Re: P5LD2 Deluxe - anyone use? - BIOS for Intel 930
- From: "R" <none@xxxxxxxx>
- Date: Wed, 26 Apr 2006 08:03:41 -0500
Thanks for all of the info.
Back in the old days I did hot flash some chips, but with the placement of
that fan connector right next to the BIOS I don't have the confidence to do
that. The only chip puller I have (and I can't find any others) is
essentially too big in the first place and all I can do is take one end of
it and try to pop up the corner of the chip opposite that stupidly placed
fan connector and carefully pry it up. I did bend the pins on one chip (out
of about 12 chips pulled) but was able to straigthten them.
"Paul" <nospam@xxxxxxxxxx> wrote in message
In article <e2jj62$miq$1@xxxxxxxxxxxxxxxxx>, "R" <none@xxxxxxxx> wrote:
I have put together about 10 systems with the P5LD2 Deluxe board. Found
that to use the 930 you have to have latest BIOS (channel still shipping
boards with BIOS that won't support the 930 which has been out for about
2 choices: either own an 830 processor to sit on your bench, install in
board and upgrade, then swap processors (and I hate the way these new
Chips & heatsinks work). OR, send $5 and your BIOS to Asus bios dept.
The first 3 BIOS that I received back from Asus worked fine. The next
did not; so I had to buy an 830. Wrote them a letter explaining, and
received 2 more back. These two won't even POST let alone run a 930.
Could Fedex be doing something to the package that would wipe out the
(4 out of the last 4 times?)
"What's more, this EPROM could be erased by exposure to X-rays."
The EEPROM still relies on the same mechanisms as the EPROM, and
uses a charge on a floating gate. What I cannot tell you, is what
exposure level would be required for erasure. If Fedex or Homeland
Security were really doing this, then all sorts of devices relying
on serial EEPROMs would be really messed up. A lot of damaged
consumer goods and unhappy etailers.
According to the Fedex web site, X-rays are mentioned as a possibility
for certain international destinations, but who is to say they
keep those pages up to date.
To investigate the nature of the problem, you need a way to
"read back" the bad devices, and see what information is stored
in them. A device receiving a thorough erasure operation, has
all of its bits in one state. Maybe this is all 1's, or hex 0xFF for
all of the bytes in the device. A partial erasure, would lead to
a number of bits being converted to 1's, and the rest being in
whatever state they were lasted programmed. Think of this as
a "fog of 1's".
This could be a mistake in configuring the programming device, as
it is unlikely they shove the chips into motherboards for the
programming. Only reading out the contents of the chip, using
a programmer, would give you some idea as to what's up. One way to
do this, would be the "hot flash" procedure, where you boot a
motherboard with a good chip, then pull the flash chip out (with
the power on!), then insert the duff chip. That must be done
with great care - the chips tend to "hop around" in the socket,
and accidentally connecting the wrong pins is a possible outcome.
Then use the Asus provided DOS flashing tool, to archive the
contents of the hot inserted chip, to a floppy. At your leisure,
you can then examine the file and make a guess as to the failure
mechanism, based on the observed erasure pattern.
I would recommend a BIOS Savior as a safe way to clone BIOS
chips, but unfortunately ioss.com.tw seems to be asleep at
the switch, and they have not made a device for the 1MB
(8 megabit) flash chips. They still seem to offer BIOS Savior
for sale, but for the latest Deluxe/Premium Intel boards from Asus,
that use the large flash chips, they would not be compatible.
In terms of the flash contents, be aware that they are not
immutable. The BIOS flash chip is used as a repository for
information recorded during POST. That means, when you use
a flash utility, to update a BIOS image, before you reset the
machine, the image might still match the file used to program it.
But, once a computer POSTs, the BIOS code actually causes the
DMI/ESCD area of the flash chip to be updated. When comparing
the read-back contents of the chip, against the file used to
create it, that should be kept in mind.
Asus has become better at retaining the contents of the boot
block, and not attempting to reprogram it. So that is another
area of a flash chip, where the command options fed to the flash
program, will determine whether that segment gets updated or not.
So, what you would be doing, when comparing the archived file
from the duff chip, is perhaps examining the upper half of
the file (hex addr 0x00000 to N/2). I think that is where the
immutable BIOS code modules live, while the rest of the changable
area is at the high addresses (as seen on my hex editor :-)).
I wish I had a practical solution to offer you. Looking at the
purchase of standalone programmer products, I don't know if
I could pick a "winner" from the plethora of crap Xbox chip
programmers. There are reputable firms, like Data-IO. They
make programmers for thousands of dollars, that will program
anything. Some of the products are socketless, so you just lay
the chip in a cavity, and the box detects where the pins are.
Part of their service, is delivering updates for their programmer
devices, when new flash chip types are created. This is not an
option except for the most wealthy operations. Maybe someone in
the business of flashing BIOS chips (like badflash.com), could
afford to use a universal programmer like that.
- Prev by Date: Re: dual channel ram configuration for dual processors on Asus K8N-DL
- Next by Date: Re: P5LD2 Deluxe - anyone use? - BIOS for Intel 930
- Previous by thread: Re: P5LD2 Deluxe - anyone use? - BIOS for Intel 930
- Next by thread: Re: P5LD2 Deluxe - anyone use? - BIOS for Intel 930