Announcement

Collapse
No announcement yet.

890GX/HD4290 dual display question

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #51
    Originally posted by agd5f View Post
    Did changing any of those bios options I suggested help?
    Nope. I've tried every option in every combination.

    One thing to note is that the available options change when there is a card in the first x16 slot... so there is definitely some kind of auto-detection going on.

    Comment


    • #52
      If the monitor connected to the DisplayPort does not even show anything in the POST screen when it is the only connected display, then you should talk to the manufacturer.

      Comment


      • #53
        Originally posted by chithanh View Post
        If the monitor connected to the DisplayPort does not even show anything in the POST screen when it is the only connected display, then you should talk to the manufacturer.
        Have been. Got a reply from them last night that was less than informed.... the person who replied apparently didn't realize that displayport is the successor to DVI rather than just being a general term that encompasses VGA, DVI, and HDMI.

        Comment


        • #54
          You might just want to send the board back to the vendor you bought it from and exchange it for a new one rather than going through the actual oem.

          Comment


          • #55
            Originally posted by agd5f View Post
            You might just want to send the board back to the vendor you bought it from and exchange it for a new one rather than going through the actual oem.
            Only problem there is that the vendor sent me a "v1.0" board when apparently "v2.0" is the current version.

            So there's a few things I'm trying to get out of the manufacturer before I decide what I'm doing about this;
            1) what is the difference between a 1.0 and a 2.0? I.e. could this be a problem with 1.0's?
            2) is it a board problem or a bios problem and do they have a beta bios that fixes it?

            Comment


            • #56
              Although this is mostly for my benefit being that I'm the one who bought this board, I do like to keep the story complete in case anyone else can gain anything out of it.

              Got another communication from the manufacturer today (apparently they tested it out and found what was wrong). Turns out that this mainboard lacks (in their words) the "circuitry required to be able to use a display port to DVI adapter". I asked for clarification if that meant that it lacked the clock signal line (required for DVI but not for DP), and they confirmed it.

              Which basically means that a passive adapter is a definite no-go on this board.

              Of course that doesn't explain why the ACTIVE adapter also doesn't work, unless for some reason the active adapter I use doesn't have its own clock generator and just passes it through. Which seems like a reasonable possibility.

              Now the REAL question since my brain has a thoroughly geekish outlook on electronics, is WHERE is the clock source on the board (since presumably, the radeon part has it and its just not hooked up), and what does it need to be connected to?

              Is it possible to hijack the clock source on the DVI plug and wire it into the DP? (I know, different monitors need different clocks, otherwise there would be no need for a clock signal, but in this case, the monitors are identical.... so... ??? )

              Comment


              • #57
                Both the DP clock and the TMDS clock are both derived from the pixel pll. In this case, it's probably more of an issue of the oem having not wired all the pins in the connector properly.

                Comment


                • #58
                  Originally posted by agd5f View Post
                  Both the DP clock and the TMDS clock are both derived from the pixel pll. In this case, it's probably more of an issue of the oem having not wired all the pins in the connector properly.
                  And with that, I jump in for another round of studying DP pinouts, pin descriptions, and DP-->DVI adapter schematics.

                  What I have uncovered is quite interesting;

                  1) There are precisely TWO pins required in a DP-->DVI adapter that are *NOT* required for DP by itself; Cable Adapter Detect, and DP_PWR (+3.3). The rest of the pins used *MUST* be present in order for DP to function.

                  2) Apparently, Cable Adapter Detect is normally held low, so presumably, pulling it HIGH is how you tell it that there is an adapter present... which interestingly enough, requires power... from DP_PWR... which is the second pin that might not be connected.

                  3) My *ACTIVE* adapter is powered by the DP plug.... and doesn't work on this board. An active adapter makes a DVI monitor (or whatever other kind of monitor) into a DP monitor, so there is no reason why this wouldn't work unless the active adapter is either seriously retarded, or there is no power running to it.

                  And a pattern emerges.....


                  Going to test the voltage on the DP_PWR pin when I get home tonight. $10 says its dead.

                  When I confirm that, rather than modifying the mainboard, I'm going to modify the DP-->DVI plug (the active one, so I don't ram +3.3 down into something when I don't know where it goes...) and add external +3.3 to it.

                  I have a good feeling about this....

                  Comment


                  • #59
                    Originally posted by droidhacker View Post
                    And with that...
                    blah, blah, blah
                    No dice.
                    The DP_PWR is just fine, and for extra confirmation (since the voltage seemed a little low), I disconnected the DP_PWR pin and connected an external 3.3 anyway -- still nothing.

                    Upon further inspection of the DP-->DVI adapter, I cant help but notice that the Cable Adapter Detect line on it is tied to the Return line on the DP plug by a resistor, which means that for some reason, the adapter detect line is being held high. Not good -- suggests that this DP-DVI adapter is trying to do something weird/wrong since it should NOT be doing that.

                    Anyway, this pretty much confirms that the Cable Adapter Detect line is N/C, since every other pin on it is either strictly necessary for DP by itself, or confirmed connected.

                    So the question is.... what pin on what chip is the Cable Adapter Detect? These specifications have to exist somewhere, right?

                    Comment


                    • #60
                      I think the pinouts are in the RS780 databooks:


                      RS880 should be pretty much the same.

                      Comment

                      Working...
                      X