Has anyone gotten Sun's VirtualBox to do more than boot from
Myron A. Calhoun
XP Home machine. I've installed Sun's VirtualBox and it will boot
command.com from a floppy disk, but it cannot "see" any hard drive.
Has anyone gotten any farther with VirtualBox?
--
--Myron A. Calhoun.
Five boxes preserve our freedoms: soap, ballot, witness, jury, and cartridge
NRA Life Member & Certified Instructor for Rifle, Pistol, & Home Firearm Safety
Also Certified Instructor for the Kansas Concealed-Carry Handgun (CCH) license
Jim Leonard
wrote:
> I have a real DOS machine but would like to "combine" it with my Windoze
> XP Home machine. I've installed Sun's VirtualBox and it will boot
> command.com from a floppy disk, but it cannot "see" any hard drive.
>
> Has anyone gotten any farther with VirtualBox?
Have you run fdisk off of the floppy to partition the virtual hard
drive? Did you set up a virtual hard drive in VirtualBox?
DOS Guy
> I have a real DOS machine but would like to "combine" it with
> my Windoze XP Home machine.
Why don't you keep your machine's hard drive formatted as 100% FAT-32
and simply install XP-home on a FAT-32 volume (even the C:\ volume).
XP pro runs fine on FAT-32 (I do it all the time). I think NTFS is for
the birds (when you dig down, there are more downsides to NTFS vs FAT32
and very little benefit).
About a year ago I did exactly what you're looking for - I formatted a
blank 40 gb drive as a single FAT-32 volume, installed DOS 7 on it, and
then installed XP-pro on it. During startup, I get a choice to boot
into DOS or XP. No special boot-handlers needed.
And it's a breath of fresh air to be able to browse your XP's file
system from a pure DOS environment.
E. S. Fabian
| XP pro runs fine on FAT-32 (I do it all the time). I think NTFS is
| for the birds (when you dig down, there are more downsides to NTFS
| vs FAT32 and very little benefit).
I am not aware of any downside to NTFS, other than the lack of reliable
drivers for DOS and older Windows. OTOH I have been using the following NTFS
features available in WinXP for years:
1/ symbolic links (technically, junctions) - they provide access to a
subdirectory via multiple paths on the same drive, so collections of files
on a subtopic used by multiple main topics can be accessed from each main
topic as if it were local. It also makes access to any one of multiple
versions of a program easier. Note that in Vista and later versions of
Windows the symbolic links can cross drive boundaries. This allows you to
add a new drive, move some of your subdirectories to it, and create a
symbolic link on the original drive to the new subdirectories, and all your
programs will behave as if never moved them.
2/ hard links - they allow files to have multiple names. In the *nix world
grep, egrep and fgrep are a typical example - the interpretation of the
command line is vastly different based on which name you used to run the
program. I have many .BTM files where the invocation name substitutes for
option specifications. When you use multiple versions of a program suite,
for each file that is identical between the versions, hard links allow you
to use catalog those files in every version directory, but use diskspace
only for one.
Both of the above usages of NTFS features conserve diskspace, and improve
file maintenance. Two other NTFS features I have not yet used, because I did
not need (#3) or it is only available in XP Pro, which I just recently
acquired (#4):
3/ compressed files
4/ encrypted files
I am about to convert one or both of my pocket drives to NTFS to increase
capacity, using methods 1, 2 and 3; I intend to put encrypt all my personal
information (4).
--
Steve
Calvin
> XP pro runs fine on FAT-32 (I do it all the time). I think NTFS is for
> the birds (when you dig down, there are more downsides to NTFS vs FAT32
> and very little benefit).
FAT-32 is a plain BAD idea. It took the already fragile FAT-16 file
system and made it ten times WORSE !
NTFS is a proper full journaling file system designed for file safety,
and in the event of hardware failure or malfunction, enhanced file
recoverability.
Calvin.
DOS Guy
> > XP pro runs fine on FAT-32 (I do it all the time). I think NTFS
> > is for the birds (when you dig down, there are more downsides to
> > NTFS vs FAT32 and very little benefit).
>
> FAT-32 is a plain BAD idea. It took the already fragile FAT-16
> file system and made it ten times WORSE !
Fabian wrote that NTFS allows for hard and symbolic links - I know of
nobody that uses them (I'm not even sure how to make them without using
third-party application software). Compressed files are unnecessary -
and have been for years given the rate of increase in the size of hard
drives. Encrypted files are also rarely used by most people.
> NTFS is a proper full journaling file system designed for file
> safety, and in the event of hardware failure or malfunction,
> enhanced file recoverability.
You've got to understand the original purpose of NTFS back when it was
designed in the early 1990's.
Hard drives were less reliable than they would be by the early 2000's.
They did not have automatic bad-sector remapping, or in-drive caching.
Journaling in NTFS was designed to overcome the pathetic fault-tolerance
and failure rates of the hard drives of that age.
NTFS was initially going to find it's way onto servers, where I agree
that multi-threaded apps and general file-level coherency was going to
be important (and where files were more likely to cross the 4gb
threshold).
A lot of people have a misconception of journaling. If there is an
interruption or hardware failure that causes the system to shut down
(not just a failure that occurrs during a write operation), your file
system is returned to the same state it had before the interruption.
New data that existed during the interruption that was being written
will be lost. NTFS sacrifices orphaned data at the expense of
maintaining a "clean" cluster allocation. FAT32 can't
roll-back incomplete transactions, so data that was being written can be
recovered, but it comes because of the unintended creation of lost
clusters or chains which can lead to a "messy" cluster allocation but
rarely (if ever) a disfunctional file system.
The primary feature of NTFS is, by design, that you can't gain
file-system access without booting the GUI and logging into the system
(or at least that's how Microsoft deployed it). I don't particularly
care for that, given that I work in a home and soho environment.
If you want to read a very good comparative explanation about NTFS vs
FAT32, I suggest you read this:
http://cquirke.blogspot.com/2006/01/bad-file-system-or-incompetent-os.html
His summary:
--------------------
I would use NTFS where:
* Users have professional-grade IT admin, including backup
* Users need to hide data more than they need to salvage it
* Applications require files over 4G in size
* Hard drive exceeds the 137G barrier
--------------------
Item 3 is arguably rarely encountered.
Item 4 is patently false in all cases for the current revision level for
2K and XP, and can be false for win-98 given the correct hardware or
drivers.
But I fully agree with this:
"..I would avoid the use of NTFS in consumer PCs."
On a performance level, because NTFS is so "busy" doing it's journaling
and house-keeping, it takes a performance hit. There is no question
about this. Given the same hardware and (most) of the same tasks, FAT32
will outperform NTFS. Certainly on a single user home or SOHO
situation. Multi-user server applications are a different story. But
in that case, you're more likely to need NTFS's permission-based access
controls anyways.
It can also be argued that NTFS exposes the file system to more
vulnerability because of it's extra house-keeping activities. People
don't notice the consequences of that vulnerability because NTFS knows
how to clean up after itself.
Most IT people see every .chk file on a FAT32 drive as an example of
each time that FAT32 has "let the user down", without knowing what those
files really represent (rarely does it represent a munged file for which
it was the only or the best copy).
He goes into some detail to describe the consequences of having to
rebuild a file system under both NTFS and FAT32 - I hope you read them.
The take home message is that you have a better chance of rebuilding a
file system after a catastrophe if it's FAT vs NTFS. As well, NTFS
certainly doesn't do away with the concept of file fragmentation (a
drive can become quite frag'd under NTFS too).
Sure, FAT32 can be more "sloppy", maybe more often than we'd like.
Blame the OS, or faulty apps or hardware for that. But even if you
don't clean it as often as you should, sloppy rarely or never equates to
disfunctional. At least FAT32 knows when it should (or must) run
scandisk during startup - but NTFS will also throw up that requirement
in certain situations.
FAT32 also gets a bad rap because Microsoft chose to increase cluster
size along with volume size, which is completely unnecessary.
I've formatted up to 500 gb FAT32 volumes using 4kb cluster size and
installed both Win-98 and XP on such volumes (!). Win-98 functions
pretty well given that configuration - aside from the fact that defrag
and scandskw can't deal with so many clusters (but DOS checkdsk can).
Hans Kjaergaard
mcal...@sdf.lNoOnSePsAtMar.org (Myron A. Calhoun) wrote:
>I have a real DOS machine but would like to "combine" it with my Windoze
>XP Home machine. I've installed Sun's VirtualBox and it will boot
>command.com from a floppy disk, but it cannot "see" any hard drive.
>
>Has anyone gotten any farther with VirtualBox?
I use Dosbox (http://dosbox.com) for things like that, it is free and
easy to configure.
/Hans
Bill Buckels
"Hans Kjaergaard" <hans.k...@post5.tele.dk> wrote:
>I use Dosbox (http://dosbox.com) for things like that, it is free and easy
>to configure.
That link again is:
Doesn't everybody:)
If not they should. DosBOX has other features like screen capture etc.
It mounts directories of legacy programs as volumes making it easy to store
REALLY old programs neatly in subdirectories, and it is also cross-platform.
BTW my Clipshop program uses DosBOX to run some very old and cool DOS stuff:
However, VirtualBox does sound very good but perhaps not for DOS. Also
systems like Linux may wish to try http://dosemu.sourceforge.net/
Here's a little on that:
http://www.aztecmuseum.ca/docs/linux.txt
Bill
http://en.wikipedia.org/wiki/User:Bill_Buckels
http://www.cpm8680.com/
http://www.appleoldies.ca/
http://www.c64classics.ca/
Jim Leonard
> XP pro runs fine on FAT-32 (I do it all the time). I think NTFS is for
> the birds (when you dig down, there are more downsides to NTFS vs FAT32
> and very little benefit).
Wow, I guess you don't like retaining filesystem sanity after a power
outage. NTFS is journaled -- FAT32 isn't.
> About a year ago I did exactly what you're looking for - I formatted a
> blank 40 gb drive as a single FAT-32 volume, installed DOS 7 on it, and
You need to clarify which version of "DOS 7" as all regular commercial
releases of PC-DOS 7 did not support FAT32. If you're using something
that does support FAT32, please write what you're using.
Jim Leonard
> Hard drives were less reliable than they would be by the early 2000's.
> They did not have automatic bad-sector remapping, or in-drive caching.
> Journaling in NTFS was designed to overcome the pathetic fault-tolerance
> and failure rates of the hard drives of that age.
That's completely incorrect. A journaling filesystem does nothing at
all to prevent against faulty drive hardware; it prevents against
corruption caused by incomplete writes, ie. a power outage.
As for the rest of your summary, it essentially tries to justify your
position by stating that "because those are features I'd never use,
NTFS is worthless". Having had to deal with many unexpected outages
AT HOME, despite having both a household UPS and individual UPSes, I
hold data integrity very highly and obviously use journaling
filesystems. (I also keep at least two full backups, but that's
outside of the scope of this conversation.)
> A lot of people have a misconception of journaling. If there is an
> interruption or hardware failure that causes the system to shut down
> (not just a failure that occurrs during a write operation), your file
> system is returned to the same state it had before the interruption.
> New data that existed during the interruption that was being written
> will be lost. NTFS sacrifices orphaned data at the expense of
> maintaining a "clean" cluster allocation. FAT32 can't
> roll-back incomplete transactions, so data that was being written can be
> recovered, but it comes because of the unintended creation of lost
> clusters or chains which can lead to a "messy" cluster allocation but
> rarely (if ever) a disfunctional file system.
You write the above like it's a bad thing incomplete transactions are
discarded. You would rather keep them? Being incomplete, their
integrity is unverifiable.
FAT32 is not a copy-on-write filesystem, so lost cluster chains are
*bad*, not good.
> He goes into some detail to describe the consequences of having to
> rebuild a file system under both NTFS and FAT32 - I hope you read them.
That's a moot argument -- one should always have backups. I've lost
data on nearly every type of filesystem and media, so while I love
sophisticated filesystems (NOT NTFS -- more like XFS or ZFS), I am not
naive to trust them.
Lest you think me a windows-lover, I run a wide variety of OSes and
machines at home, including very many DOS machines for my hobby. To
the OP, I would recommend having a FAT16 bootable DOS partition, then
his XP NTFS partition on the same drive. I ran like that for many
years and could read/write the DOS files while in Windows, and could
boot to DOS when I wanted to run something that I couldn't run inside
windows or DOSBOX.
DOS-Guy
> You write the above like it's a bad thing incomplete transactions are
> discarded. You would rather keep them? Being incomplete, their
> integrity is unverifiable.
I've lost data on NT systems with failing power supplies and failing UPS
batteries *BECAUSE* of journaling. With FAT32 I could have recovered
that data.
I know I've lost data, because I obtained intermediate versions of
certain files (log files, notably) that the OS typically keeps open for
hours or days on end. When I compare the intermediate version with the
"final" version on disk, they are different.
Had the OS been FAT32, a simple scandisk would have either adjusted the
length of the file to include the last few clusters, or give me the
choice to convert those clusters to separate files where I could deal
with them later.
> FAT32 is not a copy-on-write filesystem, so lost cluster chains
> are *bad*, not good.
If you need copy-on-write (ie - in a server situation) then sure, NTFS
is more appropriate. But the main point is that for a consumer PC,
home, SOHO situation, NTFS gives next to no benefits to the user, but
can be much more difficult to rebuild if the drive experiences a logical
or even hardware failure). You will note that NTFS is proprietary, and
the number of recovery tools that exist for it pales in comparison to
FAT tools. You also say nothing about the performance hit and extra
drive activity caused by journalling.
> You need to clarify which version of "DOS 7" as all regular
> commercial releases of PC-DOS 7 did not support FAT32.
Here is where you can get more information and even download it:
ms-dos7.hit . bg
The OS described as MS-DOS 7 or 7.1 is basically all the dos files
stripped out of a system after win-98se or ME is installed. I guess
it's what Microsoft would have sold as a separate DOS version.
With that version, you can boot into a pure DOS environment and have
long file-name support.
DOS Guy
> I have a real DOS machine but would like to "combine" it with
> my Windoze XP Home machine.
I still think that the simplest (and best) solution is to simply install
XP on a FAT-32 volume after DOS has been installed on the same drive.
XP will automatically create a boot menu, allowing you to boot into DOS
or XP at system start-up.
And I recommend DOS 7.1 because it has long file-name support.
And if you don't like the way that Microsoft's format program creates
FAT-32 volumes with large cluster size, then use one of the third-party
drive preparation tools and create a FAT-32 volume with the cluster size
(allocation unit) that you want. Get Hiren's Boot CD for that (lots of
drive tools, utilities, etc).
Rugxulo
On Oct 25, 10:12 pm, DOS Guy <D...@Guy.com> wrote:
>
> Fabian wrote that NTFS allows for hard and symbolic links - I know of
> nobody that uses them (I'm not even sure how to make them without using
> third-party application software).
Vista (and presumably Win7, which I haven't tried) has MKLINK, but I
don't think Linux's NTFS driver handles those, so you have to use the
original path. The weird thing about symlinks is that DOS apps can
access the original files through them, but their size in a dir
listing shows up as zero.
> > NTFS is a proper full journaling file system designed for file
> > safety, and in the event of hardware failure or malfunction,
> > enhanced file recoverability.
>
> You've got to understand the original purpose of NTFS back when it was
> designed in the early 1990's.
Wasn't NTFS more of a successor to HPFS, which itself was an alleged
huge improvement over FAT?
http://en.wikipedia.org/wiki/HPFS
> NTFS was initially going to find it's way onto servers, where I agree
> that multi-threaded apps and general file-level coherency was going to
> be important (and where files were more likely to cross the 4gb
> threshold).
The overhead of NTFS is allegedly quite high compared to FAT, hence
why MS whipped up exFAT for intended use on flash drives.
> The primary feature of NTFS is, by design, that you can't gain
> file-system access without booting the GUI and logging into the system
> (or at least that's how Microsoft deployed it). I don't particularly
> care for that, given that I work in a home and soho environment.
To MS, cmdline and TUI are and have always been obsolete, so they shun
that as much as possible. Plus, let's face it, most people hate
"arcane" DOS and it's alleged troubles. It's up for debate whether
they're the crazy ones or we are. :-)
> I would use NTFS where:
>
> * Applications require files over 4G in size
> * Hard drive exceeds the 137G barrier
Well, obviously the OS dictates what you can use, and even NTFS has
various subversions available (I think Vista is 5, but don't quote me
on that). So you're unlikely to be able to use NTFS (or ext2 or HPFS)
natively on any DOS. While there are 3rd-party tools, most (all?)
aren't quite good enough. So you can't choose a file system on its own
merits, you have to use what is supported. I know you know this, just
saying ... it's less ideal than it could be. For instance, Vista won't
boot off a FAT partition while XP will. Oh, and don't forget that XP
etc. can access (?) but refuse to create > 32 GB FAT32 partitions "for
speed reasons".
EDR-DOS tried to invent an extension called FAT+ for > 4 GB files, and
I think Udo collaborated with some FreeDOS folks on a rough spec, but
they never implemented it (too few volunteers). Even FreeDOS' FAT32 is
still 2 GB limited per individual file.
> But I fully agree with this:
>
> "..I would avoid the use of NTFS in consumer PCs."
>
> On a performance level, because NTFS is so "busy" doing it's journaling
> and house-keeping, it takes a performance hit. There is no question
> about this. Given the same hardware and (most) of the same tasks, FAT32
> will outperform NTFS.
Win95 could run (barely) on 4 MB of RAM, I think, but NT always needed
at least 16 MB or so. So yeah, there was a tradeoff. And we all know
that NTVDM isn't nearly as good at DOS emulation, but it's more stable
overall (probably due to better .DLL hell protection).
> As well, NTFS certainly doesn't do away with the concept of file fragmentation
> (a drive can become quite frag'd under NTFS too).
I think Vista regularly defragments in the background once every week
for me.
> FAT32 also gets a bad rap because Microsoft chose to increase cluster
> size along with volume size, which is completely unnecessary.
>
> I've formatted up to 500 gb FAT32 volumes using 4kb cluster size and
> installed both Win-98 and XP on such volumes (!). Win-98 functions
> pretty well given that configuration - aside from the fact that defrag
> and scandskw can't deal with so many clusters (but DOS checkdsk can).
Yes, that's a problem. Of course, the main issue for us is that no DOS
natively accesses NTFS, no FAT partitions exist on most installs by
default, no easy way to resize NTFS (until Vista), and you can only
convert one way: FAT to NTFS. In other words, it's their way or the
highway.
Ross Ridge
>The overhead of NTFS is allegedly quite high compared to FAT, hence
>why MS whipped up exFAT for intended use on flash drives.
And why Microsoft also uses FAT derived filesystems on their Xbox
consoles.
Ross Ridge
--
l/ // Ross Ridge -- The Great HTMU
[oo][oo] rri...@csclub.uwaterloo.ca
-()-/()/ http://www.csclub.uwaterloo.ca/~rridge/
db //
E. S. Fabian
| Rugxulo <rug...@gmail.com> wrote:
|| The overhead of NTFS is allegedly quite high compared to FAT, hence
|| why MS whipped up exFAT for intended use on flash drives.
|
| And why Microsoft also uses FAT derived filesystems on their Xbox
| consoles.
For a game, speed is everything. For a system used for important or critical
information, dependability is much more important. I'll stick with NTFS.
However, as I am about to reformat one of my drives, I may create a DOS
partition, so I can run some of my old DOS programs better.
<><
Steve
DOS Guy
> For a system used for important or critical information,
> dependability is much more important.
I think recoverability is more important. FAT32 is a more recoverable
file system. NTFS is harder to reconstruct given a certain (and not
necessarily high) amount of dammange is done to it (and more expensive
to reconstruct / repair).
Both file systems are dependable.
In fact, given the succeptibility for win-9x/me to behave in an unstable
manner in some circumstances, FAT32 functions very well and has shown me
that is can recover itself with basically zero lost files the vast
majority of the time.
I also think accessibility is important. A fat32 volume is more
accessible, from more platforms, under more conditions (user rights /
permissions, hidden files, directories and other such games) vs ntfs.
I think simplicity is important. FAT32 is a simple file system.
Microsoft has shown that it's unofficial motto - "If it works, it's not
complicated enough" to be a ridiculous and costly philosophy.
I equate simplicity with dependability.
NTFS is more suitable as a file-server or for handling large database
files - but that is a choice made from a practicality or performance
(not dependability) POV.
If you want dependability, choose your hard drive manufacturer
carefully, as well as the motherboard and power supply of your system.
bruce
> > For a system used for important or critical information,
> > dependability is much more important.
journaled. I've just recovered from a munged disk by reformatting and
reinstalling the OS! No power failure or HW problem, just filled it
too full at 95%.
Just my 2 cents.
Ross Ridge
> And why Microsoft also uses FAT derived filesystems on their Xbox
> consoles.
E. S. Fabian <ESFa...@comcast.net> wrote:
>For a game, speed is everything. For a system used for important or critical
>information, dependability is much more important. I'll stick with NTFS.
Despite what has been said in this thread, NTFS is not a full journaling
filesystem, it only journals filesystem meta-data. It doesn't journal
your data, so your important or critical information doesn't get any
special protection. Even as far as integrity of the file system, they
end up being just as reliable. With NTFS, Windows is able to use the
journal to quickly restore filesystem integrity after a power failure,
while with FAT Windows has to do a perform a full file system check to
the same job. The net result ends up being the same. Both filesystems
are in consistant state after booting, and neither provides any guarantees
that your data remains consistant if you didn't shut down properly.
Now since doing a full filesystem check on a large FAT filesystem can
take a stupidly long time, that's reason enough for me to recommend
installing Windows XP to NTFS partition. You're also probably going to
find more and more software is going to assume NTFS.
>However, as I am about to reformat one of my drives, I may create a DOS
>partition, so I can run some of my old DOS programs better.
I have a FAT16 boot partition, a FAT32 partition for games, and an NFTS
partition for Windows XP. It's possible to install Windows to a different
partition than the boot partition, so if the original poster wants to
the use boot menu to boot into MS-DOS and he can do so and still use NTFS.
Rugxulo
On Oct 26, 7:28 am, Jim Leonard <mobyga...@gmail.com> wrote:
>
> > About a year ago I did exactly what you're looking for - I formatted a
> > blank 40 gb drive as a single FAT-32 volume, installed DOS 7 on it, and
>
> You need to clarify which version of "DOS 7" as all regular commercial
> releases of PC-DOS 7 did not support FAT32. If you're using something
> that does support FAT32, please write what you're using.
MS-DOS 7 from Win95 before OSR2.5 didn't support FAT32, no, but I
think later versions did (though I could be wrong).
As far as PC-DOS, I thought I told you before about what Lucho
"found", but I haven't tried it myself as I have no interest in it
(FreeDOS works for my needs):
=============================================
"PC-DOS v7.1 is present in the IBM ServerGuide Scripting Toolkit
(MIGR-53564) which can freely be downloaded, starting at
http://www.ibm.com/systems/management/sgstk.html "
"PC-DOS is Da Best DOS kernel with LBA/FAT32 support, running even on
8086"
=============================================
In other words, I don't know how legal it is, so caveat emptor.
Calvin
> I have a FAT16 boot partition, a FAT32 partition for games, and an NFTS
> partition for Windows XP. It's possible to install Windows to a different
> partition than the boot partition, so if the original poster wants to
> the use boot menu to boot into MS-DOS and he can do so and still use NTFS.
This is the method I recommend as well.
See http://nt4ref.zcm.com.au/bigdisk.htm for details.
Calvin.
DOS Guy
> Despite what has been said in this thread, NTFS is not a full
> journaling filesystem, it only journals filesystem meta-data.
> It doesn't journal your data
Would IIS server log files be considered "meta-data", or "my data" ?
IIS log file data does become lost when an NT4 system is shut-down
unexpectedly (the log files are journalled back to their last
checkpoint).
> so your important or critical information doesn't get any special
> protection.
Journalling doesn't "protect" user information contained in files.
Journalling maintains the coherency of files or file structure (metrics
such as file size, cluster composition, etc) - not the file-data that
they may have once contained.
Every time that journalling kicks in because of some system event or
malfunction, the result will either be inconsequential to your data (no
difference) or the result will be loss of user data.
Journalling will sacrifice the last data written to a file in order to
maintain what it thinks is the proper coherency of the file tables.
Journalling is a *active* process. It doesn't happen passively. You
could, in theory, impliment journalling in FAT32 by having the OS
impliment extra structures and constructs to contain file checkpoints,
semaphores and ghost copies.
The extra read and write operations that are performed in the name of
NTFS journalling do take a toll on hard drive operational lifespan, as
do ALL or ANY read / write operations. The goal of any file system that
operates on a magneto-optical-mechanical storage device should be to
minimize the mechanical operations of the device when performing a given
read or write task.
Similarly, NTFS is not recommended for flash memory drives because the
memory elements of those drives actually have a finite life span
(re-write-ability) and again NTFS is notorious for causing arguably an
unnecessary amount of those activities in the name of journalling.
> Now since doing a full filesystem check on a large FAT
> filesystem can take a stupidly long time,
Only if you elect to perform a "surface" test (ie a check for bad
sectors) which is a legacy operation based on hard drives of 10+ years
ago that did not monitor themselves and automatically remap bad sectors
for good ones held in reserve areas.
On this system, I have a 32 gb FAT32 c: partition with 120k files in
7,300 directories. It takes about 7 minutes to perform a DOS scandisk
operation on it. I would not call that a stupidly long time.
> that's reason enough for me to recommend installing Windows
> XP to NTFS partition.
I have seen XP take a pathetically long time (30+ minutes) to perform
it's version of chkdsk on 20 gb NTFS volumes.
So your reason to use NTFS is not valid from my experience.
> You're also probably going to find more and more software
> is going to assume NTFS.
I am aware of no user application software that is aware of what file
system it is running on, save for the fact that some video editing
software seems to know that it must truncate or span the 4 gb file-size
barrier when writing to a FAT32 volume.
> I have a FAT16 boot partition, (...)
Why would anyone have a FAT16 partition in this day and age?
> if the original poster wants to the use boot menu to boot
> into MS-DOS and he can do so and still use NTFS.
And still be stuck looking for a DOS NTFS driver if he needs to be able
to access the NTFS volume from DOS.
DOS Guy
> > You need to clarify which version of "DOS 7" as all regular
> > commercial releases of PC-DOS 7 did not support FAT32.
>
> MS-DOS 7 from Win95 before OSR2.5 didn't support FAT32, no, but I
> think later versions did (though I could be wrong).
What are you people talking about?
MS-Dos 6.22 was the last real, stand-alone version of DOS sold by
Microsoft, and it supports FAT32.
The thing called "DOS 7" or 7.1 that I'm referring to is a construct put
together by some people from the DOS files contained in the release of
Win-98se and Me.
This "MS-DOS 7" is theoretically what Microsoft would have created had
it elected to release a successor to 6.22.
For some reason my NNTP server flags the following URL as "banned", but
it's a harmless URL:
ms-dos7.hit . bg
Remove the spaces between hit and bg and go to that webpage.
You will get a complete description of this "MS-DOS 7.1" and will be
able to download it.
Jim Leonard
> I think recoverability is more important. FAT32 is a more recoverable
> file system. NTFS is harder to reconstruct given a certain (and not
> necessarily high) amount of dammange is done to it (and more expensive
> to reconstruct / repair).
This makes no sense. It is much harder to damage an NTFS filesystem
due to the fact it journals, so there is less need to recover in the
first place.
It sounds like you'd rather drive a car that breaks down every weekend
as long as it's easy to fix roadside. Me, I'd rather drive a car that
breaks down only once a year, regardless of how badly it breaks down.
> I also think accessibility is important. A fat32 volume is more
> accessible, from more platforms, under more conditions (user rights /
> permissions, hidden files, directories and other such games) vs ntfs.
Because there are NO user rights, permissions, etc. So yeah, it does
a lot less, so of course it's easy to read.
> I think simplicity is important. FAT32 is a simple file system.
Simple is not always best. There is a right tool for every job; FAT32
is the right tool for some jobs but it is not the right tool for every
job.
H-Man
>
> MS-Dos 6.22 was the last real, stand-alone version of DOS sold by
> Microsoft, and it supports FAT32.
>
I'd be interested how you managed that bit of magic ;)
AFAIK there are drivers available that allow access to LFNs but there was
nothing like that AFAIR when 6.22 was released, and was not made available
by MS. If you somehow managed to get 6.22 installed on a FAT32 partition,
I'd be interested in knowing how to go about that.
--
HK
Rugxulo
On Oct 28, 8:43 am, DOS Guy <D...@Guy.com> wrote:
>
> Why would anyone have a FAT16 partition in this day and age?
I can't speak for everyone else, but I don't resize / repartition a
lot, so I just leave whatever's there. Sometimes that's the easiest
(or even) best option, esp. if the DOS in question is too old to
support FAT32 (and per chance runs something that FreeDOS doesn't,
rare but still possible).
But I will openly say FAT16 sucks for > 512 MB partitions. Better to
create and share more than one or go entirely to FAT32 (if possible).
Rugxulo
On Oct 28, 8:52 am, DOS Guy <D...@Guy.com> wrote:
> Rugxulo wrote:
>
> > MS-DOS 7 from Win95 before OSR2.5 didn't support FAT32, no, but I
> > think later versions did (though I could be wrong).
>
> MS-Dos 6.22 was the last real, stand-alone version of DOS sold by
> Microsoft, and it supports FAT32.
MS arguably still sells DOS (on their developer network thingy or
whatever it's called). MS-DOS 6.22 may be the last standalone version,
but Win9x had MS-DOS 7.10 (and WinMe had 8.00). And no, I don't think
any MS-DOS before Win95 supported FAT32 (or LFNs). Some other third-
party DOSes do, but 6.22 does not. Heck, you can't even install MS-DOS
except on the very beginning of a hard drive. DR-DOS, while better,
also doesn't (officially, i.e. without the TSR hack) support FAT32,
and it has trouble recognizing *any* FAT partitions found before the
active / bootable one. FreeDOS has none of those issues, but even it
is (for now) limited to 2 GB per file even on FAT32.
As MS later admitted, there was no technical reason to not ship it
separately from Windows at the time. It just looks like it's more
integrated, but you can indeed easily strip out the DOS-only parts for
raw booting if you want.
> For some reason my NNTP server flags the following URL as "banned", but
> it's a harmless URL:
>
> ms-dos7.hit . bg
>
> Remove the spaces between hit and bg and go to that webpage.
>
> You will get a complete description of this "MS-DOS 7.1" and will be
> able to download it.
That's not legal, as most of us are well aware. Hence we don't
recommend that, even if you erroneously think they don't sell it
anymore. (Even if they didn't, that still wouldn't be legal.)
It's better to just use FreeDOS (or even Linux or *BSD or Menuet or
OctaOS ...) if DOS freeware doesn't do what you want.
DOS Guy
> > I think recoverability is more important. FAT32 is a more
> > recoverable file system.
>
> This makes no sense. It is much harder to damage an NTFS filesystem
> due to the fact it journals, so there is less need to recover in the
> first place.
Read this:
http://cquirke.blogspot.com/2008/03/ntfs-vs-fatxx-data-recovery.html
And then tell me what your response is.
The summary of that material is:
-----------
Given the poor results I see when recovering data from NTFS, I'd have to
recommend using FATxx rather than NTFS as a data survivability
strategy. If readers can attain better results with other recovery
tools for NTFS, then please describe your mileage with these in the
comments section!
-----------
> It sounds like you'd rather drive a car that breaks down every
> weekend as long as it's easy to fix roadside.
No.
I'd rather drive a car that even if it breaks as often as another car,
it will rarely break as badly, or as expensively, as the other car.
> Me, I'd rather drive a car that breaks down only once a year,
> regardless of how badly it breaks down.
Unlike most IT-centric people who migrated away from Win-9x the moment
that win-2k came out, I continued to run win-9x on dozens of PC's during
the past 10 years, and I've seen the huge improvements in performance
and stability that came with better hardware, more system memory, better
motherboards and video cards, and better drivers for win-9x towards the
end of it's commercial life (circa 2006).
Those that left win-98 back in 2000 or 2001 have only bad memories of an
OS trying to run on 16 or 32 mb of memory with buggy AGP video drivers
that left their system hanging and resulted in many scandisk sessions
and .chk files.
I have never lost data on a FAT-32 drive due to logical file-table
errors that could not be fixed or repaired. For the past 5 years I
really haven't had to run scandisk on any win-98 system - period.
And yes, it would matter more to me how badly a file system could break
vs how often it does break, because eventually you will encounter the
"bad" break when you can least afford it and possibly not be able to
recover from it.
> > I also think accessibility is important. (user rights /
> > permissions, hidden files, directories and other such
> > games)
>
> Because there are NO user rights, permissions, etc.
> So yeah, it does a lot less, so of course it's easy
> to read.
It does exactly what it's supposed to do - which is to store files and
not play games with me, the system user and owner. User permissions,
rights, etc, have no place on a consumer desktop or laptop PC.
The concept is absurd, always has been, but Micro$haft had no choice
when they took their corporate / institutional / gov't -certified OS (NT
and it's derivatives) and shoved it down consumer's throats. And it did
consumers absolutely zero good having those "features" when their
machines got hacked and became spam zombies during 2003 - 2006.
> Simple is not always best. There is a right tool for every
> job; FAT32 is the right tool for some jobs but it is not
> the right tool for every job.
FAT32 is not a tool. It's a specification.
FAT32 is no more a tool than my tool-drawer in my tool-chest is a tool.
DOS Guy
> > MS-Dos 6.22 was the last real, stand-alone version of DOS
> > sold by Microsoft, and it supports FAT32.
>
> I'd be interested how you managed that bit of magic ;)
Aaak.
Ok, that was totally wrong.
I must have had this unconcious link in my head that 6.22 was more
recent that it actually is.
> AFAIK there are drivers available that allow access to LFNs but
> there was nothing like that AFAIR when 6.22 was released, and
> was not made available by MS. If you somehow managed to get 6.22
> installed on a FAT32 partition, I'd be interested in knowing how
> to go about that.
No, I haven't messed with DOS 6.22 since before win-95 came out.
This material might be useful:
http://www.cn-dos.net/msdos71/index.htm
Whenever I want to put DOS on a machine, I simply throw a floppy in my
PC, open a command shell and enter "format a: /s" and maybe throw fdisk
and format.exe on the floppy and I'm ready to deal with the other
machine.
Ross Ridge
> Despite what has been said in this thread, NTFS is not a full
> journaling filesystem, it only journals filesystem meta-data.
> It doesn't journal your data
DOS Guy <D...@Guy.com> wrote:
>Would IIS server log files be considered "meta-data", or "my data" ?
Your data, obviously.
>> so your important or critical information doesn't get any special
>> protection.
>
>Journalling doesn't "protect" user information contained in files.
If NTFS journaled data, it would provide the same consistancy guarantees
to data as it does to its own meta-data. A data write operation would
either entirely committed to the disk or not happen at all, and if it
does get comitted all previously issued writes would also be guaranteed
comitted in their entirety. Databases and other important and critical
forms of information storage often need properties like this, so they
have to implement there own measures to ensure data consistancy under
NTFS and FAT.
>On this system, I have a 32 gb FAT32 c: partition with 120k files in
>7,300 directories. It takes about 7 minutes to perform a DOS scandisk
>operation on it. I would not call that a stupidly long time.
Most users would consider a boot time of over seven minutes a stupidly
long time time to have to wait.
>So your reason to use NTFS is not valid from my experience.
You're free to do what ever you want on your own systems, but I don't
think anyone is taking your advice here seriously.
DOS Guy
> > http://ms-dos7.hit . bg
(for some reason, hit dot bg is listed as a "banned" domain by uribl.com
and I must mung it to get past the censorship being applied by this nntp
server)
> > You will get a complete description of this "MS-DOS 7.1" and
> > will be able to download it.
>
> That's not legal, as most of us are well aware. Hence we don't
> recommend that, even if you erroneously think they don't sell
> it anymore.
My philosophy is that Microsoft is a criminal organization - and you
can't really steal from a criminal.
Besides, I would think that most IT people (or at least the people
reading these newsgroups) would have a few bootable "win-98" floppys
hanging around. If you've got that, then you've got your FAT-32
compatible DOS right there, ready to take to another PC and install.
Is it legal? Maybe not, but neither is J-walking or spitting on the
sidewalk.
Note that when you format a volume with /s command, or when you transfer
the DOS system files using sys, you are not prompted with a EULA message
or an "I agree" query ...
I find the concept of paying for DOS (MS-dos) to be bizzare or strange.
It's like buying a PC and having to go out and pay for the bios.
Here's another way to look at it.
If Microsoft sold (in total) 1 million copies of DOS (say, DOS 6.22)
then they have nothing to complain about if people trade and copy and
give it to each other, as long as the total number of systems running
DOS 6.22 never exceeds 1 million at any point in time.
Doesn't have to be DOS 6.22. It could be windows 98, XP, etc. The
concept is the same.
DOS Guy
> > Would IIS server log files be considered "meta-data", or "my
> > data" ?
>
> Your data, obviously.
Then my experience is that journalling does not "protect" user data that
has been written to a file.
> > On this system, I have a 32 gb FAT32 c: partition with 120k files
> > in 7,300 directories. It takes about 7 minutes to perform a DOS
> > scandisk operation on it. I would not call that a stupidly long
> > time.
>
> Most users would consider a boot time of over seven minutes a
> stupidly long time time to have to wait.
Who says my system must, or does, run scandisk every time it boots?
In fact, it does not. Why do you think it does?
Scandisk is run if or when I tell it to run, or it will run upon startup
in the VERY rare case that a flag is set somewhere in the mbr as a
response to an improper shutdown or system lockup.
> ... but I don't think anyone is taking your advice here
> seriously.
There once was an emperor that had no clothes, but the people didn't
want to admit it to each other because of peer pressure and fear of
group-rejection.
Ross Ridge
> Most users would consider a boot time of over seven minutes a
> stupidly long time time to have to wait.
DOS Guy <D...@Guy.com> wrote:
>Who says my system must, or does, run scandisk every time it boots?
No one.
>> ... but I don't think anyone is taking your advice here
>> seriously.
>
>There once was an emperor that had no clothes, but the people didn't
>want to admit it to each other because of peer pressure and fear of
>group-rejection.
If you took the time to get your facts correct and didn't misrepresent
what other people have said, you might be taken more seriously.
DOS Guy
> If you took the time to get your facts correct and didn't
> misrepresent what other people have said, you might be
> taken more seriously.
Why don't you give us your comments on this material:
http://cquirke.blogspot.com/2008/03/ntfs-vs-fatxx-data-recovery.html
H-Man
> H-Man wrote:
>
>>> MS-Dos 6.22 was the last real, stand-alone version of DOS
>>> sold by Microsoft, and it supports FAT32.
>>
>> I'd be interested how you managed that bit of magic ;)
>
> Aaak.
>
> Ok, that was totally wrong.
>
> I must have had this unconcious link in my head that 6.22 was more
> recent that it actually is.
>
No sweat, could happen to anyone.
--
HK
Jim Leonard
> Read this:
>
> http://cquirke.blogspot.com/2008/03/ntfs-vs-fatxx-data-recovery.html
>
> And then tell me what your response is.
His personal observations don't factor into the conversation. Both he
and you are focusing on a tiny tiny part of a filesystem -- under
normal circumstances, *the part you see the least*. I have backups
for when a filesystem fails, so that I don't have to dick around with
the "what did or didn't get corrupted" guessing game.
If your needs never grow beyond 4G file limits, compression,
encryption, or security, then stick with FAT32. Just don't recommend
it as the best filesystem for all needs.
> And yes, it would matter more to me how badly a file system could break
> vs how often it does break, because eventually you will encounter the
> "bad" break when you can least afford it and possibly not be able to
> recover from it.
That's what backups are for.
> It does exactly what it's supposed to do - which is to store files and
> not play games with me, the system user and owner. User permissions,
> rights, etc, have no place on a consumer desktop or laptop PC.
Unless you share your machine with other members of the family.
Jim Leonard
He won't, but I will: I get the impression from that blog post that
he has spent a ton of time recovering files because he had a poor
backup strategy, and he is trying to justify his poor backup strategy
by promoting the filesystem that recovers easily from crashes.
Jim Leonard
> My philosophy is that Microsoft is a criminal organization - and you
> can't really steal from a criminal.
Then why should we take anything you say seriously? I think your
advice to run FAT32 against all odds is criminally negligent to anyone
who relies on your advice to keep their data integrity high; does that
mean it's ok to steal from you?
> Note that when you format a volume with /s command, or when you transfer
> the DOS system files using sys, you are not prompted with a EULA message
> or an "I agree" query ...
Nevertheless, the EULA is still attached to the product. Doesn't
matter if it prints out in big flashing letters when you use format.
(BTW, it *is* displayed when you install DOS 5.0 or later using the
bootable setup disks.)
> I find the concept of paying for DOS (MS-dos) to be bizzare or strange.
> It's like buying a PC and having to go out and pay for the bios.
Right, because who should deserve to get paid for their work?
> If Microsoft sold (in total) 1 million copies of DOS (say, DOS 6.22)
> then they have nothing to complain about if people trade and copy and
> give it to each other, as long as the total number of systems running
> DOS 6.22 never exceeds 1 million at any point in time.
You're under the impression that all software is a physical product
with second-sale rights, like books. Not all software is. When you
buy most commercial software, you're actually purchasing a license to
the use the software under certain conditions, not the physical media
the software comes on.
If you don't like this and wish to cease being a hypocrite, then stop
running MSDOS and start running a free OS like FreeDOS. Rugxulo would
be happy to set you up with a distribution. There's even an 8086
variant that, while slow, does indeed run on any 808x or 80286 machine
and still gives you FAT32.
H-Man
Really??
Anyone buying any of this?
--
HK
E. S. Fabian
| DOS Guy wrote:
...
|| My philosophy is that Microsoft is a criminal organization - and you
|| can't really steal from a criminal.
Are you allowed to kill a criminal?
|| I find the concept of paying for DOS (MS-dos) to be bizzare or
|| strange. It's like buying a PC and having to go out and pay for the
|| bios.
You do, when you buy the PC. The price of the BIOS is part of the price of
the PC.
|| Here's another way to look at it.
||
|| If Microsoft sold (in total) 1 million copies of DOS (say, DOS 6.22)
|| then they have nothing to complain about if people trade and copy
|| and give it to each other, as long as the total number of systems
|| running DOS 6.22 never exceeds 1 million at any point in time.
||
|| Doesn't have to be DOS 6.22. It could be windows 98, XP, etc. The
|| concept is the same.
|
| Really??
| Anyone buying any of this?
Not many. "DOS Guy" must believe that the people who developed MS-DOS don't
need food, clothing, shelter, schools for their children, etc.
--
Steve
<><
DOS Guy
> Are you allowed to kill a criminal?
Apparently the state can - if that criminal has been convicted of
murder.
And in some cases, you can use deadly force on someone who is
threatening you with bodily harm (eg home robery, home invasion)
> Not many. "DOS Guy" must believe that the people who developed
> MS-DOS don't need food, clothing, shelter, schools for their
> children, etc.
If Microsoft sells 1 million licenses of something, and if there is
never more than 1 million licenses (or copies of) in use at any one
time, then microsoft has suffered no economic harm or dammage in that
situation.
Further more, if Microsoft elects to no longer sell licenses for a given
product (ie - windows 98) then they suffer no economic harm if that
product is downloaded, copied, passed around at will, etc.
This raises the concept of a license that becomes dormant or abandoned.
I could say that my use of a particular Windows 98 product key
represents an abandoned license that Microsoft did receive payment for
once upon a time.
Again, as long as the number of systems using a licensed product does
not exceed the number of licenses sold, then you can't demonstrate
economic harm or loss to the vendor.
DOS Guy
> > My philosophy is that Microsoft is a criminal organization -
> > and you can't really steal from a criminal.
>
> Then why should we take anything you say seriously?
My criticism of Microsoft and their criminal operating methods has got
nothing to do with technical analysis and end-user experiences of their
products or constructs such as file systems.
If you are a Microsoft psycophant and appologist, then fine - you love
Microsoft and everything they do and how they do it. So we disagree
about Microsoft the corporate entity.
But that's on a different scale than any technical discussion about
their products. It's sad that you appear to not be able to
differentiate between the two realms.
> I think your advice to run FAT32 against all odds is criminally
> negligent to anyone who relies on your advice to keep their
> data integrity high; does that mean it's ok to steal from you?
You would be incorrect to ascribe a form of "criminal" negligence to my
IT practices as they pertain to desktop file systems. There is nothing
in any criminal code of any jursidiction that criminalizes a user's
choice of computer file system. So your premis is poorly crafted. I
can't be a criminal if my choice of file system (and it's possible
implications) is not a crime. Hence it's not ok to steal from me.
And my advice to "run FAT32" is not against any odds. This is not a
race track or casino we're talking about here.
> > I find the concept of paying for DOS (MS-dos) to be bizzare
> > or strange. It's like buying a PC and having to go out and
> > pay for the bios.
>
> Right, because who should deserve to get paid for their work?
How many times should Microsoft continue to get paid for the same
product? Possibly one they don't even sell any more? The programmers
don't get paid on a royalty basis. Maybe they should...
> You're under the impression that all software is a physical
> product with second-sale rights, like books. Not all software
> is.
Maybe not all software, but what about specifically DOS? Or Windoze?
> If you don't like this and wish to cease being a hypocrite,
> then stop running MSDOS and start running a free OS like
> FreeDOS.
My OS of choice, on many of the systems I interact with, is Windows
98se, all of which run Office 2k Premium. That combination, along with
some accounting and contact management software circa 2000 - 2002
satisfies all of our small business operating needs. Our developers run
XP and various virtual platforms. But I digress. As someone who's main
OS is win-98, I feel that my "shadow" OS is DOS, and I will boot into
DOS on any system I choose when I have to or need to.
We have a few technical / bench systems that run custom software in DOS,
and those machines do boot directly into DOS 7 (the win-98 version of
dos) but they don't actually have win-98 installed on them.
So my dos needs are met with DOS 7, as created simply by "format a: /s"
from a win-98 command shell. If I copy all of the c:\windows\command
dos files onto the floppy, and then copy all the remaining dos 6.22
files, then I have a full DOS if I need it. It's second nature - I
don't give it a second thought.
Klaus Meinhard
> This raises the concept of a license that becomes dormant or
> abandoned. I could say that my use of a particular Windows 98 product
> key represents an abandoned license that Microsoft did receive
> payment for once upon a time.
Now you're weaseling into the "abandonware" realm.
While I may have some sympathy for the concept, fact is that your
personal justification for infringement of copyright and license law is
still a breach of the law. Copyright and licensing laws are what they
are, until you succeed in changing them (call your congress man). Until
then your practice is against these laws, and may border on the
criminal.
--
Best Regards,
* Klaus Meinhard *
<www.4dos.info>
DOS Guy
> Copyright and licensing laws are what they are, until you succeed
> in changing them (call your congress man). Until then your practice
> is against these laws, and may border on the criminal.
I think there is a practical if not legal difference if I were to
re-package some old copy of DOS or Windoze and sell it to someone else
vs installing as many copies of those products as I want on my own
computers for my own use.
In the latter case, Micro$haft will probably never discover what I've
done, but nor could they prove in a court of law that I did not have a
license for each installation. I don't think that Macro$haft has ever
prosecuted (let alone won) such a case.
If Milkro$oft wanted to secure a proper paper trail between themselves
and end-users of their product, then they should have implimented a
different stratagy for distributing their software and not go through
countless intermediary agents. The fact is that their licensing scheme
is so watered down and invisible as to be meaningless and irrelavent to
end users.
DOS Guy
> > And then tell me what your response is.
>
> His personal observations don't factor into the conversation.
His experiences with NTFS are no different than anyone else's. NTFS
doesn't behave differently depending on who is using it.
It's just that he bothers to investigate the working details and pull
the cover off the charade that is NTFS.
> Both he and you are focusing on a tiny tiny part of a filesystem
> -- under normal circumstances, *the part you see the least*.
And it's the very part that you NTFS supporters constantly point out as
the main advancement of NTFS over FAT32.
> I have backups for when a filesystem fails, so that I don't
> have to dick around with the "what did or didn't get corrupted"
> guessing game.
What a joke.
NTFS is supposed to be the superior file system, with fallbacks and
safeguards and extra complexity built-in.
You won't let me use the excuse that I can back up my FAT32 file system
as a fall-back to it's supposed problems, so why should I let you do the
same with NTFS?
> If your needs never grow beyond 4G file limits,
Many apps limit their user files to 4gb, even under XP, vista, etc.
Mailbox size for example.
> compression,
Who uses compression these days (or - this decade?)
> encryption,
Tell me that the FBI, NSA, CIA, DEA, etc, can't de-crypt your hard drive
when they come knocking at your door. Those gov't agencies were given
the back door into your encrypted files by Meekro$haft in exchange for
dropping the antitrust suit. Next time you return to the US and have to
get through immigration and customs, tell me how well they handle your
encrypted laptop when they take it away from you and send it to the boys
at Langley for analysis.
> or security,
Security. The most over-worked and over-used word in the english
language since 9/11.
If I want security, I'll buy a vault and put my PC in it. I don't need
the "security" that NTFS offers on my desktop system at home or at
work. That same security that malware seems so able to get around.
> then stick with FAT32. Just don't recommend it as the best
> filesystem for all needs.
For student, home and SOHO situations, it is a better choice. For big
evil corporations and organizations that don't or can't trust their
employees to not use their CD-drive trays as coffee-cup holders, then
NTFS is for those goons.
> > And yes, it would matter more to me how badly a file system
> > could break vs how often it does break, because eventually
> > you will encounter the "bad" break
>
> That's what backups are for.
Like I said above, backups are another layer that can be used with any
file system and play no role in the discussion of the pro's and con's of
one file system over another.
But it's still sad that you have to raise the topic of backups as a way
to explain-away the deficiencies and vulnerabilities of an NTFS drive
that goes south.
> > It does exactly what it's supposed to do - which is to store
> > files and not play games with me, the system user and owner.
> > User permissions, rights, etc, have no place on a consumer
> > desktop or laptop PC.
>
> Unless you share your machine with other members of the family.
Hmmm.
I never considered that you might live in a third-world country, with
only one computer per household. Good point.
DOS Guy
> > Why don't you give us your comments on this material:
>
> He won't, but I will: I get the impression from that blog post
> that he has spent a ton of time recovering files because he had
> a poor backup strategy, and he is trying to justify his poor
> backup strategy by promoting the filesystem that recovers easily
> from crashes.
I'll take a file system that recovers easily vs one that requires
inordinate backup fuss any day.
Klaus Meinhard
> I think there is a practical if not legal difference if I were to
> re-package some old copy of DOS or Windoze and sell it to someone else
> vs installing as many copies of those products as I want on my own
> computers for my own use.
You probably have a license for only one or two machines. Going against
that is illegal in most countries, regardless of how you try to justify
it.
> ... Micro$haft...
> ... Milkro$oft...
I have no further interest to continue in this prepubertal microsoft
bashing.
Microsoft may deem it unnecessary to prosecute license breaks for this
old and nearly dead OS, because there's no longer any money in it. It
would have been the nice thing to make it free and open source software,
just as Rex did with 4DOS.
That is, however, no reason to dance before their doorstep and let your
thing hang out just to show how plucky you are.
Try to write some software and earn money with it. Then lets see how you
think about illegal copying.
--
Herzliche Gr��e,
Klaus Meinhard
DOS Guy
> You probably have a license for only one or two machines. Going
> against that is illegal in most countries, regardless of how you
> try to justify it.
Illegal is the wrong word.
When two parties are in conflict with each other over a licensing issue,
the concept of "illegality" does not come into play. Only the
determination if a license violation has occurred, and any court-awarded
dammages that might result. Nobody goes to jail over it.
> Microsoft may deem it unnecessary to prosecute license breaks for
> this old and nearly dead OS, because there's no longer any money
> in it.
I don't believe that's the reason. I believe the reason is that they
don't want to experience a historical precedent of failure to prosecute
an individual for using one of their products withing having a legit
license.
> That is, however, no reason to dance before their doorstep and
> let your thing hang out just to show how plucky you are.
Microsoft (the corporation that has been found guilty of many truly
illegal acts in many countries) needs to understand that even though it
has been granted defacto monopoly status in contrary to existing laws,
that it can't escape competition with itself vis-a-vie the continued (if
not "questionable") use of it's older products.
Making it known that I am depriving Microsoft of revenue by telling it
that I am using it's older defunct products is a pleasure. If the state
will not enforce it's anti-monopoly laws and enable a healthy and
competitive software industry, then I will do my part to deprive
Microsoft any future revenue by using it's older products at will,
against the terms of it's EULA. It is a sad commentary that Windows 98,
for example, is still a viable and practical alternative to XP and newer
OS's. That is to be expected in the monopoly paradigm we have to live
with.
> Try to write some software and earn money with it. Then lets
> see how you think about illegal copying.
We do write software for which we charge several thousand dollars for.
We have a simple license-key arrangement to enable the software's
functionality (no dongle).
Klaus Meinhard
> When two parties are in conflict with each other over a licensing
> issue, the concept of "illegality" does not come into play. Only the
> determination if a license violation has occurred, and any
> court-awarded dammages that might result. Nobody goes to jail over
> it.
... as long as you don't do it on a grand scale or commercially, as long
as you live in a western democracy and can afford good lawyers, and as
long as your countries jails are more than full anyway.
>> That is, however, no reason to dance before their doorstep and
>> let your thing hang out just to show how plucky you are.
> Making it known that I am depriving Microsoft of revenue by telling it
> that I am using it's older defunct products is a pleasure.
That's exactly what I meant above.
>> Try to write some software and earn money with it. Then lets
>> see how you think about illegal copying.
> We do write software for which we charge several thousand dollars for.
> We have a simple license-key arrangement to enable the software's
> functionality (no dongle).
So your software is probably so specialized that the hacker kids didn't
come across it, or so oldfangled (still writing DOS progs?) that they
took pity :-)
DOS Guy
> > When two parties are in conflict with each other over a
> > licensing issue, (...) Nobody goes to jail over it.
>
> ... as long as you don't do it on a grand scale or
> commercially,
That's more-or-less the determining factor. If the police or law
enforcement agency can lay a charge and prosecute a case without needing
assistance or input from the copyright holder. This typically happens
when a police raid stumbles upon boxes of music or movie CD's.
I don't think that any retail PC vendor loading their PC's with hacked
or duplicate XP license keys has ever been taken to court or lost a
court case, and even if they did, I don't think they'd ever get a jail
sentence. Out-of-court settlements - probably.
> as long as you live in a western democracy and can afford
> good lawyers, and as long as your countries jails are more
> than full anyway.
Nobody goes to jail over a civil-court matter - unless perhaps if you're
found in contempt of court (bad behavior in court, failure to pay a
court-ordered judgement, etc).
> > > That is, however, no reason to dance before their doorstep
> > I am using it's older defunct products - is a pleasure.
> That's exactly what I meant above.
I think it's a good reason. A way to give Macro$haft the middle finger.
> > We do write software
>
> So your software is probably so specialized that the hacker
> kids didn't come across it
Yes
> or so oldfangled (still writing DOS progs?) that they
> took pity :-)
It's mostly a suite of Windoze apps (visual studio 6) but it does have
one program that is a compiled PB program (ie 16-bit DOS) used for data
acquisition (it runs on XP with the help of a port-mapper because it
performs direct hardware I/O access).
Jim Leonard
> I'll take a file system that recovers easily vs one that requires
> inordinate backup fuss any day.
And I'll take one that doesn't need recovery in the first place. So
this is where we disagree.
Jim Leonard
> > His personal observations don't factor into the conversation.
>
> His experiences with NTFS are no different than anyone else's. NTFS
> doesn't behave differently depending on who is using it.
They are different than mine, which invalidates them if that is your
only reference sample.
> It's just that he bothers to investigate the working details and pull
> the cover off the charade that is NTFS.
There's no charade to booting without any recovery needed from a power
loss, vs. *always* needing recovery from a power loss with FAT32.
> > If your needs never grow beyond 4G file limits,
>
> Many apps limit their user files to 4gb, even under XP, vista, etc.
> Mailbox size for example.
Well, I hope you never need to work with video.
> > compression,
>
> Who uses compression these days (or - this decade?)
Just because you don't see a need for a feature doesn't mean that
feature isn't useful.
DOS Guy
> > His experiences with NTFS are no different than anyone else's.
> > NTFS doesn't behave differently depending on who is using it.
>
> They are different than mine, which invalidates them if that is
> your only reference sample.
What kind of cracked logic is that?
Cquirke's examples are "invalidated" simply because they are different
than yours?
How about this: The car you drive is different than the car I drive.
Therefore I conclude that you don't drive a car. How's that logic?
It's the same as yours.
> > It's just that he bothers to investigate the working details
> > and pull the cover off the charade that is NTFS.
>
> There's no charade to booting without any recovery needed fro
> m a power loss, vs. *always* needing recovery from a power
> loss with FAT32.
When was the last time, or how often do you run win-9x or XP from a
FAT32 volume?
I do it all the time, on about a dozen machines, for about, oh, the past
10 years. So I can tell you that the percentage of time that a FAT32
volume auto-runs scandisk at startup is about the same as an NT-based
system with NTFS does (which is very rarely).
Hey, I know that I have to do a hard or soft reset on my win-98 systems
when I bog them down with too many open apps. It's the rare time (less
than 5%) that scandisk auto-runs as a consequence.
And at least with FAT32 I can cancel scandisk if I want (with zero
consequences) but you're forced to let scandisk (or chkdisk I think it's
called in NT) run to completion.
> > > If your needs never grow beyond 4G file limits,
>
> Well, I hope you never need to work with video.
I built an XP-pro system for a friend a couple years ago, using a pair
of 250 gb SATA drives. I formatted the drives with FAT32, using 4kb and
32kb cluster size (4kb for drive 1, and 32kb for drive 2). The OS and
apps are installed on drive 1. Drive 2 is used for large multimedia
files.
I torrented a bunch of Adobe Premiere CS3 video editing software and
installed it on the system. It had no problem with the FAT32 volumes
reading and writing and editing the various video files from a digital
video camera. I guess it knows to span the 4 gb file boundary if it
needs to.
The only place I see really large files (more than 4 gb) are for virtual
images. And not too many people run virtual sessions. And it might
even be the case that VM-ware and other virtual machine managers would
also span the 4gb file boundary if they needed to.
> > > compression,
> >
> > Who uses compression these days (or - this decade?)
>
> Just because you don't see a need for a feature doesn't mean
> that feature isn't useful.
Do you use the NTFS file-compression feature? Do you know anyone that
does?
Klaus Meinhard
may I remind you that this thread is for 4DOS and its successors 4NT,
TCC and TCC/LE, powerfull command line interpreters by JP Software, not
general discussions about DOS topics.
If you want to continue this discussion, I'd recommand alt.msdos, where
you would find a greater audience which might take an interest.
Ragnarok
mcalhoun escribi�:
> From Newsgroup: comp.os.msdos.programmer
>
> I have a real DOS machine but would like to "combine" it with my Windoze
> XP Home machine. I've installed Sun's VirtualBox and it will boot
> command.com from a floppy disk, but it cannot "see" any hard drive.
>
> Has anyone gotten any farther with VirtualBox?
>
lastest virtualbox put the virtual disk on SATA bus, that (i think ) dos
cant see it
try to change it to IDE
--- Synchronet 3.15 Hack by Ragnarok rev. a-Linux NewsLink 1.91
* Dock Sud BBS - Avellaneda, Bs As, Argentina - telnet://bbs.docksud.com.ar
Ragnarok
DOS Guy
> lastest virtualbox put the virtual disk on SATA bus, that (i think)
> dos cant see it try to change it to IDE
What do you mean - SATA "bus" ?
Do you mean SATA drive?
> lastest virtualbox put the virtual disk on SATA bus
Didn't it as you which drive to put it on?
> SATA bus that (i think) dos cant see it
DOS can see SATA drives just fine - as long as they contain FAT-32
volumes.
Peter Schepers
>
>DOS can see SATA drives just fine - as long as they contain FAT-32
>volumes.
_Only_ if the BIOS has the SATA drive controller set to Legacy or
Compatible, or some such. Set to AHCI and DOS will not see it. Some
Toshiba laptops can't turn SATA to Legacy, and likely newer machines also
cannot switch the SATA modes, thus further limiting older OS's.
PS
Ross Ridge
>_Only_ if the BIOS has the SATA drive controller set to Legacy or
>Compatible, or some such. Set to AHCI and DOS will not see it. Some
>Toshiba laptops can't turn SATA to Legacy, and likely newer machines also
>cannot switch the SATA modes, thus further limiting older OS's.
Actually, MS-DOS doesn't support SATA, IDE, ST-506 or any other kind of
hard drive interface directly. It uses the BIOS to access hard drives.
hard disks. Since boot sectors also use the BIOS to access the hard
drive, if you can boot any operating system at all on a SATA drive you
should be also be able to access it from MS-DOS. You will however be
limitted to whatever size restrictions imposed by the version of MS-DOS
you're using.
Ross Ridge
--
l/ // Ross Ridge -- The Great HTMU
[oo][oo] rri...@csclub.uwaterloo.ca
-()-/()/ http://www.csclub.uwaterloo.ca/~rridge/
db //
H-Man
To be clear MS-DOS < v6.22 cannot read the FAT32 FS. I know you know this,
but it should be made clear. And DOS cannot read just any FAT drive. Set
your BIOS to AHCI mode for the SATA drives and try to get DOS to read the
drive. An AHCI drive can still be formatted in FAT or FAT32. It gets even
more difficult with a RAID array, and yes, these can still be FAT32. DOS
uses the BIOS to access the drive interface, so the drive interface either
needs to be something that DOS knows about, or you need a driver/TSR to
intercept the BIOS calls and translate them. Quite doable with DOS but not
common. Newer interfaces, like AHCI for any chipset I've ever seen, do not
have a DOS driver/TSR available for them. I doubt we'll ever see one. Most
BIOSs allow you to set the SATA interface mode to IDE, which is something
DOS can read and write to directly via the BIOS. With the right tools DOS
can see anything the BIOS can, but you can't necessarily do anything
constructive with it, or use it in any conventional sense.
--
HK
H-Man
> Peter Schepers <sche...@ist.uwaterloo.ca> wrote:
>>_Only_ if the BIOS has the SATA drive controller set to Legacy or
>>Compatible, or some such. Set to AHCI and DOS will not see it. Some
>>Toshiba laptops can't turn SATA to Legacy, and likely newer machines also
>>cannot switch the SATA modes, thus further limiting older OS's.
>
> Actually, MS-DOS doesn't support SATA, IDE, ST-506 or any other kind of
> hard drive interface directly. It uses the BIOS to access hard drives.
> hard disks. Since boot sectors also use the BIOS to access the hard
> drive, if you can boot any operating system at all on a SATA drive you
> should be also be able to access it from MS-DOS. You will however be
> limitted to whatever size restrictions imposed by the version of MS-DOS
> you're using.
>
The boot sectors don't use anything, the BIOS accesses the boot sectors for
instructions on what to do next. If I can boot Windows 7 from a drive set
to AHCI mode in the BIOS, and formatted in NTFS, I'd be really interested
in how you would have DOS do anything even remotely constructive with the
drive or it's data. You can't even partition the drive and format it. I
don't think DOS will even see it as installed hardware. You would need a
third party driver or TSR to translate the BIOS calls to something the AHCI
drive will understand, and you'd need a third party NTFS driver/TSR. I do
not believe these even exist in tandem.
--
HK
Peter Schepers
Ross Ridge <rri...@csclub.uwaterloo.ca> wrote:
>Peter Schepers <sche...@ist.uwaterloo.ca> wrote:
>>_Only_ if the BIOS has the SATA drive controller set to Legacy or
>>Compatible, or some such. Set to AHCI and DOS will not see it. Some
>>Toshiba laptops can't turn SATA to Legacy, and likely newer machines also
>>cannot switch the SATA modes, thus further limiting older OS's.
>
>Actually, MS-DOS doesn't support SATA, IDE, ST-506 or any other kind of
>hard drive interface directly. It uses the BIOS to access hard drives.
>hard disks. Since boot sectors also use the BIOS to access the hard
>drive, if you can boot any operating system at all on a SATA drive you
>should be also be able to access it from MS-DOS. You will however be
>limitted to whatever size restrictions imposed by the version of MS-DOS
>you're using.
I trust you have read some of the responses other have posted. It
certainly appears from the wording in your posting that this is something
of which you have no practical experience on. The issue of accessing hard
disks from DOS on AHCI-enabled BIOS's is something I deal with on a daily
basis. And no, it doesn't work when AHCI (or a non-legacy/compatible
setting) is enabled. There are drivers to read AHCI-enabled CD drives but
not the hard disks from DOS. Even Windows XP can't access the hard disk
without SATA drivers in the boot sequence.
PS.
E. S. Fabian
| With the right tools DOS can see anything the BIOS can, but you
| can't necessarily do anything constructive with it, or use it in any
| conventional sense.
Technically speaking, there is no reason one could not write an interface
driver so "DOS" could have full access to an NTFS drive, or any other file
system, as long as the BIOS can provide the physical access, or one is
willing to expand the BIOS to do it. Neither data structures on disk, nor
physical access are magic. The real issue is that nobody wants to devote the
labor to do it, when alternates are abundant. DOS was designed for small,
slow computers (in today's technology). I spent 600.00 USD on my first hard
drive, with a whopping 10 MB! And I also had to write my own BIOS to access
it from CP/M. I would not want to do the same job over again.
However, this whole discussion is OT - it has no relationship to 4DOS.
--
Steve
DOS Guy
> The issue of accessing hard disks from DOS on AHCI-enabled BIOS's
> is something I deal with on a daily basis. And no, it doesn't work
> when AHCI (or a non-legacy/compatible setting) is enabled.
I don't see why not.
Given a motherboard with on-board SATA ports, are you saying that the
BIOS int13 and int13h routines will not provide access to a SATA drive
unless the SATA sub-system has been placed into legacy /
IDE-compatability mode?
My own experience with installing win-98 on an ASROCK DUAL-VSTA
motherboard and 500 gb Western Digital SATA hard drive is that:
a) it's necessary to _NOT_ enable IDE/Legacy mode because I do not want
Win-98's protected mode hard-drive driver (ESDI_506.pdr) to touch the
drive because of it's 128 gb access limitation. By keeping the
controller in native SATA mode, win-98 installs itself but must resort
to "DOS-mode" compatability drive access until I have access to the
desktop and can install the 32-bit SATA drivers.
NT-based OS's are crippled in this regard, as they don't seem to be able
to use DOS-mode access (INT-13h) as an interim solution until the user
is able to feed the drivers. Instead, the SATA drivers must be fed to
the OS early during the installation process.
b) Given the above hardware and SATA bios setting, I can boot the system
into DOS and still have full access to the drive in question, including
scandisk and chkdsk (it's a fallacy that DOS scandisk has a 128 gb size
limitation).
I don't see why enabling AHCI would have any effect on the INT13
routines and their ability to provide an interface to the SATA
subsystem.
Ross Ridge
> Actually, MS-DOS doesn't support SATA, IDE, ST-506 or any other kind of
> hard drive interface directly. It uses the BIOS to access hard drives.
> hard disks. Since boot sectors also use the BIOS to access the hard
> drive, if you can boot any operating system at all on a SATA drive you
> should be also be able to access it from MS-DOS. You will however be
> limitted to whatever size restrictions imposed by the version of MS-DOS
> you're using.
>The boot sectors don't use anything, the BIOS accesses the boot sectors for
>instructions on what to do next.
The BIOS loads the boot sector on the hard drive in to memory and
then the boot sector uses the BIOS to access the hard drive in order
to continue the bootstrap proces. There just isn't enough space in a
512 byte boot sector for it to be accessing the ACHI device directly,
never mind an IDE device as well.
Ross Ridge
>The issue of accessing hard disks from DOS on AHCI-enabled BIOS's is
>something I deal with on a daily basis. And no, it doesn't work when AHCI
>(or a non-legacy/compatible setting) is enabled.
I have hard time believing you deal with something on daily basis that
you think is impossible.
>There are drivers to read AHCI-enabled CD drives but
>not the hard disks from DOS.
There are also drivers for reading IDE CD-ROM drives on MS-DOS, but not
for reading IDE hard drives. MS-DOS doesn't need a driver to access
hard drives, it uses the BIOS.
>Even Windows XP can't access the hard disk without SATA drivers in the
>boot sequence.
And how do you think the Windows XP boot sequence loads the SATA drivers
off the SATA drive?
Klaus Meinhard
> H-Man wrote:
>> With the right tools DOS can see anything the BIOS can, but you
>> can't necessarily do anything constructive with it, or use it in any
>> conventional sense.
>
> Technically speaking, there is no reason one could not write an
> interface driver so "DOS" could have full access to an NTFS drive, or
> any other file system, as long as the BIOS can provide the physical
> access, or one is willing to expand the BIOS to do it.
Looking at it realistically, DOS is dead.
A modern computer doesn't even have floppy drives. That means you cannot
install any old DOS from the original media (FreeDOS may have an CD iso
file, I haven't looked lately). So you need at least 1 modern system to
build an old DOS system.
Interaction with hardware today is mostly SATA and USB, filesystem
mostly NTFS. You have to find and install drivers for these (and not all
are very reliable), in addition to the other DOS drivers for screen,
keyboard, mouse(?), CD drive, compatibilty, network and internet access,
which may leave you without enough memory for any serious work even
using a third party mem manager. Of course you have the alternative to
dumb down a modern, fast machine to run DOS on it, but how intelligent
is that?
You then have to find printer drivers for all your printing apps,
preferably for a very old needle printer from a flee market. How long
will it last?
You then have to find apps to do anything still worthwhile, which isn't
easy and in many cases at least dubious if not illegal, even if nobody
sues you. I know I couldn't fulfill my needs starting from internet
access, media center, office apps, games, synching with my mobile etc.
with an old DOS machine and an 80x25 character screen.
Taken together this means that you put so much of your time into
building and maintaining such a system that you can justify it today
only as a hobby for diehards, which is okay. But it isn't even a viable
OS for a family computer today, much less a viable business environment.
If the defenders of the virtue of DOS would take the trouble to look at
XP, Vista or Wiindows 7, they would learn that they can have their
beloved DOS, even with several instances on the same screen, several
different versions at the same time,
Of course all of this is off topic here, but look at the time the last
on_topic message was posted here. Even the name of this group now
precludes users of TCC/LE, the free NT version of 4DOS, or its
commercial sisters 4NT, Take Command, TCC etc. to find it as a forum for
their questions.
So really, should we care (taking into consideration the time to
admonish offenders, and the inabilty to do something if not heeded)?
Jim Leonard
> b) Given the above hardware and SATA bios setting, I can boot the system
> into DOS and still have full access to the drive in question, including
> scandisk and chkdsk (it's a fallacy that DOS scandisk has a 128 gb size
> limitation).
Not all DOS is equal. When most people think of DOS they think of MS-
DOS 6.22, which does have that limitation.
When I talk about something in DOS to hobbyist friends, I am very
careful to qualify what version I'm using (PC DOS 2000) and other
friends are careful to do the same (ie. FreeDOS, or DRDOS 7.03, etc.).
You need to stop saying that "DOS" has all these great features when
you are specifically talking about the version of MS-DOS that ships
with Windows 98.
H-Man
> Ross Ridge wrote:
>> Actually, MS-DOS doesn't support SATA, IDE, ST-506 or any other kind of
>> hard drive interface directly. It uses the BIOS to access hard drives.
>> hard disks. Since boot sectors also use the BIOS to access the hard
>> drive, if you can boot any operating system at all on a SATA drive you
>> should be also be able to access it from MS-DOS. You will however be
>> limitted to whatever size restrictions imposed by the version of MS-DOS
>> you're using.
>
> H-Man <Sp...@bites.fs> wrote:
>>The boot sectors don't use anything, the BIOS accesses the boot sectors for
>>instructions on what to do next.
>
> The BIOS loads the boot sector on the hard drive in to memory and
> then the boot sector uses the BIOS to access the hard drive in order
> to continue the bootstrap proces. There just isn't enough space in a
> 512 byte boot sector for it to be accessing the ACHI device directly,
> never mind an IDE device as well.
>
Yes, I agree, but the boot sectors themselves do not access anything, the
sectors are accessed. The BIOS loads the code in the MBR or VBR and
effectively hands over control to that code. I get what you are saying, but
the terminology is not entirely correct. A sector is a container, it can't
access anything, it's contained code does the accessing and it does that
from RAM. The BIOS is required it provides some read access to the drive
interface but not nearly the access required for a full OS, again, this
would require the OS understands the interface, MS-DOS understands only the
level of access the BIOS gives, and the BIOS does not fully understand AHCI
(in this instance, RAID arrays would also apply here with some exceptions)
drive access. This is why you would require you set the BIOS to provide the
IDE translation for the drive. Having done this you forgo the benefit of
the AHCI (native) interface and are limited to IDE access only. As AHCI is
handled differently depending on the chipset / controller used, it makes
sense that you would require third party code to properly access the drive
as an OS.
If I'm wrong here I'm open to correction.
Where the BIOS does not provide full R/W access to the drive, DOS will not
be able to use this drive. As I can boot an AHCI drive in AHCI mode in XP
and I cannot in DOS, I'd suggest that saying "if you can boot any operating
system at all on a SATA drive you should be also be able to access it from
MS-DOS" would be misleading at best and incorrect for certain.
--
HK
H-Man
Steve,
You are correct, the BIOS would be required to provide full support and the
file system would need to be supported. There is no reason it could not be
done. One could even write a driver to hook the BIOS calls and provide
access from there as an expanded BIOS set. I believe Ontrack had a disk
manager software once that would allow int13h access to larger drives
(>504MB) where the BIOS did not support it.
Yes this has gotten OT and I appologize, so I am not crossposting this.
--
HK
H-Man
Ross,
You have my sincerest apologies, I am entirely incorrect regarding AHCI and
DOS access. I saw DosGuy's post and thought I would have to try this for
myself. I freed up a disk on a new machine (Intel server MB) and booted DOS
from a USB stick. Although DOS would only see 8GB of drive there (it is a
1TB drive), I could partition it, format it and read and write to it. So, I
completely admit, I was mistaken and was wrong. Although I have not tried
it, I wonder if one set up a RAID array if the RAID BIOS would allow the
same transparancy?
Again, I apologize.
--
HK
H-Man
As per one of my previous posts. I am incorrect regarding my assertion BIOS
will not provide complete access to AHCI drives for DOS. This was an
incorrect assertion on my part and I apologize to the group for this. My
bad.
--
HK
H-Man
> Peter Schepers wrote:
>
>> The issue of accessing hard disks from DOS on AHCI-enabled BIOS's
>> is something I deal with on a daily basis. And no, it doesn't work
>> when AHCI (or a non-legacy/compatible setting) is enabled.
>
> I don't see why not.
>
> Given a motherboard with on-board SATA ports, are you saying that the
> BIOS int13 and int13h routines will not provide access to a SATA drive
> unless the SATA sub-system has been placed into legacy /
> IDE-compatability mode?
It appears you are correct in this regard and it is my recent experience
that the BIOS does in fact provide int13 and int13h access to the drive.
>
> My own experience with installing win-98 on an ASROCK DUAL-VSTA
> motherboard and 500 gb Western Digital SATA hard drive is that:
>
> a) it's necessary to _NOT_ enable IDE/Legacy mode because I do not want
> Win-98's protected mode hard-drive driver (ESDI_506.pdr) to touch the
> drive because of it's 128 gb access limitation. By keeping the
> controller in native SATA mode, win-98 installs itself but must resort
> to "DOS-mode" compatability drive access until I have access to the
> desktop and can install the 32-bit SATA drivers.
>
> NT-based OS's are crippled in this regard, as they don't seem to be able
> to use DOS-mode access (INT-13h) as an interim solution until the user
> is able to feed the drivers. Instead, the SATA drivers must be fed to
> the OS early during the installation process.
This also would appear to be the case. Stupid Microsoft!! How much simpler
would it make things to be able to step up after the OS is installed as you
do with '98. I can't tell you how difficult it was to get Win2K SAerver
loaded onto new hardware simply because the Intel drivers would not work.
They were written for XP, everything said they would work with 2K, but
noooooooo. Stupid Microsoft. I ended up using older drivers that did work.
Even the drivers that came with the MB didn't work.
>
> b) Given the above hardware and SATA bios setting, I can boot the system
> into DOS and still have full access to the drive in question, including
> scandisk and chkdsk (it's a fallacy that DOS scandisk has a 128 gb size
> limitation).
>
> I don't see why enabling AHCI would have any effect on the INT13
> routines and their ability to provide an interface to the SATA
> subsystem.
It appears it doesn't.
--
HK
Ross Ridge
> b) Given the above hardware and SATA bios setting, I can boot the system
> into DOS and still have full access to the drive in question, including
> scandisk and chkdsk (it's a fallacy that DOS scandisk has a 128 gb size
> limitation).
Jim Leonard <moby...@gmail.com> wrote:
>Not all DOS is equal. When most people think of DOS they think of MS-
>DOS 6.22, which does have that limitation.
Actually, MS-DOS 6.22 uses the old CHS BIOS services, and so has a 8Gb
size limitation. MS-DOS 7 (in Windows 95/98/ME) can use the newer
64-bit LBA BIOS services and so can access hard drives of any size,
or at least any size the BIOS supports. Hmm... the 2.2Tb limit of the
MBR partition table is an effective limit on what MS-DOS 7 can support
without additional software, though I suppose you could format a drive
without a partition table to get past it.
Rugxulo
Apologies in advance for cross-posting, but I doubt you'd see it if I
don't also cross-post, so ... ;-)
On Nov 13, 4:33 am, "Klaus Meinhard" <K_Meinh...@gmx.de> wrote:
>
> Looking at it realistically, DOS is dead.
It isn't that dead or else MS, DR, etc. would give away their products
(and not whine when someone uploads 'em online). And they haven't
budged. So it must have some value to them. No, they don't update
anymore, but others do (e.g. FreeDOS, ROM-DOS). MS said DOS was dead
with the arrival of XP. (They also say Win7 was "made by me", heh. O
RLY?)
And realistically, something is useless when it's dead, but DOS
continues to work fine on real hardware or even under emulation or
virtualization. DOSBox doesn't really count as true DOS, but NTVDM,
DOSEMU, OS/2, VirtualBox + FreeDOS, etc. do.
I don't see DOS dying any worse than Atari or Amiga. They might not
have the largest fan base, but since when did that matter? Who cares?
We're doing this for fun, not to compete with other people.
> A modern computer doesn't even have floppy drives.
That's the OEM's fault, not that of the OS or users. You can buy a USB
floppy drive (3.5").
> That means you cannot
> install any old DOS from the original media (FreeDOS may have an CD iso
> file, I haven't looked lately). So you need at least 1 modern system to
> build an old DOS system.
I have some old MS-DOS 5 floppies that need 5.25", and that indeed
might be hard to find (although my old machine has one). However,
there's no reason to bother with that when I have FreeDOS (which
indeed comes on .ISO that can be burned to CD-R). Actually, you can
automatically make any floppy image into a CD image, and making a CD
image from scratch is pretty easy too. Some have even gotten CD
burning to work in DOS (which I've never tried personally, though).
Even if you did need another system, that doesn't prevent you, as most
people have multiple machines.
> Interaction with hardware today is mostly SATA and USB, filesystem
> mostly NTFS. You have to find and install drivers for these (and not all
> are very reliable),
The first Linux kernel wasn't so perfect either. I don't understand
the whole "DOS is dead" mentality. Why is it Linux (which is updated
every day) is considered eternally better than FreeDOS? Both can be
extended. Both can be improved. It just needs someone to do it.
Neither is alive or dead, just one has more volunteers (and commercial
sponsors, big whoop).
> in addition to the other DOS drivers for screen,
> keyboard, mouse(?), CD drive, compatibilty, network and internet access,
> which may leave you without enough memory for any serious work even
> using a third party mem manager.
You mean conventional memory? (Anyways, most of that other stuff is
taken care of under DOSEMU, which could be considered one way of
"upgrading" your setup.) You must because DOS can access fairly large
amounts (ask Jack Ellis), e.g. 2 GB.
> Of course you have the alternative to
> dumb down a modern, fast machine to run DOS on it, but how intelligent
> is that?
It's only as dumb or intelligent as the user himself. "A poor
carpenter blames his tools", remember? It depends on what you want to
do.
> You then have to find printer drivers for all your printing apps,
> preferably for a very old needle printer from a flee market. How long
> will it last?
You can't even find printer drivers on modern Windows!! You think
everything works there? It doesn't. (Ahem, digital camera, XP = good,
Vista = bad.) Vista (and presumably 7) don't even support the same
drivers as XP, which is annoying. For some extremely confusing reason,
Microsoft either doesn't seem to care or cares very little. It boggles
the mind. They already (officially!) want XP to be obsolete and are
heavily pushing that fact. And, as you probably know, XP worked well
(better, even, in some ways, than Vista, ahem NTVDM).
> You then have to find apps to do anything still worthwhile, which isn't
> easy and in many cases at least dubious if not illegal,
You can write or port your own apps. It depends on your skills, needs,
interests. I regularly compile stuff myself with DJGPP. OpenWatcom's
not too shabby either. There is still enough DOS software to be
interesting (to me, at least).
> even if nobody
> sues you. I know I couldn't fulfill my needs starting from internet
> access, media center, office apps, games, synching with my mobile etc.
> with an old DOS machine and an 80x25 character screen.
Internet access is done all the time in DOS (hi Udo, DOS386 !), but
I'll admit it's not exactly transparent. Media is also done in DOS (hi
DOS386 !). Games? If anything, that's the stereotype that DOS was/is
only good for "old" games. Syncing with your mobile? Blame the
carrier, not us. 80x25? I guess you know you can easily use up to
132x32 in DOS, you're not limited (except by your video card).
> Taken together this means that you put so much of your time into
> building and maintaining such a system that you can justify it today
> only as a hobby for diehards, which is okay. But it isn't even a viable
> OS for a family computer today, much less a viable business environment.
It's only viable if you have software. An OS is nothing without
software. And an OS that doesn't run because the requirements are too
high (or price is too ridiculous) is also useless. Also useless is
when they try to force you to learn yet another entirely different
environment and re-port (or repurchase) all your software. If you
enjoy that, then good for you, but most don't.
Obviously nobody said, "DOS should take over the world!" It's not
quite the same attitude that is prevalent in Linux and Windows
circles. Yes, it's only hobbyists because we're the only ones who
care! It has nothing to do with marketshare or money, only fun and
usefulness. :-)
> If the defenders of the virtue of DOS would take the trouble to look at
> XP, Vista or Wiindows 7, they would learn that they can have their
> beloved DOS, even with several instances on the same screen, several
> different versions at the same time,
No. Modern Windows only emulates DOS, it's not real DOS. It's not as
good (ahem, weak or non-existant support for VESA, SB). Vista on up
don't even work as well as XP (DPMI limit -> needs registry hack, no
full-screen, various other bugs), which is ultra annoying! And I think
you mean "different instances" as you're only able to use the provided
MS-DOS 5 variant (unlike OS/2 or eCS where you can use any DOS, or so
I'm told).
Yes, DOSEMU should be more popular as that solves many problems. The
"problem" with that is that it's in "multiverse" (for some unknown
reason), and no Linux distro includes it by default (although a very
few include DOSBox, which is mostly good but not real DOS and very
very slow).
> Of course all of this is off topic here, but look at the time the last
> on_topic message was posted here. Even the name of this group now
> precludes users of TCC/LE, the free NT version of 4DOS, or its
> commercial sisters 4NT, Take Command, TCC etc. to find it as a forum for
> their questions.
>
> So really, should we care (taking into consideration the time to
> admonish offenders, and the inabilty to do something if not heeded)?
I'm not sure why it was cross-posted to the 4DOS group, honestly, but
as you mention, traffic is low (except for spam, ugh), so it shouldn't
matter.
Ross Ridge
>Although I have not tried it, I wonder if one set up a RAID array if
>the RAID BIOS would allow the same transparancy?
Under the same "if you can boot from it, MS-DOS should support it"
principle, yes. In the past I've had MS-DOS access a non-RAID configured
IDE drive through one of those cheap motherboard RAID controllers.
I'd be suprised if it wouldn't have worked in a RAID configuration,
but in that case I just wanted to connect a fifth IDE device.
Note, one potential pitfall is that in some cases the BIOS will only
enable support for the drive if you boot from it, like when you booted
DOS from your USB stick.
Klaus Meinhard
> Apologies in advance for cross-posting, but I doubt you'd see it if I
> don't also cross-post, so ... ;-)
I hope I've stopped that with this post.
> It isn't that dead or else MS, DR, etc. would give away their products
> (and not whine when someone uploads 'em online). And they haven't
> budged.
My guess would be they did it first not to hurt the sales of their next
OS, and because some of it was still buried in their NT type OS up until
today. and second on principle.
> And realistically, something is useless when it's dead, but DOS
> continues to work fine on real hardware or even under emulation or
> virtualization. DOSBox doesn't really count as true DOS, but NTVDM,
> DOSEMU, OS/2, VirtualBox + FreeDOS, etc. do.
As Ive said and your post so clearly illucidates:
Taken together this means that you put so much of your time into
building and maintaining such a system that you can justify it today
only as a hobby for diehards, which is okay. But it isn't even a viable
OS for a family computer today, much less a viable business environment.
So maybe DOS is not dead, but Greepeace would surely put it on a red
list of dying species :-)