Pete Rynas wrote:
> well tech support finally emailed me and said my video cards are not
> compatable so i wrote back asking them for the " List " of incompatable
> cards - Duh!! it looks like paul and i are the only people w/any input
> anymore on this site?
> anyway, i was wondering if perhaps this samsung 932b monitor of mine needs
> a video cable the digital one with every pin in it?? i have a few with the
> few pins missing in a couple of spots but no full d-pinout plug - but that
> is what came w/ the monitor was the cable w/a few missing pins so that along
> with the fact that the analog doesn't work any better suggests that the
> cable isn't the case. ECS should list incompatable cards!!! i have a Fry's
> brand EVGA geforce 7200GS and a 7600GT , they work flawlessly once windows
> is running , but you have to reset, reboot , unplug , turn power switch in
> rear of machine off several times just to initialize the card( get video
> even for Bios)?? i have been at this comp building for years so i'm not
> stupid but i will admit i learn every day and may be missing something
> simple or obvious or ECS better come up with a "list" or refund / advance
> RMA this pile to something that will truely support PCI-E slotted card
>
>
Maybe there is a forum somewhere, where all the ECS fun takes place :-)
And you're missing out :-)
I don't think there is a particular reason to suppose the 945GT is
incompatible. Like I said before, all 16 lanes look to be wired,
judging by the pictures. And other companies have used that chip,
and I bet they don't have the same problems.
Have you been able to run any tests, after you get into Windows ?
Does everything work under load ?
Does the board pass memtest86+ without errors ?
Can you run Prime95 while in Windows ?
Once the video card driver is in place, you have DirectX installed
and so on, does a version of 3DMark run OK on the video card ?
If you're in Windows, and do a "restart", do the odds improve of
the motherboard starting properly ?
It sounds like there is some kind of initialization problem,
almost as if the BIOS is checking the video card before it
is ready, or there is just a problem with the BIOS design.
Since you say you've built a number of systems before, is
there a chance of you popping a PCI Port 80 display card in
the first PCI slot ? Those cards have a 2 digit hex display
and codes are written to the display by the BIOS. The codes
are not error codes - they are progress codes, indicating
when a certain activity is occurring.
The thing about the Port 80 display card, is figuring out
what is happening from the port card, is virtually impossible.
The progress codes tend to bundle several activities
together, so you don't know which one is at fault. And
in any case, the description of the activity is obscure
enough, that it doesn't mean much of anything.
But having said so many bad things about it, it does have
the benefit of being an "activity" indicator. In a system
where the motherboard starts inconsistently, you'd want to see
what the port code display looks like, for the working and
non-working cases.
Another example of an activity indicator, is some people
own RAM that has some kind of LED activity display on the
edge of the DIMM. Some people have used to good effect,
noting that when a BIOS won't run, the activity LEDs never
flash. That is almost as effective as the Port 80 card.
OK, so where does that leave us ?
When the BIOS starts, in your case, you have built-in graphics
and you have a PCI Express video card slot. (I'd have to look
it up, but it is also possible the video card slot also supports
the Intel ADD2 display card, so the PCI Express lanes have more
than one definition. With the ADD2 card, a digital stream is
passed to the ADD2 card, and it has things like a TMDS transmitter
chip on it, to drive a DVI connector.)
The BIOS should check the video card slot, to see if something
is plugged in. If a video card is detected, then at least on
some motherboards, the BIOS would disable the built-in 945GC
graphics, and rely on the video card instead. The BIOS should
load the code stored in the video BIOS chip on the video card.
On one of my FX5200 cards, the video BIOS actually displays
a banner prompt on the screen, when the video BIOS is loaded.
It is the first text to be displayed on the screen. After
that, the normal BIOS display appears. For the video card
to work normally, I would have expected that loading the
BIOS would have been an early activity. Not all video BIOS
display something on the screen, so you cannot rely on that
as a debugging feature. I have two FX5200 cards, one has
the banner, and the other one doesn't.
The video BIOS should define some standard display modes
(VESA?). A couple of those are "standard", and for the
BIOS to be able to display, the appropriate modes have
to exist. Matrox had a couple cards, that failed for some
reason, to meet that requirement, and the old Matrox forum
had a number of reports about it. But I cannot say that I've
read of ATI or Nvidia having problems like that.
Your problem could be a BIOS problem or it could be
a motherboard problem of some sort. Or even a power
supply problem for that matter. As I suggested before,
you can partially eliminate some of those possibilities,
like testing the video cards in another system. Swapping
power supplies doesn't guarantee anything, as there have
been cases where a particular model and brand of power
supply, doesn't work with a bunch of different motherboards.
(Antec comes to mind.) In the Antec case, it almost looked
like a timing problem, like the power supply output was slow
to reach full voltage or something.
At some point, you're going to have to move on, and try
another motherboard. Perhaps your ECS board will make
new friends on Ebay...
I found a page for 932B here. It is a 19" with 1280x1024
resolution. That should not cause any grief, for any interface.
That resolution can be handled by a "single link" DVI digital
output (half the pins populated).
http://www.samsung.com/za/consumer/d...LS19PEBSBV/EDC
http://en.wikipedia.org/wiki/Dvi
Paul
Similar Threads
Linear Mode
