Time for some upgrades to the retro-PC. But which one? I have three!

(Some people might ask, "why do you need THREE retro PCs?" To the many
people asking that incredibly on-point and justified question, allow
me to respond by saying, "Shut up!" ;-)

WHAT AND WHY:
------------------------------------
So, of the three retro-PCs in question, there is the 'original'
retro-PC that I built a few years back: an old Athlon-XP 5200 with a
GeForce 7950 running Windows XP. Then there is my 'newest' retro-PC
that I just slapped together a week ago, with its Pentium II-400 and a
Voodoo3, running DOS/Windows95. And finally, I have a third retro-PC,
built back in March: a Pentium II-400 that /used/ to have a Voodoo3,
before I moved that video-card to the Windows95 machine. It is this
third machine which is getting the upgrade.

I am upgrading partly because it seemed silly to have /two/ Pentium
II-400 MHz computers, but mostly because the only other video-card I
had available from that era wasn't compatible with the existing
motherboard. And since a computer without a video-card isn't very
exciting, a new motherboard it would get!

THE UPGRADE:
------------------------------------
Digging through my stash, I decided on an ABit KT7A-RAID motherboard
and an Athlon 900MHz CPU. I was fortunate to find a heat-sink/fan that
fit the board; I'd long discarded the original and needed to buy a new
one. Oddly enough, heat-sinks compatible with 20-year old motherboards
just aren't as common today as they used to be. ;-)

I was actually somewhat surprised the motherboard still worked; it had
for years been pinned to the wall (along with numerous other
motherboard and expansion cards) as part of a geeky art project. I had
taken some care with the mounting, of course, but the board had more
or less sat unprotected in the open air for nearly a decade.
Installation was easy, although a lack of a PC speaker made
troubleshooting difficult since I couldn't hear the beep-codes.

(I literally dragged another retro-PC over to plug its speaker into
the new board just to get past those initial tests. I really should
invest in a $5 PC speaker though, dontcha think?)

Other than that, the only tricky part was getting the motherboard and
USB drivers to work. The ones that came on the CD-ROM with the
motherboard were particularly flaky (at least on Windows98). I finally
had to burn a CD-R just to move some updated drivers to the machine.

(For obvious reasons, I'm not keen on letting this PC onto the
Internet)

I coupled the new motherboard with a GeForce4 AGP 420MX video-card,
which arguably is a bit too new for a Windows98-era machine (the card
itself released in 2002). Still, an "entry level" video-card from 2002
is equal to a "high performance" card from 1998, so I think I can live
with the date mismatch. Besides, it's not like I had a lot of
alternatives on hand. I junked a lot of my older cards a few years
back, thinking I'd never have use for them.

Other than that, the PC is the same as it was before the upgrade: it
has the same RAM, the same hard-drive, the same soundcard, etc.
Disappointingly, while the new motherboard does have a USB 2.0 header,
it uses a proprietary design made only to be used with its proprietary
rear-panel USB port. That means I still can't use the chassis'
built-in front-panel USB ports or the SD-Card reader. Bummer. Also,
I'm not entirely sure that the 250W power-supply will be sufficient
when all the components are at maximum draw. We'll see how it goes; I
don't really have an alternative so hopefully that aging PSU can keep
up.

Another downside is that the motherboard's faster external bus speed
causes some difficulties with my Sidewinder 3D Pro joystick. There are
some hacks that /might/ help with this, though, so all hope is not
lost. Still, even if I can't get it working, I'm not that upset. The
'stick works well enough with my older Win95 retro-PC, so if I'm
forced to use a different joystick with this Windows98 computer, I'm
okay with that too.

RESULTS & FINAL THOUGHTS:
------------------------------------
Performance wise, this upgrade has made a definite difference. The
7-Zip benchmark (good for testing combined CPU and disk performance)
jumped from 259kb/s to 360kb/s; nearly a 140% speed boost. Synthetic
benchmarks saw similar improvements (3DMark99 jumped from 3034 to
5669, a 180% jump; 3DMark2000 jumped from 1414 to 4873, a 344%
improvement!). Games are likewise snappier, although - since most of
the games already installed on that PC were selected with the slower
processor in mind - the difference isn't /that/ noticeable. Everything
is a bit smoother and faster, but I wasn't dissatisfied with the
performance before the upgrade.

I'm generally happy with this upgrade. It went fairly smoothly and all
the components worked as expected. I might upgrade the OS to "Windows
ME" so its OS better matches the hardware; then again, maybe not.
Windows98SE works well enough. It really depends on what I end up
doing with the Windows95 computer; I would really like some more
'distance' between the that PC and this one.

I still need to do some cosmetic work on the Win98 PC. The motherboard
is missing its IO back panel; right now, there's a gaping hole around
the serial/parallel/ps2 ports. I'll need to kludge up some sort of
replacement (too cheap to buy one on EBay, and lacking a 3D-printer,
it will probably be made out of cardboard). I need to clean up the
internal cabling (again); it's a rat's nest in there now. And
eventually I got to fix that broken reset button.

But all-in-all, a worthwhile effort that better differentiated my two
Win9x computers at a minimum cost, and put some older hardware to good
use.

New Specs
----------------------------
CPU: AMD Athlon K7 "Thunderbolt" 900MHz (was P2-400)
Mobo: Abit KT7A-RAID (was IBM Via v6MM)
RAM: 384MB Micron PC-133 SDRAM
GPU: GeForce4 MX 420 AGP with 64MB (was Voodoo3 2000)
HDD: Maxtor 120GB PATA-133
CD: ASUS 52X CDROM & TSST DVD
FDD: Zip-100 IDE
Audio: Soundblaster Live 5.1 PCI
NIC: Netgear 10/100 Ethernet
Case: Xoxide X-Purity EXE Edition
PSU: 250W
OS: Windows98SE

On 7/26/2023 10:21 AM, Spalls Hurgenson wrote:
>
> Time for some upgrades to the retro-PC. But which one? I have three!
>
> (Some people might ask, "why do you need THREE retro PCs?" To the many
> people asking that incredibly on-point and justified question, allow
> me to respond by saying, "Shut up!" ;-)
>
Because "Number of Games" is not the only Holy Number that Must Increase.

--
I've done good in this world. Now I'm tired and just want to be a cranky
dirty old man.

On Wed, 26 Jul 2023 16:40:38 -0700, Dimensional Traveler
<dtravel@sonic.net> wrote:

>On 7/26/2023 10:21 AM, Spalls Hurgenson wrote:
>>
>> Time for some upgrades to the retro-PC. But which one? I have three!
>>
>> (Some people might ask, "why do you need THREE retro PCs?" To the many
>> people asking that incredibly on-point and justified question, allow
>> me to respond by saying, "Shut up!" ;-)

>Because "Number of Games" is not the only Holy Number that Must Increase.

I hope 'number of IEC power cords' is in the Sacred Numerology,
because if it is, I'm a sure bet for getting into the Good Place. I
seem to gather those even faster than I do video games. ;-)

I'm still debating whether owning so many retro-PCs is commendable,
because I'm repurposing hardware people were just tossing in the bin
and making it useful again, or despicable because it represents an
excess of materialism.

Got the Sidewinder 3D Pro joystick working (or is it a Sidewinder Pro
3D? I can't remember and am too knackered too check). The drivers for
that device are extremely sensitive to external bus speeds. Originally
PCs had an 8MHz bus speed; that went up to 33MHz with PCI, and kept
increasing as CPUs got faster and faster. The Sidewinder joysticks
could keep up to 66MHz bus speeds, but after that it was very hit or
miss if the drivers detected them.

Fortunately, somebody a lot smarter (or, at least, a lot more
dedicated to the task) than me hacked together a driver pack that -
while a kludge to install - actually fixed the problem.
https://www.vogons.org/viewtopic.php?t=17941
So if you're in that infinitesimally small group of people who are
using retro computers AND own a Sidewinder joystick, take a look.

Other than that, the upgrade went swimmingly. I'm still having some
issues; USB remains extremely finicky, for instance, but that might be
a hardware problem (as mentioned, the mainboard had been used as an
art exhibit for years so some damage is possible ;-).

Game compatibility has drastically decreased too. I am not sure if
this is a result of the in-place upgrade (e.g., underlying problems
because Windows isn't really designed to have mainboards swapped out
from beneath it), or because the games were installed expecting a
Voodoo and now have to use a Direct3D card, or compatibility issues
with the games themselves (most of the installed games are from the
mid- to late-90s, when 3DFX was the expected high-end card, and were
optimized for MiniGL rather than Direct3D). So right now, gaming on
this PC is a lot worse than it was before the upgrade. Still, I
suspect I can hammer out most of the problems, although it may mean
re-installing a lot of the games.

But aggravating as these problems can be, it's the sort of thing I
enjoy doing - possibly even more than actually playing the games on
the PCs once they're up and running.

On Thu, 27 Jul 2023 10:02:08 -0400, Spalls Hurgenson
<spallshurgenson@gmail.com> wrote:

>
>Other than that, the upgrade went swimmingly. I'm still having some
>issues; USB remains extremely finicky, for instance, but that might be
>a hardware problem (as mentioned, the mainboard had been used as an
>art exhibit for years so some damage is possible ;-).

Well, it looks like I may have spoken too soon. Booting up the Win98
PC yesterday resulted in a 'system disk not found' error. Initial
thoughts were that the HDD had suffered some sort of error; a logical
error if I was lucky, a hardware fault if I was not. After all,
mechanical drives are one of the more fragile parts of
retro-computing, so it wasn't completely unexpected. It meant I would
have to reinstall the OS, drivers, apps and a ton of games, but I was
already inclined towards doing that anyway.

But it turns out it may be a bit more serious. The error message
appears regardless of what medium I try to boot from: whether it is an
alternate HDD, the ZIP-100 drive, or a bootable CD, all present the
same error message. The hardware is all detected by the BIOS; the PC
just doesn't seem to find the MBR to progress beyond that point. The
only thing that /does/ allow me to boot is if I use a floppy disk, and
while it /is/ possible to use a computer like that, I'm not ready to
go /that/ retro.

I still haven't given up on the machine but I think it may be a
motherboard issue. I've replaced pretty much every component in the PC
except for the motherboard. Of course, limited as I am to booting from
floppies, I have a much smaller cache of diagnostic tools than I
usually do. It may be a bad BIOS; perhaps some corruption has it
looking in the wrong place for the MBR. I'll see if I can find an
updated BIOS to reflash the ROMS. Still, I suspect that it is more
likely a fault on the board itself, if only because the southbridge
chip is burning hot to the touch.

Maybe hanging the motherboard on a wall as part of an art project
wasn't the best way to preserve the hardware after all. ;-)

I'll keep fiddling with the board, but if I can't find a solution I
may have to revert back to the Via66M/Pentium-II... except then I'll
be without a video card.

Ah well. Thankfully these retro-PCs are just a hobby and not something
I depend upon as my daily driver. And there's always PCEm if I really
need to play a game that only runs on Win9x...

Spalls Hurgenson <spallshurgenson@gmail.com> wrote:
>But it turns out it may be a bit more serious. The error message
>appears regardless of what medium I try to boot from: whether it is an
>alternate HDD, the ZIP-100 drive, or a bootable CD, all present the
>same error message. The hardware is all detected by the BIOS; the PC
>just doesn't seem to find the MBR to progress beyond that point. The
>only thing that /does/ allow me to boot is if I use a floppy disk, and
>while it /is/ possible to use a computer like that, I'm not ready to
>go /that/ retro.

If you can boot it from floppy and then access the hard disk normally,
I suspect that the problem isn't hardware related. I that case I would
trying running "fdisk /mbr" to rewrite the boot sector on the hard drive.
If that doesn't work, I'd check the BIOS settings, in particular the ones
affecting boot and boot order. Disable other boot options other than
your hard drive and floppy disk. If your BIOS has a key you can press
to sellect what you can boot from try using that to select the hard drive.
Clearing the CMOS to reset all the BIOS settings may also be a good idea.

If you have to, booting Windows 9x from a floppy isn't that bad. I used
to do that for specialized CONFIG.SYS and AUTOEXEC.BAT settings in order
to get various games to work.

--
l/ // Ross Ridge -- The Great HTMU
[oo][oo] rridge@csclub.uwaterloo.ca
-()-/()/ http://www.csclub.uwaterloo.ca:11068/
db //

On Sun, 30 Jul 2023 20:28:21 -0000 (UTC), rridge@csclub.uwaterloo.ca
(Ross Ridge) wrote:

>If you can boot it from floppy and then access the hard disk normally,
>I suspect that the problem isn't hardware related. I that case I would
>trying running "fdisk /mbr" to rewrite the boot sector on the hard drive.
>If that doesn't work, I'd check the BIOS settings, in particular the ones
>affecting boot and boot order. Disable other boot options other than
>your hard drive and floppy disk. If your BIOS has a key you can press
>to sellect what you can boot from try using that to select the hard drive.
>Clearing the CMOS to reset all the BIOS settings may also be a good idea.

As they used to say, 'Don't teach your grandmother to suck eggs'. I've
tried all the tricks ;-)

It's not a logical error of a single drive. Would that it were so
simple. Multiple hard-drives aren't recognized, despite being seen in
the BIOS. But it's more than that; the computer won't boot from CD-ROM
or the ZIP drive either, despite it formerly having that capability.
And the boot-to-floppy capability has - since the earlier post -
failed as well. So basically, all methods to boot no longer work.

Which makes any attempt to reflash the BIOS just that little bit
harder (not necessarily impossible since some BIOS bypass the usual
boot process and just read the data off the floppy directly, but
harder). Whether it's a spurious command in the ROMs or a damaged IO
chip, something is keeping the PC from locating the MBR on any devices
plugged into it. It's possible I could trace the fault; it could just
be a broken trace somewhere for all I know. Or maybe just a bad
capacitor.

(Back in the day I might even have pulled out an oscilliscope and
checked the chips directly. It's not really an option with modern
boards though, what with everything being combined into proprietary
single-chip designs and a lack of adequate documentation. Even if you
can figure out which chip is at fault, you probably won't be able to
source a replacement.)

But it is probably not worth the effort. Instead, I'm relegating the
motherboard to the 'art exhibit' pile again, and pulling out my next
spare motherboard. This time we're moving up to an Athlon XP 3000,
which is /far/ beyond what a Windows98 install needs, but I've only a
limited selection of old motherboards on hand and beggars can't be
choosers.

Luckily, the new board - an ASUS A7N8X - uses Socket-462, so I can
reuse the CPU cooler I bought for the Athlon 900 (fortunately, its TDP
is within the cooler's capability). Other than that, the hardware will
alrgely remain the same. Sure, the mainboard /could/ support much
faster hardware, but in this case I don't WANT the speed. Honestly,
even the 900MHz CPU was a bit overpowered for my needs.

The only downside is that the 250Watt power-supply I've been using
isn't going to be sufficient, and I don't have a viable spare on hand.
So I have to decide whether I'm going to buy a new PSU or wait for the
Bin Gods to bequeath me a suitable replacement.

(It's not that the cost is prohibitive; I just don't feel like
spending cash on this old hardware ;-)

So, not the end of the world. All hardware fails eventually. It just
amuses me that it happened so soon after I boasted about it being an
easy upgrade.

And maybe one day I'll go back and give the busted motherboard another
chance; retest the caps, look for obvious problems. It's just not
something I'm that interested in doing now, since I've an easier
alternative on hand.

Spalls Hurgenson <spallshurgenson@gmail.com> wrote:
>Which makes any attempt to reflash the BIOS just that little bit
>harder (not necessarily impossible since some BIOS bypass the usual
>boot process and just read the data off the floppy directly, but
>harder). Whether it's a spurious command in the ROMs or a damaged IO
>chip, something is keeping the PC from locating the MBR on any devices
>plugged into it. It's possible I could trace the fault; it could just
>be a broken trace somewhere for all I know. Or maybe just a bad
>capacitor.

The basic legacy boot process is really simple. For standard floppy
drives all that happens is that it reads the first sector of the floppy
and executes it. For standard hard drives this process is extended to
check whether the last two bytes of the boot sector read into memory
have the required magic value before executing it.

There's no "locating the MBR" going on here, the boot sector can only
be in one place. Except for the last two bytes of boot sector of hard
drives, the BIOS doesn't care whats written there. There can otherwise
be complete garbage written on the first sector of the disk and the BIOS
will still try to exectute it.

Things get can more complicated than this for non-standard drives, like
anything USB connected, as the BIOS has to emulate these drives, and for
CD-ROMs as they use a completely different boot method. However for
standard floppy and hard drives it's very simple. The only thing
that can prevent the BIOS from booting off of a floppy is an error
that prevents it from reading the first sector of a floppy. The only
thing that can prevent a hard drive from booting is that it can't read
the first sector of the drive or that first sector doesn't contain the
special magic number at the end.

--
l/ // Ross Ridge -- The Great HTMU
[oo][oo] rridge@csclub.uwaterloo.ca
-()-/()/ http://www.csclub.uwaterloo.ca:11068/
db //

On Mon, 31 Jul 2023 17:39:01 -0000 (UTC), rridge@csclub.uwaterloo.ca
(Ross Ridge) wrote:

>There's no "locating the MBR" going on here, the boot sector can only
>be in one place. Except for the last two bytes of boot sector of hard
>drives, the BIOS doesn't care whats written there. There can otherwise
>be complete garbage written on the first sector of the disk and the BIOS
>will still try to exectute it.

ROM BIOS /do/ need to locate the MBR. It does so by checking the start
of the boot sector device (head 1, cylinder 1, sector 1 in older CHS
drives, or logical block address 0 in larger LBA drives). If the last
two bytes in the code there has a signature value of 55AAh, it loads
the code into RAM and the ROM transfers control to the code in the
MBR.

(if it doesn't find code ending with 55AAh the ROM goes on to check
the next boot device until it either finds a valid MBR, or it finds
none, in which it triggers an interrupt and you get an error message.)

Once the MBR code is loaded into RAM, the MBR code takes over the boot
process: locating the partition table and active partition, then
finding the starting sector where the OS boot record is located, which
then bootstraps the machine all the way up to usability.

(note: this is how legacy BIOS did it. I honestly haven't been keeping
up with the intricacies of the UEFI boot process, which may be
different. But it probably isn't; legacy standards are hard to shake
;-)

You are correct that - aside from the signature value 55AAh in the
last two bytes - it doesn't really matter to the ROMs what code is in
the first sector. But the BIOS needs to look for, and find, the MBR on
a valid boot device before it can proceed.

>Things get can more complicated than this for non-standard drives, like
>anything USB connected, as the BIOS has to emulate these drives, and for
>CD-ROMs as they use a completely different boot method. However for
>standard floppy and hard drives it's very simple. The only thing
>that can prevent the BIOS from booting off of a floppy is an error
>that prevents it from reading the first sector of a floppy. The only
>thing that can prevent a hard drive from booting is that it can't read
>the first sector of the drive or that first sector doesn't contain the
>special magic number at the end.

This isn't a logical error with the disk media. If it were, the
problem wouldn't be repeated on multiple hard-drives drives and/or
different media. Remember, I've pulled the drives (and cables, and
RAM, and the CPU, and the PSU, and...) and tested replacements for
them all.

This is a hardware issue: possibly with the ROM-BIOS, or a bad chip in
the IO-System, or a cut trace on the board, or a capacitor not juicing
up sufficiently. The fact that the southbridge chip was hot to the
touch makes me suspect the second; there was probably a shorting
transistor in there somewhere. Thus, no amount of FDISK /MBR is going
to fix the problem when its the hardware that's faulty. Something is
preventing the BIOS from finding the signature value 55AAh that
indicates its found the MBR.

(alternately, the ROM could be invoking an interrupt to pull the 'disk
boot failure' error message regardless of whether it finds an MBR or
not. It might even be a problem with the CMOS RAM having a bad bit
storing erroneous data that misdirects the boot process to the wrong
location. Both are unlikely... but possible)

Since the PC won't boot anymore, short of pulling out the test probes
to see if there's short,* figuring out where the hardware has failed
can be tricky (and probably not worth the effort, except for
entertainment value).

(Award BIOS of the era had a boot sector block in their ROMs that
allowed the PC to boot up just enough to locate a BIN file on a floppy
and flash it to the ROMs. Thus you could 'boot' your PC with a floppy
drive that didn't have an OS on it (it still needed an MBR, of course.
And the archived ROM file). Given the already precarious state of the
boot process, I was hesitant at the time to try that method; I had no
confidence the ROM data would have been read correctly from the drive.
Now it's not even an option.)

But trust me when I say I've experience enough with the electronics to
deduce when its a error with the drive's partition/file-system
configuration versus a problem with the actual hardware. ;-)

* and a pair of magnifying glasses. Traces are really tiny and my eyes
aren't quite what they were. ;-)

Am 27.07.23 um 16:02 schrieb Spalls Hurgenson:
> I'm still debating whether owning so many retro-PCs is commendable,
> because I'm repurposing hardware people were just tossing in the bin
> and making it useful again, or despicable because it represents an
> excess of materialism.
I use a mister for retro pc gaming and/or dosbox!
Both fullfill my needs pretty well. The MiSTer uses an MT32 Pi for the
Roland MT32/Midi Emulation!
The Mt32Pi is also usable on a pc via some special midi connector hats,
if you are interested into getting midi in old dos games working!

Am 01.08.23 um 08:33 schrieb Werner P.:
> Am 27.07.23 um 16:02 schrieb Spalls Hurgenson:
>> I'm still debating whether owning so many retro-PCs is commendable,
>> because I'm repurposing hardware people were just tossing in the bin
>> and making it useful again, or despicable because it represents an
>> excess of materialism.
> I use a mister for retro pc gaming and/or dosbox!
> Both fullfill my needs pretty well. The MiSTer uses an MT32 Pi for the
> Roland MT32/Midi Emulation!
> The Mt32Pi is also usable on a pc via some special midi connector hats,
> if you are interested into getting midi in old dos games working!
>
Btw another new interesting option is PCEM, which also emulates a full
pc but goes way higher than dosbox in its compatibility list (3dfx
etc... is also supported) Also it is a lot easier to handle

https://pcem-emulator.co.uk/

On Tue, 1 Aug 2023 08:36:03 +0200, "Werner P." <werpu@gmx.at> wrote:
>Am 01.08.23 um 08:33 schrieb Werner P.:
>> Am 27.07.23 um 16:02 schrieb Spalls Hurgenson:

>> I use a mister for retro pc gaming and/or dosbox!
>> Both fullfill my needs pretty well. The MiSTer uses an MT32 Pi for the
>> Roland MT32/Midi Emulation!
>> The Mt32Pi is also usable on a pc via some special midi connector hats,
>> if you are interested into getting midi in old dos games working!
>Btw another new interesting option is PCEM, which also emulates a full
>pc but goes way higher than dosbox in its compatibility list (3dfx
>etc... is also supported) Also it is a lot easier to handle
>https://pcem-emulator.co.uk/

I'm a big fan of emulation. I've a DOSBox collection on disk with
every DOS game I own (and I own a lot of DOS games) installed on it.
It has a major advantage over running on real hardware in that you can
configure DOSBox - often from the command line itself - to simulate
different hardware and software combinations; thus, I can quickly
switch from playing "Ultima 7" (no memory manager) to "Tomb Raider"
(3DFX) to "Test Drive 3" (extremely finicky speed requirements)
without rebooting the emulator. For DOS-era games, it's probably the
best option.

PCEM is pretty good too, and I quite like it (so much so I posted
praising it a few months back). But it is not /quite/ as good a
platform for running games as DOSBox. For one thing, being a low-level
emulator, you're limited to a single hardware configuration per
session. AFAIK, doesn't (yet?) let you launch emulated apps directly
from the host's command line. And - even with a fast hostcomputer,
speed issues remain a problem with some games. PCEm is getting there,
but it's not quite the perfect replacement for real hardware.

But sometimes I just want to play with real hardware. These older
machines are much finickier beasts than modern PCs (oh, the joys of
IRQ sharing and memory management and motherboards that don't have any
embedded IO!) and I enjoy the challenge of working out the issues. Not
to mention the fun of trying to coax 20 or 30 year old hardware back
to life.

(albeit not always successfully: see earlier posts ;-)

So playing with real hardware is its own joy. When you hear it
running, with loud fans roaring, hard-disks clicking, and
floppy-dirives making their recognizable errrr-errrr-errrr seek
sound... it's just a delight. Plus, as good as emulation is, you're
limited to a much smaller selection of hardware; no Haupage TV Tuners
or VLB video cards, or MediaVision PAS-16 soundcards in PCEm.

But I get it if people prefer to avoid the the hassle of the hardware
and just use emulation. It is easier to get things running, it is
cheaper, and doesn't demand you dedicate a lot of desk space to a
bulky tower and CRT.

rridge@csclub.uwaterloo.ca (Ross Ridge) wrote:
>There's no "locating the MBR" going on here, the boot sector can only
>be in one place. Except for the last two bytes of boot sector of hard
>drives, the BIOS doesn't care whats written there. There can otherwise
>be complete garbage written on the first sector of the disk and the BIOS
>will still try to exectute it.

Spalls Hurgenson <spallshurgenson@gmail.com> wrote:
>ROM BIOS /do/ need to locate the MBR. It does so by checking the start
>of the boot sector device (head 1, cylinder 1, sector 1 in older CHS
>drives, or logical block address 0 in larger LBA drives). If the last
>two bytes in the code there has a signature value of 55AAh, it loads
>the code into RAM and the ROM transfers control to the code in the
>MBR.

Why are you repeating something I just said? Even in your own words the
BIOS is still not "locating the MBR". Maybe this bullshit like this
works to convince yourself that you're not wrong here, but there's no
need to waste my time or anyone else's with this self-delusional crap.

--
l/ // Ross Ridge -- The Great HTMU
[oo][oo] rridge@csclub.uwaterloo.ca
-()-/()/ http://www.csclub.uwaterloo.ca:11068/
db //