Configuring a Samsung 2253BW - Suse

This is a discussion on Configuring a Samsung 2253BW - Suse ; Hello I recently bought a Samsung 2253BW monitor and I have been trying to use it with Suse 10.2 (dual-head configuration using xinerama). The computer has a Nvidia GeForce4 MX 440 so I am using the Nvidia version: Installed rpms: ...

+ Reply to Thread
Results 1 to 3 of 3

Thread: Configuring a Samsung 2253BW

  1. Configuring a Samsung 2253BW

    Hello

    I recently bought a Samsung 2253BW monitor and I have been trying
    to use it with Suse 10.2 (dual-head configuration using xinerama).
    The computer has a Nvidia GeForce4 MX 440 so I am using the Nvidia
    version:

    Installed rpms:
    xorg-x11-server-7.2-30.15
    nvidia-gfx-kmp-default-1.0.9631_2.6.18.2_34-0.1
    x11-video-nvidia-1.0.9631-0.1

    Looking at the X log, my problem starts with this:

    (--) NVIDIA(0): Samsung SyncMaster (DFP-0): 112.0 MHz maximum pixel clock
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    The X server thinks that there is a 112.0 MHz limit!
    I have no idea from where this comes (this is not yet EDID information),
    and more importantly, how to make X ignore it!

    (--) NVIDIA(0): Samsung SyncMaster (DFP-0): Internal Single Link TMDS
    (--) NVIDIA(0): Samsung SyncMaster (DFP-0): Native FlatPanel Scaling is
    (--) NVIDIA(0): supported
    (--) NVIDIA(0): Samsung SyncMaster (DFP-0): DFP modes are limited to 60 Hz
    (--) NVIDIA(0): refresh rate
    (--) NVIDIA(0):
    (--) NVIDIA(0): --- EDID for Samsung SyncMaster (DFP-0) ---
    (--) NVIDIA(0): EDID Version : 1.3
    (--) NVIDIA(0): Manufacturer : SAM
    (--) NVIDIA(0): Monitor Name : Samsung SyncMaster
    (--) NVIDIA(0): Product ID : 892
    (--) NVIDIA(0): 32-bit Serial Number : 1095840306
    (--) NVIDIA(0): Serial Number String : H9XQ602179
    (--) NVIDIA(0): Manufacture Date : 2008, week 23
    (--) NVIDIA(0): DPMS Capabilities : Active Off
    (--) NVIDIA(0): Prefer first detailed timing : Yes
    (--) NVIDIA(0): Supports GTF : No
    (--) NVIDIA(0): Maximum Image Size : 470mm x 300mm
    (--) NVIDIA(0): Valid HSync Range : 30 kHz - 81 kHz
    (--) NVIDIA(0): Valid VRefresh Range : 56 Hz - 75 Hz
    (--) NVIDIA(0): EDID maximum pixel clock : 146.2 MHz
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    although EDID says the limit is 146.2 MHz !

    (--) NVIDIA(0):
    (--) NVIDIA(0): Established Timings:
    (--) NVIDIA(0): 640 x 480 @ 60 Hz
    (--) NVIDIA(0): 640 x 480 @ 72 Hz
    (--) NVIDIA(0): 640 x 480 @ 75 Hz
    (--) NVIDIA(0): 800 x 600 @ 56 Hz
    (--) NVIDIA(0): 800 x 600 @ 60 Hz
    (--) NVIDIA(0): 800 x 600 @ 72 Hz
    (--) NVIDIA(0): 800 x 600 @ 75 Hz
    (--) NVIDIA(0): 1024 x 768 @ 60 Hz
    (--) NVIDIA(0): 1024 x 768 @ 70 Hz
    (--) NVIDIA(0): 1024 x 768 @ 75 Hz
    (--) NVIDIA(0): 1280 x 1024 @ 75 Hz
    (--) NVIDIA(0):
    (--) NVIDIA(0): Standard Timings:
    (--) NVIDIA(0): 1280 x 1024 @ 60 Hz
    (--) NVIDIA(0): 1280 x 960 @ 60 Hz
    (--) NVIDIA(0): 1152 x 864 @ 75 Hz
    (--) NVIDIA(0):
    (--) NVIDIA(0): Detailed Timings:
    (--) NVIDIA(0): 1680 x 1050 @ 60 Hz
    (--) NVIDIA(0): Pixel Clock : 119.00 MHz
    (--) NVIDIA(0): HRes, HSyncStart : 1680, 1728
    (--) NVIDIA(0): HSyncEnd, HTotal : 1760, 1840
    (--) NVIDIA(0): VRes, VSyncStart : 1050, 1053
    (--) NVIDIA(0): VSyncEnd, VTotal : 1059, 1080
    (--) NVIDIA(0): H/V Polarity : +/-
    (--) NVIDIA(0):

    It even gets a timing line for the 1680x1050 mode.

    The problem is that later it rejects this 1680x1050 mode (despite
    obtaining it from EDID) :

    (II) NVIDIA(1): --- Building ModePool for Samsung SyncMaster (DFP-0) ---
    (II) NVIDIA(1): Validating Mode "1680x1050":
    (II) NVIDIA(1): 1680 x 1050 @ 60 Hz
    (II) NVIDIA(1): For use as DFP backend.
    (II) NVIDIA(1): Mode Source: EDID
    (II) NVIDIA(1): Pixel Clock : 119.00 MHz
    (II) NVIDIA(1): HRes, HSyncStart : 1680, 1728
    (II) NVIDIA(1): HSyncEnd, HTotal : 1760, 1840
    (II) NVIDIA(1): VRes, VSyncStart : 1050, 1053
    (II) NVIDIA(1): VSyncEnd, VTotal : 1059, 1080
    (II) NVIDIA(1): H/V Polarity : +/-
    (WW) NVIDIA(1): Mode is rejected: PixelClock (119.0 MHz) too high for
    (WW) NVIDIA(1): Display Device (Max: 112.0 MHz).


    I wrote a modeline using a 112 MHz clock:

    ModeLine "1680x1050" 112.00 1680 1696 1728 1768 1050 1051 1054 1055

    but is its also being rejected, despite being considered "valid" :

    (II) NVIDIA(1): Validating Mode "1680x1050":
    (II) NVIDIA(1): 1680 x 1050 @ 60 Hz
    (II) NVIDIA(1): Mode Source: X Configuration file ModeLine
    (II) NVIDIA(1): Pixel Clock : 112.00 MHz
    (II) NVIDIA(1): HRes, HSyncStart : 1680, 1696
    (II) NVIDIA(1): HSyncEnd, HTotal : 1728, 1768
    (II) NVIDIA(1): VRes, VSyncStart : 1050, 1051
    (II) NVIDIA(1): VSyncEnd, VTotal : 1054, 1055
    (II) NVIDIA(1): H/V Polarity : +/+
    (II) NVIDIA(1): Mode is valid.
    (II) NVIDIA(1):
    (WW) NVIDIA(1): Unable to use mode "1680x1050" for Samsung SyncMaster (DFP-0);
    (WW) NVIDIA(1): cannot compute backend DFP timings (mode is larger than
    (WW) NVIDIA(1): native backend 1280 x 1024).

    Why is it thinking the "native backend" is 1280 x 1024 ? Possibly because
    it rejected the 1680x1050 mode above when looking for modes "For use as
    DFP backend."

    I tried a number of options to try to make it ignore automatic information
    but nothing seems to work

    Option "UseEDID" "false"
    Option "IgnoreEdid" "yes"
    Option "NoEdid"
    Option "ModeValidation" "NoDFPNativeResolutionCheck"
    Option "ModeValidation" "NoVirtualSizeCheck"

    Recent versions of X are really annoying in the way they ignore what is
    written in xorg.conf

    Any ideas ?

    --
    http://www.mat.uc.pt/~rps/

    ..pt is Portugal| `Whom the gods love die young'-Menander (342-292 BC)
    Europe | Villeneuve 50-82, Toivonen 56-86, Senna 60-94

  2. Re: Configuring a Samsung 2253BW

    Rui Pedro Mendes Salgueiro wrote:

    > Hello
    >
    > I recently bought a Samsung 2253BW monitor and I have been trying
    > to use it with Suse 10.2 (dual-head configuration using xinerama).
    > The computer has a Nvidia GeForce4 MX 440 so I am using the Nvidia
    > version:
    >
    > Installed rpms:
    > xorg-x11-server-7.2-30.15
    > nvidia-gfx-kmp-default-1.0.9631_2.6.18.2_34-0.1
    > x11-video-nvidia-1.0.9631-0.1
    >
    > Looking at the X log, my problem starts with this:
    >
    > (--) NVIDIA(0): Samsung SyncMaster (DFP-0): 112.0 MHz maximum pixel clock
    > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    > The X server thinks that there is a 112.0 MHz limit!
    > I have no idea from where this comes (this is not yet EDID information),
    > and more importantly, how to make X ignore it!
    >
    > (--) NVIDIA(0): Samsung SyncMaster (DFP-0): Internal Single Link TMDS
    > (--) NVIDIA(0): Samsung SyncMaster (DFP-0): Native FlatPanel Scaling is
    > (--) NVIDIA(0): supported
    > (--) NVIDIA(0): Samsung SyncMaster (DFP-0): DFP modes are limited to 60 Hz
    > (--) NVIDIA(0): refresh rate
    > (--) NVIDIA(0):
    > (--) NVIDIA(0): --- EDID for Samsung SyncMaster (DFP-0) ---
    > (--) NVIDIA(0): EDID Version : 1.3
    > (--) NVIDIA(0): Manufacturer : SAM
    > (--) NVIDIA(0): Monitor Name : Samsung SyncMaster
    > (--) NVIDIA(0): Product ID : 892
    > (--) NVIDIA(0): 32-bit Serial Number : 1095840306
    > (--) NVIDIA(0): Serial Number String : H9XQ602179
    > (--) NVIDIA(0): Manufacture Date : 2008, week 23
    > (--) NVIDIA(0): DPMS Capabilities : Active Off
    > (--) NVIDIA(0): Prefer first detailed timing : Yes
    > (--) NVIDIA(0): Supports GTF : No
    > (--) NVIDIA(0): Maximum Image Size : 470mm x 300mm
    > (--) NVIDIA(0): Valid HSync Range : 30 kHz - 81 kHz
    > (--) NVIDIA(0): Valid VRefresh Range : 56 Hz - 75 Hz
    > (--) NVIDIA(0): EDID maximum pixel clock : 146.2 MHz
    > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    > although EDID says the limit is 146.2 MHz !
    >
    > (--) NVIDIA(0):
    > (--) NVIDIA(0): Established Timings:
    > (--) NVIDIA(0): 640 x 480 @ 60 Hz
    > (--) NVIDIA(0): 640 x 480 @ 72 Hz
    > (--) NVIDIA(0): 640 x 480 @ 75 Hz
    > (--) NVIDIA(0): 800 x 600 @ 56 Hz
    > (--) NVIDIA(0): 800 x 600 @ 60 Hz
    > (--) NVIDIA(0): 800 x 600 @ 72 Hz
    > (--) NVIDIA(0): 800 x 600 @ 75 Hz
    > (--) NVIDIA(0): 1024 x 768 @ 60 Hz
    > (--) NVIDIA(0): 1024 x 768 @ 70 Hz
    > (--) NVIDIA(0): 1024 x 768 @ 75 Hz
    > (--) NVIDIA(0): 1280 x 1024 @ 75 Hz
    > (--) NVIDIA(0):
    > (--) NVIDIA(0): Standard Timings:
    > (--) NVIDIA(0): 1280 x 1024 @ 60 Hz
    > (--) NVIDIA(0): 1280 x 960 @ 60 Hz
    > (--) NVIDIA(0): 1152 x 864 @ 75 Hz
    > (--) NVIDIA(0):
    > (--) NVIDIA(0): Detailed Timings:
    > (--) NVIDIA(0): 1680 x 1050 @ 60 Hz
    > (--) NVIDIA(0): Pixel Clock : 119.00 MHz
    > (--) NVIDIA(0): HRes, HSyncStart : 1680, 1728
    > (--) NVIDIA(0): HSyncEnd, HTotal : 1760, 1840
    > (--) NVIDIA(0): VRes, VSyncStart : 1050, 1053
    > (--) NVIDIA(0): VSyncEnd, VTotal : 1059, 1080
    > (--) NVIDIA(0): H/V Polarity : +/-
    > (--) NVIDIA(0):
    >
    > It even gets a timing line for the 1680x1050 mode.
    >
    > The problem is that later it rejects this 1680x1050 mode (despite
    > obtaining it from EDID) :
    >
    > (II) NVIDIA(1): --- Building ModePool for Samsung SyncMaster (DFP-0) ---
    > (II) NVIDIA(1): Validating Mode "1680x1050":
    > (II) NVIDIA(1): 1680 x 1050 @ 60 Hz
    > (II) NVIDIA(1): For use as DFP backend.
    > (II) NVIDIA(1): Mode Source: EDID
    > (II) NVIDIA(1): Pixel Clock : 119.00 MHz
    > (II) NVIDIA(1): HRes, HSyncStart : 1680, 1728
    > (II) NVIDIA(1): HSyncEnd, HTotal : 1760, 1840
    > (II) NVIDIA(1): VRes, VSyncStart : 1050, 1053
    > (II) NVIDIA(1): VSyncEnd, VTotal : 1059, 1080
    > (II) NVIDIA(1): H/V Polarity : +/-
    > (WW) NVIDIA(1): Mode is rejected: PixelClock (119.0 MHz) too high for
    > (WW) NVIDIA(1): Display Device (Max: 112.0 MHz).
    >
    >
    > I wrote a modeline using a 112 MHz clock:
    >
    > ModeLine "1680x1050" 112.00 1680 1696 1728 1768 1050 1051 1054 1055
    >
    > but is its also being rejected, despite being considered "valid" :
    >
    > (II) NVIDIA(1): Validating Mode "1680x1050":
    > (II) NVIDIA(1): 1680 x 1050 @ 60 Hz
    > (II) NVIDIA(1): Mode Source: X Configuration file ModeLine
    > (II) NVIDIA(1): Pixel Clock : 112.00 MHz
    > (II) NVIDIA(1): HRes, HSyncStart : 1680, 1696
    > (II) NVIDIA(1): HSyncEnd, HTotal : 1728, 1768
    > (II) NVIDIA(1): VRes, VSyncStart : 1050, 1051
    > (II) NVIDIA(1): VSyncEnd, VTotal : 1054, 1055
    > (II) NVIDIA(1): H/V Polarity : +/+
    > (II) NVIDIA(1): Mode is valid.
    > (II) NVIDIA(1):
    > (WW) NVIDIA(1): Unable to use mode "1680x1050" for Samsung SyncMaster
    > (DFP-0);
    > (WW) NVIDIA(1): cannot compute backend DFP timings (mode is larger
    > than
    > (WW) NVIDIA(1): native backend 1280 x 1024).
    >
    > Why is it thinking the "native backend" is 1280 x 1024 ? Possibly because
    > it rejected the 1680x1050 mode above when looking for modes "For use as
    > DFP backend."
    >
    > I tried a number of options to try to make it ignore automatic information
    > but nothing seems to work
    >
    > Option "UseEDID" "false"
    > Option "IgnoreEdid" "yes"
    > Option "NoEdid"
    > Option "ModeValidation" "NoDFPNativeResolutionCheck"
    > Option "ModeValidation" "NoVirtualSizeCheck"
    >
    > Recent versions of X are really annoying in the way they ignore what is
    > written in xorg.conf
    >
    > Any ideas ?
    >


    Are you using DVI or VGA connection? IIRC, X needs to probe and trial/error
    with VGA. However, with DVI, X can ask the monitor for its details.

    I have a SyncMaster 2253BW but I'm using openSUSE 10.3 not dual-head.

    My Xorg.0.log contains:

    (--) NVIDIA(0): Samsung SyncMaster (DFP-0): 330.0 MHz maximum pixel clock

    According to nvdock I have:

    GeForce 8600 GT 256MB PCI Express 16X

    My xorg.conf (written by sax2) contains:

    Section "Monitor"
    Identifier "Monitor[0]"
    VendorName "Unknown"
    ModelName "Unknown"
    UseModes "Modes[0]"
    HorizSync 30.0 - 40.0
    VertRefresh 43.0 - 75.0
    Option "CalcAlgorithm" "XServerPool"
    Option "DPMS"
    EndSection

    I have not done anything special concerning this monitor. When I attached it
    to my tower, it "Just Worked".

    Could it be that X in openSUSE 10.2 is too old?

    I have xorg-x11-server-7.3-118.3

    Any particular reason you haven't upgraded from 10.2?

    --
    Chris

  3. Re: Configuring a Samsung 2253BW

    Chris wrote:
    > Rui Pedro Mendes Salgueiro wrote:


    > > I recently bought a Samsung 2253BW monitor and I have been trying
    > > to use it with Suse 10.2 (dual-head configuration using xinerama).
    > > The computer has a Nvidia GeForce4 MX 440 so I am using the Nvidia
    > > version:


    > > Any ideas ?
    > >


    > Are you using DVI or VGA connection?


    DVI.

    > IIRC, X needs to probe and trial/error
    > with VGA. However, with DVI, X can ask the monitor for its details.


    The problem is when it gets the wrong answer and there is no way to
    override it

    > I have a SyncMaster 2253BW but I'm using openSUSE 10.3 not dual-head.


    That is great. It means that this monitor can work with X. Are you
    using the nvidia driver ?

    > My Xorg.0.log contains:


    > (--) NVIDIA(0): Samsung SyncMaster (DFP-0): 330.0 MHz maximum pixel clock


    That makes more sense than 112 MHz

    > According to nvdock I have:


    > GeForce 8600 GT 256MB PCI Express 16X


    > My xorg.conf (written by sax2) contains:


    > Section "Monitor"
    > Identifier "Monitor[0]"
    > VendorName "Unknown"
    > ModelName "Unknown"
    > UseModes "Modes[0]"
    > HorizSync 30.0 - 40.0
    > VertRefresh 43.0 - 75.0


    > Option "CalcAlgorithm" "XServerPool"


    Hmm, does somebody happens to know what this line does ?

    > Option "DPMS"
    > EndSection


    > I have not done anything special concerning this monitor. When I attached it
    > to my tower, it "Just Worked".


    I will have to try tomorrow to connect this monitor to another computer
    and check if that works.

    > Could it be that X in openSUSE 10.2 is too old?


    I hope so.

    > I have xorg-x11-server-7.3-118.3


    > Any particular reason you haven't upgraded from 10.2?


    This is my work computer in which I only kill the X-windows when I really
    need to (the last time was nearly 2 months ago). Upgrading now means
    nearly a day without being able to use it and probably some more time
    solving various problems. So I was planning to only do that when 10.2
    is dropped (and probably go straight to 11.1 if it is out at that time).

    OTOH, my power supply almost failed today (the fan stopped. It was lucky
    that I noticed the smell on time and that the fan restarted), so I will
    have to turn the computer off today to replace it


    Thanks for the reply.

    --
    http://www.mat.uc.pt/~rps/

    ..pt is Portugal| `Whom the gods love die young'-Menander (342-292 BC)
    Europe | Villeneuve 50-82, Toivonen 56-86, Senna 60-94

+ Reply to Thread