ADV7611: Internal EDID stays disabled

Are you using our ADV7611 evaluation board ( EVAL-ADV7611EB1Z / EVAL-ADV7611EB2Z )?

-Matt

No, it's a custom hardware.

Konstantin

Hi Konstantin,

Are you using our example scripts for ADV7611?  If you look in ADV7611-VER.2.9c.txt, you will find the following script which will configure the ADV7611 with internal EDID data.

:ADV7611 EDID 8 bit only NO DSD or HBR Support:

Then if you reconnect your source it should detect this new EDID.

-Matt

Hi Matt,

finaly we figured out what the problems were. The edid was loaded via too long i2c commands. After switching to a single byte loading of the edid it worked. The second problem was a hardware issue, which caused the HPA line to remain high once the source is connected leadig to a situation where the source never requested the newly loaded edid after startup.

Thanks & regards

Konstantin

Thanks for posting the solution for your issue!

-Matt

Matt,

I am using the eval board and used the script you refer to above.

My display is detected by the computer and I have forced free run but when I go to the HDMI tab the front porch, back porch, etc values don't change to those values in my EDID.

What settings do I need to have to show my specific timing characteristics in the HDMI values?

Thank you,

MS.

I have a problem with this. I cannot get EDID working on ADV7611, but the scripts I find around this website have references to undocumented registers. Looking at the ADV7604 linux driver source I've found on the web, it looks like these regs actually apply to ADV7604, not ADV7611.  

Specifically:

64 77 00 ; Disable the Internal EDID ==> ADV7611 shows EDID ena reg at 74, not 77

...

64 77 00 ; Set the Most Significant Bit of the SPA location to 0 ==> Same command as above, different behaviour?

64 52 20 ; Set the SPA for port B.  ==> ADV7611 doesn't define reg 52 in REPEATER bank

64 53 00 ; Set the SPA for port B. ==> ADV7611 doesn't define reg 53 in REPEATERbank

64 70 9E ; Set the Least Significant Byte of the SPA location  ==> ADV7611 doesn't define reg 70 in REPEATER bank

64 74 03 ; Enable the Internal EDID for Ports ==> ADV7611 only specifies b0 as EDID_A_ENABLE

It also mentions KSV a lot, although n the 7611 UM the same virtual IIC slave refers to REPEATER, not KSV.

All in all, it is incredibly confusing, and frustrating (because I can't get it running).  Perhaps they 'work anyway', but it isn't helping me understand what is going on, nor help me get my system up and running.

Any help would be appreciated.

Thanks in advance.

Edit: typos

Hi,

  Yes, EDID enable bit is 0x74 but not 0x77
   Please refer below link
          ADV7630 internal EDID issue 

ADV7611 doesn't define reg 52 in REPEATER bank?
  Ans : This is not mentioned in adv7611 documentation,Please refer this in other HDMI receiver chipset in ADV7619 design support file.Refer SPA or EDID details in ADV7619 Design Support file
           Did you refer this thread?
           ADV7619 EDID on Port B 

           Beleive that this would be helpful.

Thanks,

Poornima

Thanks Poornima. 

I'm not sure at which point we are expected to assume that 7630 and 7619 have compatible registers with 7611; I didn't even know about those parts until this morning.  Regardless, there seems to be helpful info there, although I haven't seen an EDID response from 7611 at this point (still working on it).

Note that my specific configuration using 7611 uses internal EDID RAM (I don't think there is an external EEPROM option / SPI bus), that I stuff using IIC writes to slave dev 6C.  Given my lack of success, do you know where I can find details of which things can prevent a response on the DDC bus?  Is there a working set of regs anywhere specific to this part? None of the scripts, in part or whole, seem to have worked for me.

Needless to say, I will post my solution when I finally have one.

Thank you

Rhys.

P.S. I'm particularly interested in this line from this page: ADV7630 internal EDID issue

64 77 04 ; Default 0x00 - Recalculate checksums

I haven't seen that documented anywhere either.

Nor this: 

64 7D 00 ; Default 0x00 - second EDID image access not enabled on Ports A, B, C, D

It seems like it would be practically impossible to get this running without the 'magic scripts' provided by Analog engineers.  That is concerning.