    I am using astro VG - 870.
    Do you have a VG-870?

    Please check the attached document for details.

    thank you.

    TMDS171sys.pdf

  • Hello Masazumi

    I am checking your information, do you mean that the solution to your problem is set 1 to APPLY_RXTX_CHANGES? We don't have the same signal generator, but we have another signal generator. I can see in your diagram that you are receiving HDMI to the EVM, is that correct?
    Can you please explain how to change to from retimer mode to redriver mode?
    I've did a simple test with a source (computer) to sink (monitor) and I can't see any problem with HDMI signals. Have you tested this system with an HDMI source? if so, do you have the same problem?

    Regards
    Francisco
  • Hello Masazumi

    Do you have feedback?

    Regards
    Francisco
  • Hello.

    Francisco-san.

    I'm sorry for the late reply.
    I talked with the customers at the end where phenomena are occurring this time.

    In the summer of this year, due to another problem,
    we sent measuring equipment of TMDS171 to your company and got the cause investigated.
    (TO:Julie Nirchi-san.)
    Is it possible to send you the equipment of the system that is happening this time and consider it?

    If the phenomenon can not be reproduced, I think that investigating the cause will not proceed.
    Can you contact me directly by email if it is ok?

    Thank you for your consideration.

  • Hello Masazumi

    I am in touch with you by email. Let's continue the support by it.

    Regards
    Francisco