How do you tell whether a display card's firmware is dead?

I got a GeForce 8500GT that suddenly malfunctioned. Both its DVI and
HDMI ports are dead.

And is it possible that it's damaged by malware?

I have had a 7600GT and a Radeon 6850 that suddenly malfunctioned as well.

--
@~@ Remain silent! Drink, Blink, Stretch! Live long and prosper!!
/ v \ Simplicity is Beauty!
/( _ )\ May the Force and farces be with you!
^ ^ (x86_64 Ubuntu 9.10) Linux 2.6.39.3
�借貸! ��騙! ��交! �打交! �打劫! �自殺! 請考慮綜� (CSSA):
http://www.swd.gov.hk/tc/index/site_...sub_addressesa

Mr. Man-wai Chang wrote:

I got a GeForce 8500GT that suddenly malfunctioned. Both its DVI and
HDMI ports are dead.

And is it possible that it's damaged by malware?

I have had a 7600GT and a Radeon 6850 that suddenly malfunctioned as well.


Probably not.

Could be a power related failure (check ATX supply
outputs with a multimeter).

Could be a heat related failure (check that the cooling
fan still rotates when the card is powered). On some cards,
the fan fails first, the chip goes above 100C, and the
plastic on the fan body starts to melt. And that eventually
kills the chip.

*******

The EEPROM on the video card can be re-flashed.

I flash upgraded an ATI 9800Pro years ago. I changed
a Mac version of the card, to a PC version. It required
four soldering modifications, as well as a new flash image.
The Macintosh cards have a 2x sized flash chip (128KB),
making it easy to reflash a 64KB PC image into it. The
converse is not true, and to convert a PC card to
Mac, needs a larger EEPROM.

The flash chip contains a VESA declaration amongst
other things. It also happens to contain a CAS parameter
for the DRAM on the video card. For example, the EEPROM
code on my 9800Pro was loaded with a CAS3 setting for the
DRAM. The flash image you use, has to correspond to the
hardware. You check the RAM chips on the video card,
and *only* use a firmware image for the EEPROM chip, with
the correct RAM timing. When the BIOS does POST, the code
in the EEPROM contains a subroutine that sets up the
RAM on the video card. The video card must be commissioned
before the display can start during POST.

When flashing my 9800Pro, I inserted an FX5200 PCI video
card at the same time. The monitor cable connected to the
FX5200. The ATI card would not be expected to display
anything, during the flash process. While some people do
the flash process "blind", I chose to have a second video card
present so I could watch the progress of the flash
operation.

So if "malware" were to ever attack that card, I know
how to fix it.

But it is very unlikely that anything like this has happened.

If the card works at BIOS level, but doesn't work in the
OS, then you're going to need a "driver theory" as to
why it does not work.

Paul

On 30/6/2017 1:19 AM, Paul wrote:
Could be a power related failure (check ATX supply
outputs with a multimeter).

No.


Could be a heat related failure (check that the cooling
fan still rotates when the card is powered). On some cards,
the fan fails first, the chip goes above 100C, and the
plastic on the fan body starts to melt. And that eventually
kills the chip.

Fan still spinning fine.


The EEPROM on the video card can be re-flashed.
....
So if "malware" were to ever attack that card, I know
how to fix it.

But it is very unlikely that anything like this has happened.

If the card works at BIOS level, but doesn't work in the
OS, then you're going to need a "driver theory" as to
why it does not work.

Problem is I don't know anyone who's willing to try repairing this old
display card free-of-charge.

It's still interesting to find out whether it's a firmware issue or
really hardware damage.

--
@~@ Remain silent! Drink, Blink, Stretch! Live long and prosper!!
/ v \ Simplicity is Beauty!
/( _ )\ May the Force and farces be with you!
^ ^ (x86_64 Ubuntu 9.10) Linux 2.6.39.3
�借貸! ��騙! ��交! �打交! �打劫! �自殺! 請考慮綜� (CSSA):
http://www.swd.gov.hk/tc/index/site_...sub_addressesa