Scandisk and external 320GB USB HD

[MEB]

A CORRECTION the segment which reads:

A single partition on a 320 [or the 127 per the other discussion] gig drive
would potentially have how many using that same test?


SHOULD READ for clarification:

A single partition on a 320 [or 127 per the other discussion] gig drive
would potentially have how many files and folders using that same test to
fill the drive?

--
MEB
http://peoplescounsel.orgfree.com
________

 

[MEB]

"MEB" <meb@not here@hotmail.com> wrote in message
news:uFS78WuyHHA.3916@TK2MSFTNGP02.phx.gbl...
| A CORRECTION the segment which reads:
|
| A single partition on a 320 [or the 127 per the other discussion] gig
drive
| would potentially have how many using that same test?
|
|
| SHOULD READ for clarification:
|
| A single partition on a 320 [or 127 per the other discussion] gig drive
| would potentially have how many files and folders using that same test to
| fill the drive?
|
| --
| MEB
| ________
|
|

ADDENDUM: so these issues need not be separately addressed in some other
discussion or later here:

It would also be poignant to your [Franc] usage, and the testing and or
purported results, to have:

A. the hardware/chipset and other information which purports to allow such
usage. related to the intended use and/or to support the findings

B. IN ADDITION to information which may also have effect such as:

1. Any.system file replacements, examples: shell32, kernel32, user, browser
dlls, etc.. and whether IE is installed and what version, and update
history to the testing system(s).

2. Any registry tweaks or other like activities such as: shell replacements,
explorer replacements, etc. which also may have an effect.

3. What was used in the fdisk and formatting process, and cluster size, etc.
pursuant those activities.

4. Tools/applications for routine disk maintenance of these large
disks/partitions.


BTW Franc, I have no intention of conversing with 98 Guy on any more related
issues or in this discussion, If you or others wish to do so then .....

--
MEB
http://peoplescounsel.orgfree.com
________

 

[Franc Zabkar]

On Thu, 19 Jul 2007 17:49:11 -0400, 98 Guy <98@Guy.com> put finger to
keyboard and composed:

>What I have found is this:

>- The claimed max number of files per directory is 65535.

>http://www.microsoft.com/technet/technetmag/issues/2006/07/WindowsConfidential/


The limit appears to be 65534, but who's counting:
http://guysalias.batcave.net/images/65534-entries.jpg

In alt.msdos.batch, [a different?] Guy writes:

=================================================================
I think...

There is a limit to the number of entries in a given directory,
whether root or a subdirectory. That limit is 65,534 entries or less
if LFNs.

Let's see...

FOR /L %%I IN (1,1,66000) DO ECHO.>%%I.TXT

DIR G:\

[...]

65533 File(s) 131,106 bytes
0 Dir(s) 4,966,756,352 bytes free

65533 Files + RECYCLER = 65,534 entries
=================================================================

He is using an NT class OS, not Win9x, but I can't see how this would
alter the result. His test uses 8.3 filenames which would require the
least space.

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.

 

[Curt Christianson]

<
<BTW Franc, I have no intention of conversing with 98 Guy on <any more
related
<issues or in this discussion, If you or others wish to do so <then .....

--
MEB

*Not* likely MEB! Whew


--
HTH,
Curt

Windows Support Center
www.aumha.org
Practically Nerded,...
http://dundats.mvps.org/Index.htm

"MEB" <meb@not here@hotmail.com> wrote in message
news:uyMPQhwyHHA.1576@TK2MSFTNGP03.phx.gbl...
|
|
|
|
| "MEB" <meb@not here@hotmail.com> wrote in message
| news:uFS78WuyHHA.3916@TK2MSFTNGP02.phx.gbl...
|| A CORRECTION the segment which reads:
||
|| A single partition on a 320 [or the 127 per the other discussion] gig
| drive
|| would potentially have how many using that same test?
||
||
|| SHOULD READ for clarification:
||
|| A single partition on a 320 [or 127 per the other discussion] gig drive
|| would potentially have how many files and folders using that same test to
|| fill the drive?
||
|| --
|| MEB
|| ________
||
||
|
| ADDENDUM: so these issues need not be separately addressed in some other
| discussion or later here:
|
| It would also be poignant to your [Franc] usage, and the testing and or
| purported results, to have:
|
| A. the hardware/chipset and other information which purports to allow such
| usage. related to the intended use and/or to support the findings
|
| B. IN ADDITION to information which may also have effect such as:
|
| 1. Any.system file replacements, examples: shell32, kernel32, user,
browser
| dlls, etc.. and whether IE is installed and what version, and update
| history to the testing system(s).
|
| 2. Any registry tweaks or other like activities such as: shell
replacements,
| explorer replacements, etc. which also may have an effect.
|
| 3. What was used in the fdisk and formatting process, and cluster size,
etc.
| pursuant those activities.
|
| 4. Tools/applications for routine disk maintenance of these large
| disks/partitions.
|
|
| BTW Franc, I have no intention of conversing with 98 Guy on any more
related
| issues or in this discussion, If you or others wish to do so then .....
|
| --
| MEB
| http://peoplescounsel.orgfree.com
| ________
|
|
|

 

[wheely]

> Look dude, I'd plonk your stupid stuff, but then you would have free range
> here...
>
> These are the EXACT questions you asked and were answered repeatedly in
ALL
> your discussions

MEB, i've lurked and lurked, and among others i believe the both
of you to be Excellent procurators of win98.

buuuut, MEB, you have let me down by going for a humans throat
instead of the 'ms supplied informations throat'.
98Guy is as always on track,, this due to his demeanor of seeking truth,
i support him 100%.
MEB, you too are always on track, so why not try to 'merge' with 98guy
for a while and Co-produce the truths...... surely it has to be a
worthy challenge? a grand challenge to both! who knows, you could
both come out the other end as friends, and that is a Win situation in
any wise mans language.

I support you both, you both have qualities.

 

[Franc Zabkar]

On Fri, 20 Jul 2007 15:57:43 -0400, "MEB" <meb@not here@hotmail.com>
put finger to keyboard and composed:

>"MEB" <meb@not here@hotmail.com> wrote in message
>news:uFS78WuyHHA.3916@TK2MSFTNGP02.phx.gbl...
>| A CORRECTION the segment which reads:
>|
>| A single partition on a 320 [or the 127 per the other discussion] gig
>drive
>| would potentially have how many using that same test?
>|
>|
>| SHOULD READ for clarification:
>|
>| A single partition on a 320 [or 127 per the other discussion] gig drive
>| would potentially have how many files and folders using that same test to
>| fill the drive?
>|
>| --
>| MEB
>| ________
>|
>|
>
> ADDENDUM: so these issues need not be separately addressed in some other
>discussion or later here:
>
> It would also be poignant to your [Franc] usage, and the testing and or
>purported results, to have:
>
>A. the hardware/chipset and other information which purports to allow such
>usage. related to the intended use and/or to support the findings

I'm using a USB hard drive. As a USB port talks to a HD in the same
way as it talks to a mouse, I would think that the motherboard chipset
would be irrelevant in this case. FWIW, the USB controller is an SiS
7001 / SiS 7002, and the motherboard chipset is SiS 746.

>B. IN ADDITION to information which may also have effect such as:
>
>1. Any.system file replacements, examples: shell32, kernel32, user, browser
>dlls, etc.. and whether IE is installed and what version, and update
>history to the testing system(s).

I used Win98SE's format.com and Microsoft's updated fdisk.exe.

FDISK EXE 64,460 05-18-00 8:35a FDISK.EXE
FORMAT COM 49,575 04-23-99 10:22p FORMAT.COM

The IE version is 6.0.2800.1106

>2. Any registry tweaks or other like activities such as: shell replacements,
>explorer replacements, etc. which also may have an effect.

In method#1 I used the MKDIR command in QBasic. In method#2 I used
4dos.com and "counted FOR loops". There is also a Win98 batch method
involving reentrant CALL statements that I haven't yet tried.

>3. What was used in the fdisk and formatting process, and cluster size, etc.
>pursuant those activities.

I left it up to fdisk and format to decide, not third party utilities.
Fdisk was executed from within a Windows DOS box.

DOS Chkdsk reports the following:

312,492,320 kilobytes total disk space
310,234,080 kilobytes free

32,768 bytes in each allocation unit
9,765,385 total allocation units on disk
9,694,815 available allocation units on disk

>4. Tools/applications for routine disk maintenance of these large
>disks/partitions.

Win98SE's scandisk and defrag both report "not enough memory".

Win ME's scandisk also complains of "not enough memory".

Win ME's defrag runs OK.

Testing is extremely slow, as my USB 2.0 port appears to be operating
at 1.1 speeds. So far it has taken me just under an hour to create
~69000 directories and about 0.5 hour to create 65534 zero length
files. This is on my Athlon XP 2500+ box. I don't think I want to try
for 1 million folders. :-(

What's complicating things is that, if I play around with Explorer
while the batch is running, the directory structure can become
corrupted, and sometimes the machine, or HD, hangs. Otherwise, if I
leave things alone, then the batch completes without incident. It took
me a while to figure this out, but I still don't understand why it
happens.

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.

 

[MEB]

"wheely" <onmyboard@ramp.com> wrote in message
news:uBNidkSzHHA.536@TK2MSFTNGP06.phx.gbl...
|
| > Look dude, I'd plonk your stupid stuff, but then you would have free
range
| > here...
| >
| > These are the EXACT questions you asked and were answered repeatedly in
| ALL
| > your discussions
|
| MEB, i've lurked and lurked, and among others i believe the both
| of you to be Excellent procurators of win98.
|
| buuuut, MEB, you have let me down by going for a humans throat
| instead of the 'ms supplied informations throat'.

You miss the point entirely. I am apparently one of the few here that WILL
address almost any issue, but ONLY if evidence is presented to support those
claims or reasonable argument is presented thereto.

This 98 Guy entity is repeatedly directed to supply the defining proof to
support the issues broached, yet none is ever produced which does not fail
due to: refusal to supply accurate and complete testing, or, via that to
which he has been directed.
Instead, we find reversion to the *Micr$shaft* attitude, which is
insufficient [and boring] when dealing with any issues not discussing that
aspect. Is it by design? Is there misleading information? Ah yeah,
sooooooo.....

REALITY:
Is there any group or individual who ACTUALLY expects to change Microsoft,
or the market, or public perception. Microsoft controls its segment, AND the
manufacturers in that segment. Even the government [under this present
administration] now LOVES Microsoft. It has NO problem issuing baseless
patents to Microsoft, and allowing Microsoft to destroy other's patents [see
the recent Microsoft - AT&T rulings, among others] or absorbing other
companies, which would have once been questioned as violations of the
anti-trust Laws and others.
But enough of that....

Here's an example: in 98 Guy's *Win98 can use SATA* rants, he fails to
fully address the issues of *DOS compatibility mode* which he had to use.
Moreover, it COMPLETELY ignores the full ramifications of such use... NOT
because others did not attempt to address these variables during those
discussions they did, and attempted to fill these various discussions with
vast amounts of documents, test results, etc .. but because it was
obviously counter to what this 98 Guy entity WISHED to present, rather than
the cold reality of facts to support such.
Opinions or personal findings are great, we all have them, but evidence
provides pursuassion lack of it provides nothing, or worse provides false
perception and potential damage.

WHEN 98 Guy then returns using that same information to which he was
directed, now waving it as if it some new found thing:
is it that the original presenters did not read the materials [duh, bet
they did]
is it that these issues were not tested [duh, bet they were]
is it anything new [duh, bet it can be found on hundreds of sites already]
is it a new tested area [duh, must not be],
and more importantly, did 98 Guy proof his presentation before addressing
it, with his own work?
Regretfully there is no DUH for that last one, the track record is plain,
and would have been the first aspect of any presentation were there
substantial or convincing proof FROM 98 Guy. And here's is the problem,
frankly I would NOT accept any now, it would be suspicious [and that, even
if I knew it true].

These issues were not un-addressed previously within this group. Search
Google within this discussion group and you WILL find such discussions:
from: USB issues to adapters which WILL supply the ability to circumvent
SOME of the limitations of the 9X OS to discussions of such OS limitations
and, attempts to address or circumvent other of those limitations via file
replacements and otherwise [though these still have areas which need
addressed].

AND yes, even the MVPs {supposedly or openly so disliked by many who
question Microsoft and its motive and methods} have attempted to work
through these issues.
Yet are "blasted" when they point out their PERSONAL
findings, just because they point out that Microsoft DID produce some
documentation and their testing found such as true OR their OPINION is
otherwise counter to some posters ideals.
In fact, one can do historical search related to the PUBLIC posters and
MVPs in this group, and see other forums/sites [and at times here] where
they have openly attacked or tested some of those Microsoft limitations, or
exposed such, including, at times, questions concerning Microsoft methods
and motives.
Attacking the MVPs personally is ludicrous, they obtained the
classification because they CONSISTANTLY attempt to help users, WITHOUT pay,
providing the best information they have [which may at times be wrong, but
*let he who is without sin...*{and that presents NO religious aspect}] NOT
because they are in some LOVE AFFAIR with Microsoft [though some may be
- ) ].
Heck, some use Linux and/or MAC and/or SUN [or other] in addition to
Microsoft products. BUT the topic on these servers and in these groups IS
Microsoft products. Will they question or trounce upon those who aren't
doing the *Microsoft Way* thing, at times but if you can't protect your
position in some discussion in a civil fashion, then why shouldn't they
point that out, you are, after all, in a Microsoft product forum.
There are thousands of other forums, many which would embrace any
particular or like ideal. Should this group bow to the whims of a few, if
those few can not fend for themselves or defend their ideals? Why would it.

MOREOVER, If you truly have been lurking, then you must have noticed the
lengthy discussions on Fat16 and Fat32 [and NTFS as well] which have
occurred over the last three or so years [and of course before then].

WITHOUT QUESTION: The high QUALITY of the materials and issues presented
within this group, should be maintained regardless of personal feelings or
desires.
Rant and rave, attack others if you wish, but do so in the thousands of
other USENET news groups out there, its expected.
HERE we are or should be attempting to supply support to each other for a
dead OS with essentially dead manufacturer support,... that knowledge,
reality, and understanding should NEVER be left behind or lost, or ignored.
This is definitely a Microsoft product, but an entirely new world for users
of it. This is the same world that Linux users have been
in for years... however they at least, without the constraints that
9X/ME/2000 users face, the LEGAL aspects...

| 98Guy is as always on track,, this due to his demeanor of seeking
truth,
| i support him 100%.

Such is your right to think such, I will not attempt to dissuade you.

| MEB, you too are always on track,

I would NOT say that. There are times when I have taken the role of
antagonist or protagonist, just to present the other side, or to bring to
light areas of interest or understanding..

| so why not try to 'merge' with 98guy
| for a while and Co-produce the truths...... surely it has to be a
| worthy challenge? a grand challenge to both! who knows, you could
| both come out the other end as friends, and that is a Win situation in
| any wise mans language.
|
| I support you both, you both have qualities.
|
|

Well, I have previously tried, but I have no further intention of doing so,
nor of re-doing prior testing [unless it suit MY task at hand or whim], when
this 98 Guy entity fails in that part [his part] of the issue... None.. nor
will I tolerate presentations which place unsuspecting individuals at risk,
merely for argument's sake. I intend to uphold my part of the bargain in
here until I fade to black... there ARE other things that call

Prior flames were ignored prior failures were ignored but my limit has
been reached, I will no longer do so with the party at issue.
Call me, or think of me, however you wish, it matters not, life continues
til the day it ceases to exist.
Nuff said...... that wasted far more time and band width than it was likely
worth, I doubt its valve will be digested by the majority.

You, however, are more than welcome to discuss 98 Guy's issues with him....

--
MEB
http://peoplescounsel.orgfree.com
________

 

[MEB]

Not much to add to your test yet, perhaps clarifications and thought:

"Franc Zabkar" <fzabkar@iinternode.on.net> wrote in message
news:c6baa3d2dc9k983tkm72oo020mf1pj55kb@4ax.com...
| On Fri, 20 Jul 2007 15:57:43 -0400, "MEB" <meb@not here@hotmail.com>
| put finger to keyboard and composed:
|
| >"MEB" <meb@not here@hotmail.com> wrote in message
| >news:uFS78WuyHHA.3916@TK2MSFTNGP02.phx.gbl...
| >| A CORRECTION the segment which reads:
| >|
| >| A single partition on a 320 [or the 127 per the other discussion] gig
| >drive
| >| would potentially have how many using that same test?
| >|
| >|
| >| SHOULD READ for clarification:
| >|
| >| A single partition on a 320 [or 127 per the other discussion] gig drive
| >| would potentially have how many files and folders using that same test
to
| >| fill the drive?
| >|
| >| --
| >| MEB
| >| ________
| >|
| >|
| >
| > ADDENDUM: so these issues need not be separately addressed in some other
| >discussion or later here:
| >
| > It would also be poignant to your [Franc] usage, and the testing and or
| >purported results, to have:
| >
| >A. the hardware/chipset and other information which purports to allow
such
| >usage. related to the intended use and/or to support the findings
|
| I'm using a USB hard drive. As a USB port talks to a HD in the same
| way as it talks to a mouse, I would think that the motherboard chipset
| would be irrelevant in this case. FWIW, the USB controller is an SiS
| 7001 / SiS 7002, and the motherboard chipset is SiS 746.

The USB controller you reference is in the USB enclosure?
The manufacturer?
Or are you referring to the on board USB?

|
| >B. IN ADDITION to information which may also have effect such as:
| >
| >1. Any.system file replacements, examples: shell32, kernel32, user,
browser
| >dlls, etc.. and whether IE is installed and what version, and update
| >history to the testing system(s).
|
| I used Win98SE's format.com and Microsoft's updated fdisk.exe.
|
| FDISK EXE 64,460 05-18-00 8:35a FDISK.EXE
| FORMAT COM 49,575 04-23-99 10:22p FORMAT.COM

For reference, you chose this method instead of the manufacturer's tool(s)
for what reason or were there none available or recommended?

|
| The IE version is 6.0.2800.1106
|
| >2. Any registry tweaks or other like activities such as: shell
replacements,
| >explorer replacements, etc. which also may have an effect.
|
| In method#1 I used the MKDIR command in QBasic. In method#2 I used
| 4dos.com and "counted FOR loops". There is also a Win98 batch method
| involving reentrant CALL statements that I haven't yet tried.
|
| >3. What was used in the fdisk and formatting process, and cluster size,
etc.
| >pursuant those activities.
|
| I left it up to fdisk and format to decide, not third party utilities.
| Fdisk was executed from within a Windows DOS box.
|
| DOS Chkdsk reports the following:
|
| 312,492,320 kilobytes total disk space
| 310,234,080 kilobytes free
|
| 32,768 bytes in each allocation unit
| 9,765,385 total allocation units on disk
| 9,694,815 available allocation units on disk
|
| >4. Tools/applications for routine disk maintenance of these large
| >disks/partitions.
|
| Win98SE's scandisk and defrag both report "not enough memory".
|
| Win ME's scandisk also complains of "not enough memory".
|
| Win ME's defrag runs OK.
|
| Testing is extremely slow, as my USB 2.0 port appears to be operating
| at 1.1 speeds. So far it has taken me just under an hour to create
| ~69000 directories and about 0.5 hour to create 65534 zero length
| files. This is on my Athlon XP 2500+ box. I don't think I want to try
| for 1 million folders. :-(

Doesn't sound like that million plus is feasible in this situation.

CLARIFICATION:
Are you using the in built support drivers for USB from 98SE, or are you
using something like the Maximus Decim drivers, and/or the manufacturer's
enclosure drivers (if any)?

Are there other USB devices attached which might be affecting the bandwidth
or power available [e.g., keyboard, printer, modem, etc., and device
positioning/assigment/settings any additional hub involved]?

|
| What's complicating things is that, if I play around with Explorer
| while the batch is running, the directory structure can become
| corrupted, and sometimes the machine, or HD, hangs. Otherwise, if I
| leave things alone, then the batch completes without incident. It took
| me a while to figure this out, but I still don't understand why it
| happens.
|
| - Franc Zabkar
| --
| Please remove one 'i' from my address when replying by email.

Did you setup or run these batches with standard, or background, or
modified priority settings?
Did you setup or run these batches with automatic pause, or other
re-assignment, should the system and/or applications require more memory or
priority?
Did you setup or run these batches with exclusive, standard, or modified
memory settings?
Are you allowing or using or attempting any visual aspects within or
related to running these batches [e.g., DOS box, monitoring, or otherwise]?

A THOUGHT:
Explorer will attempt to PREPARE to show the drive [and its contents] when
running [visually evidenced by the drive name {the label if any} and
assignment]. If the drive is constantly changing during the attempt, there
will or may, be conflict. There are also *refresh cycles* involved and which
should be taken under consideration.
Perhaps someone can advise of how to temporarily exclude that drive from
Explorer access during the batch drive fill runs. Or if this would have the
desired effect, or any effect.

BTW Franc, thanks for doing the test in this forum. This is the *on-hands*,
*I watched this occur* activity which will hopefully verify the results and
clarify any issues. My hat is tipped to you ...
I see another test is also being presented in the group, perhaps two issues
may be resolved.

--
MEB
http://peoplescounsel.orgfree.com
________

 

[Franc Zabkar]

On Tue, 24 Jul 2007 11:59:40 -0400, "MEB" <meb@not here@hotmail.com>
put finger to keyboard and composed:

>Not much to add to your test yet, perhaps clarifications and thought:
>
>"Franc Zabkar" <fzabkar@iinternode.on.net> wrote in message
>news:c6baa3d2dc9k983tkm72oo020mf1pj55kb@4ax.com...

>| I'm using a USB hard drive. As a USB port talks to a HD in the same
>| way as it talks to a mouse, I would think that the motherboard chipset
>| would be irrelevant in this case. FWIW, the USB controller is an SiS
>| 7001 / SiS 7002, and the motherboard chipset is SiS 746.
>
> The USB controller you reference is in the USB enclosure?
> The manufacturer?
> Or are you referring to the on board USB?

The SiS 7001 / 7002 controllers are on the motherboard. The SiS 7001
is the 1.1 controller, and the SiS 7002 is 2.0.

I have been struggling to get the drive to connect at USB 2.0. No
matter what I do, Usbview always shows that the device is attached to
the SiS 7001. :-(

FWIW, I'm using the Orangeware driver. It installs OK, but maybe it is
not enabled for SiS controllers. ??? BTW, USB 2.0 is enabled in the
BIOS.

Anyway, despite my initial enthusiasm, it looks like any HD testing is
going to be unbearably slow. If it's going to take a whole day for
scandisk to run, then it's not worth pursuing this idea.

>| >B. IN ADDITION to information which may also have effect such as:
>| >
>| >1. Any.system file replacements, examples: shell32, kernel32, user,
>browser
>| >dlls, etc.. and whether IE is installed and what version, and update
>| >history to the testing system(s).
>|
>| I used Win98SE's format.com and Microsoft's updated fdisk.exe.
>|
>| FDISK EXE 64,460 05-18-00 8:35a FDISK.EXE
>| FORMAT COM 49,575 04-23-99 10:22p FORMAT.COM
>
> For reference, you chose this method instead of the manufacturer's tool(s)
>for what reason or were there none available or recommended?

I wanted to see whether standard procedures would suffice. In any case
all my files will be huge (the drive will be a video jukebox), so
large cluster sizes are not a problem for me.

>| Testing is extremely slow, as my USB 2.0 port appears to be operating
>| at 1.1 speeds. So far it has taken me just under an hour to create
>| ~69000 directories and about 0.5 hour to create 65534 zero length
>| files. This is on my Athlon XP 2500+ box. I don't think I want to try
>| for 1 million folders. :-(
>
> Doesn't sound like that million plus is feasible in this situation.

No, life is too short.

>CLARIFICATION:
>Are you using the in built support drivers for USB from 98SE, or are you
>using something like the Maximus Decim drivers, and/or the manufacturer's
>enclosure drivers (if any)?

I'm using the files provided with the Maximus Decim driver set.

Driver: USBNTMAP.SYS
File Size: 7136 (0x1BE0)
File Date: 2/17/2003 6:29 AM
Company Name: Microsoft Corporation
File Version: 4.90.3000

Driver: USBSTOR.SYS
File Size: 21040 (0x5230)
File Date: 2/17/2003 6:29 AM
Company Name: Microsoft Corporation
File Version: 4.90.3000.1

> Are there other USB devices attached which might be affecting the bandwidth
>or power available [e.g., keyboard, printer, modem, etc., and device
>positioning/assigment/settings any additional hub involved]?

No, its purely a 1.1 versus 2.0 issue. I have a USB printer, but it's
turned off.

>| What's complicating things is that, if I play around with Explorer
>| while the batch is running, the directory structure can become
>| corrupted, and sometimes the machine, or HD, hangs. Otherwise, if I
>| leave things alone, then the batch completes without incident. It took
>| me a while to figure this out, but I still don't understand why it
>| happens.
>|
>| - Franc Zabkar

> Did you setup or run these batches with standard, or background, or
>modified priority settings?
> Did you setup or run these batches with automatic pause, or other
>re-assignment, should the system and/or applications require more memory or
>priority?
> Did you setup or run these batches with exclusive, standard, or modified
>memory settings?
> Are you allowing or using or attempting any visual aspects within or
>related to running these batches [e.g., DOS box, monitoring, or otherwise]?

I'm just running them in a Windows DOS box with nothing else running
in the background. No special settings.

FWIW, I have been corresponding with another fellow via email whose
execution times are much worse than mine.

>A THOUGHT:
> Explorer will attempt to PREPARE to show the drive [and its contents] when
>running [visually evidenced by the drive name {the label if any} and
>assignment]. If the drive is constantly changing during the attempt, there
>will or may, be conflict. There are also *refresh cycles* involved and which
>should be taken under consideration.
> Perhaps someone can advise of how to temporarily exclude that drive from
>Explorer access during the batch drive fill runs. Or if this would have the
>desired effect, or any effect.

All is OK if I don't open an Explorer window, or if I leave it alone
when I do.

>BTW Franc, thanks for doing the test in this forum. This is the *on-hands*,
>*I watched this occur* activity which will hopefully verify the results and
>clarify any issues. My hat is tipped to you ...
> I see another test is also being presented in the group, perhaps two issues
>may be resolved.

Sorry, but unless I can resolve this USB speed issue, then I'm not
going to get very far. I had no idea that execution times were going
to be so long, even via a high speed SATA or PATA interface.

Here is a thread at alt.msdos.batch where some people have provided
batch routines:

http://groups.google.com/group/alt....b0210?lnk=st&q=&rnum=1&hl=en#6f94e846928b0210

http://tinyurl.com/2s634r

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.

 

[MEB]

"Franc Zabkar" <fzabkar@iinternode.on.net> wrote in message
news:dlina35gum3rkmnckk17hlbac0mh9nq5if@4ax.com...
| On Tue, 24 Jul 2007 11:59:40 -0400, "MEB" <meb@not here@hotmail.com>
| put finger to keyboard and composed:
|
| >Not much to add to your test yet, perhaps clarifications and thought:
| >
| >"Franc Zabkar" <fzabkar@iinternode.on.net> wrote in message
| >news:c6baa3d2dc9k983tkm72oo020mf1pj55kb@4ax.com...
|
| >| I'm using a USB hard drive. As a USB port talks to a HD in the same
| >| way as it talks to a mouse, I would think that the motherboard chipset
| >| would be irrelevant in this case. FWIW, the USB controller is an SiS
| >| 7001 / SiS 7002, and the motherboard chipset is SiS 746.
| >
| > The USB controller you reference is in the USB enclosure?
| > The manufacturer?
| > Or are you referring to the on board USB?
|
| The SiS 7001 / 7002 controllers are on the motherboard. The SiS 7001
| is the 1.1 controller, and the SiS 7002 is 2.0.
|
| I have been struggling to get the drive to connect at USB 2.0. No
| matter what I do, Usbview always shows that the device is attached to
| the SiS 7001. :-(

Is it possible to disable that 7001 in Device Manager forcing 2.0 [or
remove the 7001]? Or did you check to see what Device Manger shows [like its
not configured to use 1.1 compatibility or something]?

|
| FWIW, I'm using the Orangeware driver. It installs OK, but maybe it is
| not enabled for SiS controllers. ??? BTW, USB 2.0 is enabled in the
| BIOS.

Q: Can you tell me briefly about your driver?
The USB 2.0 software drivers give the user the ability to utilize USB 2.0
High Speed devices with their NEC, Intel and many more devices. These
software drivers are compatible with the Mac OS X, and with the Windows 98,
98SE, Me, 2000, XP and Server 2003.

The software driver is customized with the customer's unique subvendor and
subdevice ID numbers. The customization process requires that the USB 2.0
device have an EEPROM chip. The customer programs the EEPROM chip with its
unique subvendor and subdevice ID combination. OrangeWare engineers program
the ID combination directly into the software code preventing the software
driver from operating on any unlicensed hardware. The software driver must
detect the unique subvendor and subdevice ID combination in the EEPROM chip
and match it with one of the combinations programmed in the software code in
order for the driver to work.
http://www.orangeware.com/developers/faq_usb2.html


|
| Anyway, despite my initial enthusiasm, it looks like any HD testing is
| going to be unbearably slow. If it's going to take a whole day for
| scandisk to run, then it's not worth pursuing this idea.

Looks that way, particularly with your USB limited access.

If you noted, there is another offering on MGDX that supplies an
alternative using the old Helix Nuts & Bolts disk maintenance utilities
[which were a heck of a lot faster than default Microsoft tools]. Then
again, I have no indication [or test results] they would even work upon
disks as large as those being tested...
http://www.mdgx.com/files/diskmind.php
http://www.mdgx.com/add.htm
http://www.mdgx.com/files/DISKMIND.EXE
FROM THE INCLUDED DISKMIND.txt:
DMindr32 + DmDOS support FAT12, FAT16 + FAT32.
Both should *theoretically* support all (E)IDE/(U)DMA/(P)ATA/SATA hard
drives/partitions up to 2 TB (TeraBytes). More testing is needed. [?]
DMindr32 supports Long File Names (LFNs).
DmDOS does NOT support LFNs because native MS-DOS does NOT support LFNs
without a dedicated tool like DOSLFN (free open source):
http://www.geocities.com/jadoxa/doslfn/
loaded in memory (as TSR) from native MS-DOS command line [or autoexec.bat]

REFERENCED HERE:
http://www.msfn.org/board/index.php?showtopic=46581&view=findpost&p=323675

I do remember [back in the day] there was some trick involved, where you
could replace the default scandisk with these tools.. [but what, my memory
isn't that good /renaming to scandisk or something\]

|
| >| >B. IN ADDITION to information which may also have effect such as:
| >| >
| >| >1. Any.system file replacements, examples: shell32, kernel32, user,
| >browser
| >| >dlls, etc.. and whether IE is installed and what version, and update
| >| >history to the testing system(s).
| >|
| >| I used Win98SE's format.com and Microsoft's updated fdisk.exe.
| >|
| >| FDISK EXE 64,460 05-18-00 8:35a FDISK.EXE
| >| FORMAT COM 49,575 04-23-99 10:22p FORMAT.COM
| >
| > For reference, you chose this method instead of the manufacturer's
tool(s)
| >for what reason or were there none available or recommended?
|
| I wanted to see whether standard procedures would suffice. In any case
| all my files will be huge (the drive will be a video jukebox), so
| large cluster sizes are not a problem for me.

Well a 32k cluster isn't going to kill anyone anyway.. particularly for
your intended use... perhaps your test is more in line with tools and large
file manipulation concerns anyway... [then again you still have fat and
other 98 OS aspects to consider for your 98 contact]

|
| >| Testing is extremely slow, as my USB 2.0 port appears to be operating
| >| at 1.1 speeds. So far it has taken me just under an hour to create
| >| ~69000 directories and about 0.5 hour to create 65534 zero length
| >| files. This is on my Athlon XP 2500+ box. I don't think I want to try
| >| for 1 million folders. :-(
| >
| > Doesn't sound like that million plus is feasible in this situation.
|
| No, life is too short.

WAAAAAAAAAAAAy too short.. [or too long depending upon your outlook]

|
| >CLARIFICATION:
| >Are you using the in built support drivers for USB from 98SE, or are you
| >using something like the Maximus Decim drivers, and/or the manufacturer's
| >enclosure drivers (if any)?
|
| I'm using the files provided with the Maximus Decim driver set.
|
| Driver: USBNTMAP.SYS
| File Size: 7136 (0x1BE0)
| File Date: 2/17/2003 6:29 AM
| Company Name: Microsoft Corporation
| File Version: 4.90.3000
|
| Driver: USBSTOR.SYS
| File Size: 21040 (0x5230)
| File Date: 2/17/2003 6:29 AM
| Company Name: Microsoft Corporation
| File Version: 4.90.3000.1

Hmm, that didn't help me, I show those in the 23E and the 31 version. Which
version did you use?

|
| > Are there other USB devices attached which might be affecting the
bandwidth
| >or power available [e.g., keyboard, printer, modem, etc., and device
| >positioning/assigment/settings any additional hub involved]?
|
| No, its purely a 1.1 versus 2.0 issue. I have a USB printer, but it's
| turned off.

I think your correct, iron that out and you will at least have full
bandwidth..

|
| >| What's complicating things is that, if I play around with Explorer
| >| while the batch is running, the directory structure can become
| >| corrupted, and sometimes the machine, or HD, hangs. Otherwise, if I
| >| leave things alone, then the batch completes without incident. It took
| >| me a while to figure this out, but I still don't understand why it
| >| happens.
| >|
| >| - Franc Zabkar
|
| > Did you setup or run these batches with standard, or background, or
| >modified priority settings?
| > Did you setup or run these batches with automatic pause, or other
| >re-assignment, should the system and/or applications require more memory
or
| >priority?
| > Did you setup or run these batches with exclusive, standard, or modified
| >memory settings?
| > Are you allowing or using or attempting any visual aspects within or
| >related to running these batches [e.g., DOS box, monitoring, or
otherwise]?
|
| I'm just running them in a Windows DOS box with nothing else running
| in the background. No special settings.

Well I would attempt to setup something else, though that's likely hit or
miss... then again you're really only testing for your intended use > DivX
and 98, will they play well together with this large disk...

|
| FWIW, I have been corresponding with another fellow via email whose
| execution times are much worse than mine.

Yeah, this is a whole new world.... bigger disks = bigger headaches
Differring BIOS access schemes, differing SATA support, etc...

|
| >A THOUGHT:
| > Explorer will attempt to PREPARE to show the drive [and its contents]
when
| >running [visually evidenced by the drive name {the label if any} and
| >assignment]. If the drive is constantly changing during the attempt,
there
| >will or may, be conflict. There are also *refresh cycles* involved and
which
| >should be taken under consideration.
| > Perhaps someone can advise of how to temporarily exclude that drive from
| >Explorer access during the batch drive fill runs. Or if this would have
the
| >desired effect, or any effect.
|
| All is OK if I don't open an Explorer window, or if I leave it alone
| when I do.
|
| >BTW Franc, thanks for doing the test in this forum. This is the
*on-hands*,
| >*I watched this occur* activity which will hopefully verify the results
and
| >clarify any issues. My hat is tipped to you ...
| > I see another test is also being presented in the group, perhaps two
issues
| >may be resolved.
|
| Sorry, but unless I can resolve this USB speed issue, then I'm not
| going to get very far. I had no idea that execution times were going
| to be so long, even via a high speed SATA or PATA interface.

Well here's a good reason most keep their partitions small.. one big disk
and you really can't expect fast times, to much "real estate"... even with
new algorithms and disk access schemes... and supposed OSs to handle them
with dual or quad processors... so though it may seem blazingly fast just
how much time and effort was actually used....


|
| Here is a thread at alt.msdos.batch where some people have provided
| batch routines:
|
|
http://groups.google.com/group/alt....b0210?lnk=st&q=&rnum=1&hl=en#6f94e846928b0210
|
| http://tinyurl.com/2s634r
|
| - Franc Zabkar
| --
| Please remove one 'i' from my address when replying by email.

Well I guess I need to look over there...

--
MEB
http://peoplescounsel.orgfree.com
________

 

[MEB]

"MEB" <meb@not here@hotmail.com> wrote in message
news:O7t09Ja0HHA.464@TK2MSFTNGP02.phx.gbl...
| "Franc Zabkar" <fzabkar@iinternode.on.net> wrote in message
| news:dlina35gum3rkmnckk17hlbac0mh9nq5if@4ax.com...
| | On Tue, 24 Jul 2007 11:59:40 -0400, "MEB" <meb@not here@hotmail.com>
| | >"Franc Zabkar" <fzabkar@iinternode.on.net> wrote in message
| | >news:c6baa3d2dc9k983tkm72oo020mf1pj55kb@4ax.com...
|
| |
| | Here is a thread at alt.msdos.batch where some people have provided
| | batch routines:
| |
| |
|
http://groups.google.com/group/alt....b0210?lnk=st&q=&rnum=1&hl=en#6f94e846928b0210
| |
| | http://tinyurl.com/2s634r
| |
| | - Franc Zabkar
| | --
| | Please remove one 'i' from my address when replying by email.
|
| Well I guess I need to look over there...
|
| --
| MEB
| http://peoplescounsel.orgfree.com
| ________
|
|

Perhaps your not really interested, but having looked at those
scripts/batches, I would tend to think that something in VB would much
better suit your needs. At least that gives 32bit virtual access with native
support in Win98. You should have at least one version's [5, 6] DLLs already
installed, perhaps both..

That would also negate the need to run in a dos box.. and allow pure
background activity and priority..

--
MEB
http://peoplescounsel.orgfree.com
________

 

[Franc Zabkar]

On Sun, 29 Jul 2007 01:37:07 -0400, "MEB" <meb@not here@hotmail.com>
put finger to keyboard and composed:

>| The SiS 7001 / 7002 controllers are on the motherboard. The SiS 7001
>| is the 1.1 controller, and the SiS 7002 is 2.0.
>|
>| I have been struggling to get the drive to connect at USB 2.0. No
>| matter what I do, Usbview always shows that the device is attached to
>| the SiS 7001. :-(
>
> Is it possible to disable that 7001 in Device Manager forcing 2.0 [or
>remove the 7001]?

The problem is that I don't know whether the chipset is switching
between USB controllers on its own, or whether the driver is doing it.

Anyway, I disabled both USB 1.1 controllers in DM and got a blue
screen error (openhci.sys) when I launched usbview.exe. I launched it
a second time and this time I saw that the SiS 7001 entries were still
present but there were no data associated with them. The USB 2.0
controller showed up as usual but the HD was not attached to it.

>Or did you check to see what Device Manger shows [like its
>not configured to use 1.1 compatibility or something]?

I don't see anything like that.

>| FWIW, I'm using the Orangeware driver. It installs OK, but maybe it is
>| not enabled for SiS controllers. ??? BTW, USB 2.0 is enabled in the
>| BIOS.
>
>Q: Can you tell me briefly about your driver?

>The USB 2.0 software drivers give the user the ability to utilize USB 2.0
>High Speed devices with their NEC, Intel and many more devices. These
>software drivers are compatible with the Mac OS X, and with the Windows 98,
>98SE, Me, 2000, XP and Server 2003.
>
>The software driver is customized with the customer's unique subvendor and
>subdevice ID numbers. The customization process requires that the USB 2.0
>device have an EEPROM chip. The customer programs the EEPROM chip with its
>unique subvendor and subdevice ID combination. OrangeWare engineers program
>the ID combination directly into the software code preventing the software
>driver from operating on any unlicensed hardware. The software driver must
>detect the unique subvendor and subdevice ID combination in the EEPROM chip
>and match it with one of the combinations programmed in the software code in
>order for the driver to work.

>http://www.orangeware.com/developers/faq_usb2.html

Thanks for that. I had already noticed that and I can confirm that the
subvendor and subdevice IDs that appear in my registry are not
supported in OrangeWare's INF file. I had considered editing the INF
to include them, but after reading OrangeWare's FAQ I decided that it
would be pointless. I'll do it anyway, though, as I'm getting
desperate.

BTW, I looked around the southbridge chip for the EEPROM but found
only unpopulated IC locations. Maybe the southbridge is internally
programmed for ECS by SiS ???

FWIW, my M571 socket 7 board has the same SiS 7001 controller but its
sub IDs are zeroes.

>| Anyway, despite my initial enthusiasm, it looks like any HD testing is
>| going to be unbearably slow. If it's going to take a whole day for
>| scandisk to run, then it's not worth pursuing this idea.
>
> Looks that way, particularly with your USB limited access.

Well, USB 1.1 maxes out at 12 Mbit/s, so this would mean that it would
take about 2.5 days to fill the whole disc with data. :-(

> If you noted, there is another offering on MGDX that supplies an
>alternative using the old Helix Nuts & Bolts disk maintenance utilities
>[which were a heck of a lot faster than default Microsoft tools]. Then
>again, I have no indication [or test results] they would even work upon
>disks as large as those being tested...
>http://www.mdgx.com/files/diskmind.php
>http://www.mdgx.com/add.htm
>http://www.mdgx.com/files/DISKMIND.EXE
>FROM THE INCLUDED DISKMIND.txt:
>DMindr32 + DmDOS support FAT12, FAT16 + FAT32.
>Both should *theoretically* support all (E)IDE/(U)DMA/(P)ATA/SATA hard
>drives/partitions up to 2 TB (TeraBytes). More testing is needed. [?]
>DMindr32 supports Long File Names (LFNs).
>DmDOS does NOT support LFNs because native MS-DOS does NOT support LFNs
>without a dedicated tool like DOSLFN (free open source):
>http://www.geocities.com/jadoxa/doslfn/
>loaded in memory (as TSR) from native MS-DOS command line [or autoexec.bat]
>
>REFERENCED HERE:
>http://www.msfn.org/board/index.php?showtopic=46581&view=findpost&p=323675
>
> I do remember [back in the day] there was some trick involved, where you
>could replace the default scandisk with these tools.. [but what, my memory
>isn't that good /renaming to scandisk or something\]

I'll have to get to this later because at the moment I can't keep my
external HD or enclosure from hanging. Originally I thought it was an
Explorer issue, but now I have managed to make it hang while running
Xcopy in a DOS box on my socket 7 machine.

Tomorrow I intend to hook up the kit to my friend's XP / USB 2.0 PC.
That should help me narrow down the problem.

<snip>

>| I'm using the files provided with the Maximus Decim driver set.
>|
>| Driver: USBNTMAP.SYS
>| File Size: 7136 (0x1BE0)
>| File Date: 2/17/2003 6:29 AM
>| Company Name: Microsoft Corporation
>| File Version: 4.90.3000
>|
>| Driver: USBSTOR.SYS
>| File Size: 21040 (0x5230)
>| File Date: 2/17/2003 6:29 AM
>| Company Name: Microsoft Corporation
>| File Version: 4.90.3000.1
>
> Hmm, that didn't help me, I show those in the 23E and the 31 version. Which
>version did you use?

I can't remember, but my usbstor.inf file has this annotation:

Maximus Decim modified - 06/14/2006, 01.02

>| > Are there other USB devices attached which might be affecting the
>bandwidth
>| >or power available [e.g., keyboard, printer, modem, etc., and device
>| >positioning/assigment/settings any additional hub involved]?
>|
>| No, its purely a 1.1 versus 2.0 issue. I have a USB printer, but it's
>| turned off.
>
> I think your correct, iron that out and you will at least have full
>bandwidth..

There are two 1.1 root hubs with three ports each. The printer is on a
different hub.

>| >| What's complicating things is that, if I play around with Explorer
>| >| while the batch is running, the directory structure can become
>| >| corrupted, and sometimes the machine, or HD, hangs. Otherwise, if I
>| >| leave things alone, then the batch completes without incident. It took
>| >| me a while to figure this out, but I still don't understand why it
>| >| happens.
>| >|
>| >| - Franc Zabkar
>|
>| > Did you setup or run these batches with standard, or background, or
>| >modified priority settings?
>| > Did you setup or run these batches with automatic pause, or other
>| >re-assignment, should the system and/or applications require more memory
>or
>| >priority?
>| > Did you setup or run these batches with exclusive, standard, or modified
>| >memory settings?
>| > Are you allowing or using or attempting any visual aspects within or
>| >related to running these batches [e.g., DOS box, monitoring, or
>otherwise]?
>|
>| I'm just running them in a Windows DOS box with nothing else running
>| in the background. No special settings.
>
> Well I would attempt to setup something else, though that's likely hit or
>miss... then again you're really only testing for your intended use > DivX
>and 98, will they play well together with this large disk...

I've copied a bunch of test AVI files to the drive and had them
playing just fine via the USB 1.1 port on my $40 supermarket DVD
player.

My next step is to determine whether the DVD player will recognise
more than one partition, and whether it has any file limits, etc.

Incidentally there were no hangs while I was playing these files
through my DVD. Maybe it's a write issue.

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.

 

[MEB]

"Franc Zabkar" <fzabkar@iinternode.on.net> wrote in message
news:lke0b3dtbgcrrgop072bls7fp0kvksfifa@4ax.com...
| On Sun, 29 Jul 2007 01:37:07 -0400, "MEB" <meb@not here@hotmail.com>
| put finger to keyboard and composed:
|
| >| The SiS 7001 / 7002 controllers are on the motherboard. The SiS 7001
| >| is the 1.1 controller, and the SiS 7002 is 2.0.
| >|
| >| I have been struggling to get the drive to connect at USB 2.0. No
| >| matter what I do, Usbview always shows that the device is attached to
| >| the SiS 7001. :-(
| >
| > Is it possible to disable that 7001 in Device Manager forcing 2.0 [or
| >remove the 7001]?
|
| The problem is that I don't know whether the chipset is switching
| between USB controllers on its own, or whether the driver is doing it.

Did you try something like filemon or regmon by sysinternals to observe
what was going on?
Potential registry settings?

|
| Anyway, I disabled both USB 1.1 controllers in DM and got a blue
| screen error (openhci.sys) when I launched usbview.exe. I launched it
| a second time and this time I saw that the SiS 7001 entries were still
| present but there were no data associated with them. The USB 2.0
| controller showed up as usual but the HD was not attached to it.

So you were apparently using the 1.1 aspect, killed that, BSODed, restarted
the computer, and the 2.0 didn't kick in?
Did you do a *Find New Harware* to correct registry, VMM, and other
aspects?

|
| >Or did you check to see what Device Manger shows [like its
| >not configured to use 1.1 compatibility or something]?
|
| I don't see anything like that.

Opps, my error, sometimes I forget what I'm discussing...

|
| >| FWIW, I'm using the Orangeware driver. It installs OK, but maybe it is
| >| not enabled for SiS controllers. ??? BTW, USB 2.0 is enabled in the
| >| BIOS.
| >
| >Q: Can you tell me briefly about your driver?
|
| >The USB 2.0 software drivers give the user the ability to utilize USB 2.0
| >High Speed devices with their NEC, Intel and many more devices. These
| >software drivers are compatible with the Mac OS X, and with the Windows
98,
| >98SE, Me, 2000, XP and Server 2003.
| >
| >The software driver is customized with the customer's unique subvendor
and
| >subdevice ID numbers. The customization process requires that the USB 2.0
| >device have an EEPROM chip. The customer programs the EEPROM chip with
its
| >unique subvendor and subdevice ID combination. OrangeWare engineers
program
| >the ID combination directly into the software code preventing the
software
| >driver from operating on any unlicensed hardware. The software driver
must
| >detect the unique subvendor and subdevice ID combination in the EEPROM
chip
| >and match it with one of the combinations programmed in the software code
in
| >order for the driver to work.
|
| >http://www.orangeware.com/developers/faq_usb2.html
|
| Thanks for that. I had already noticed that and I can confirm that the
| subvendor and subdevice IDs that appear in my registry are not
| supported in OrangeWare's INF file. I had considered editing the INF
| to include them, but after reading OrangeWare's FAQ I decided that it
| would be pointless. I'll do it anyway, though, as I'm getting
| desperate.
|
| BTW, I looked around the southbridge chip for the EEPROM but found
| only unpopulated IC locations. Maybe the southbridge is internally
| programmed for ECS by SiS ???
|
| FWIW, my M571 socket 7 board has the same SiS 7001 controller but its
| sub IDs are zeroes.
|
| >| Anyway, despite my initial enthusiasm, it looks like any HD testing is
| >| going to be unbearably slow. If it's going to take a whole day for
| >| scandisk to run, then it's not worth pursuing this idea.
| >
| > Looks that way, particularly with your USB limited access.
|
| Well, USB 1.1 maxes out at 12 Mbit/s, so this would mean that it would
| take about 2.5 days to fill the whole disc with data. :-(
|
| > If you noted, there is another offering on MGDX that supplies an
| >alternative using the old Helix Nuts & Bolts disk maintenance utilities
| >[which were a heck of a lot faster than default Microsoft tools]. Then
| >again, I have no indication [or test results] they would even work upon
| >disks as large as those being tested...
| >http://www.mdgx.com/files/diskmind.php
| >http://www.mdgx.com/add.htm
| >http://www.mdgx.com/files/DISKMIND.EXE
| >FROM THE INCLUDED DISKMIND.txt:
| >DMindr32 + DmDOS support FAT12, FAT16 + FAT32.
| >Both should *theoretically* support all (E)IDE/(U)DMA/(P)ATA/SATA hard
| >drives/partitions up to 2 TB (TeraBytes). More testing is needed. [?]
| >DMindr32 supports Long File Names (LFNs).
| >DmDOS does NOT support LFNs because native MS-DOS does NOT support LFNs
| >without a dedicated tool like DOSLFN (free open source):
| >http://www.geocities.com/jadoxa/doslfn/
| >loaded in memory (as TSR) from native MS-DOS command line [or
autoexec.bat]
| >
| >REFERENCED HERE:
|
>http://www.msfn.org/board/index.php?showtopic=46581&view=findpost&p=323675
| >
| > I do remember [back in the day] there was some trick involved, where you
| >could replace the default scandisk with these tools.. [but what, my
memory
| >isn't that good /renaming to scandisk or something\]
|
| I'll have to get to this later because at the moment I can't keep my
| external HD or enclosure from hanging. Originally I thought it was an
| Explorer issue, but now I have managed to make it hang while running
| Xcopy in a DOS box on my socket 7 machine.

Ah, that code and EEPROM activity/identity is also related to the actual
device.......

|
| Tomorrow I intend to hook up the kit to my friend's XP / USB 2.0 PC.
| That should help me narrow down the problem.

Well don't forget that WILL at least add information to the first track of
the disk,,,, remember??

|
| <snip>
|
| >| I'm using the files provided with the Maximus Decim driver set.
| >|
| >| Driver: USBNTMAP.SYS
| >| File Size: 7136 (0x1BE0)
| >| File Date: 2/17/2003 6:29 AM
| >| Company Name: Microsoft Corporation
| >| File Version: 4.90.3000
| >|
| >| Driver: USBSTOR.SYS
| >| File Size: 21040 (0x5230)
| >| File Date: 2/17/2003 6:29 AM
| >| Company Name: Microsoft Corporation
| >| File Version: 4.90.3000.1
| >
| > Hmm, that didn't help me, I show those in the 23E and the 31 version.
Which
| >version did you use?
|
| I can't remember, but my usbstor.inf file has this annotation:
|
| Maximus Decim modified - 06/14/2006, 01.02

I only have the 23E, 24 update, and the 31 versions locally so I can't tell
you what the full 24 would have been, BUT within the two full versions
usbstor.inf:

23E -
Maximus Decim modified - 06/14/2006, 01.02

31 -
Maximus Decim modified - 06/14/2006, 01.02
Maximus Decim modified - 08/31/2006, 01.03 add iPod, fix Sony Digital
Camera
Maximus Decim modified - 09/02/2006, 01.04 add LayoutFile
Maximus Decim modified - 25/01/2007, 01.05 add wdmstub, remove USB FDD,
add CnMemory
Maximus Decim modified - 07/03/2007, 01.06 remove wdmstub,

There were warnings regarding the 31 that ALL USB drivers, hubs, etc.,
needed removed BEFORE installing, but looking within the zip and the
contained files and INFs, I think that was not only advisable, but for full
effect of what this version addressed. It was a MAJOR new driver
setup/generic support ... designed to address a number of issues observed
concerning the older versions, and difficulties *cross driving* [necessity
to use extra/other drivers] USB devices. I wish I could say I tested it, but
I have no present USB 2.0 devices.

|
| >| > Are there other USB devices attached which might be affecting the
| >bandwidth
| >| >or power available [e.g., keyboard, printer, modem, etc., and device
| >| >positioning/assigment/settings any additional hub involved]?
| >|
| >| No, its purely a 1.1 versus 2.0 issue. I have a USB printer, but it's
| >| turned off.
| >
| > I think your correct, iron that out and you will at least have full
| >bandwidth..
|
| There are two 1.1 root hubs with three ports each. The printer is on a
| different hub.
|
| >| >| What's complicating things is that, if I play around with Explorer
| >| >| while the batch is running, the directory structure can become
| >| >| corrupted, and sometimes the machine, or HD, hangs. Otherwise, if I
| >| >| leave things alone, then the batch completes without incident. It
took
| >| >| me a while to figure this out, but I still don't understand why it
| >| >| happens.
| >| >|
| >| >| - Franc Zabkar
| >|
| >| > Did you setup or run these batches with standard, or background, or
| >| >modified priority settings?
| >| > Did you setup or run these batches with automatic pause, or other
| >| >re-assignment, should the system and/or applications require more
memory
| >or
| >| >priority?
| >| > Did you setup or run these batches with exclusive, standard, or
modified
| >| >memory settings?
| >| > Are you allowing or using or attempting any visual aspects within or
| >| >related to running these batches [e.g., DOS box, monitoring, or
| >otherwise]?
| >|
| >| I'm just running them in a Windows DOS box with nothing else running
| >| in the background. No special settings.
| >
| > Well I would attempt to setup something else, though that's likely hit
or
| >miss... then again you're really only testing for your intended use >
DivX
| >and 98, will they play well together with this large disk...
|
| I've copied a bunch of test AVI files to the drive and had them
| playing just fine via the USB 1.1 port on my $40 supermarket DVD
| player.
|
| My next step is to determine whether the DVD player will recognise
| more than one partition, and whether it has any file limits, etc.
|
| Incidentally there were no hangs while I was playing these files
| through my DVD. Maybe it's a write issue.

I would think its more the attempts in these two hard drive discussions, to
use DOS... makes me question whether its a DOS test or Windows.... if it is
DOS, then your in an entirely OLD world ... really, what do you expect,
everything to run properly and fast??? There is a whole other set of
limitations and constraints in that area.

|
| - Franc Zabkar
| --
| Please remove one 'i' from my address when replying by email.

--
MEB
http://peoplescounsel.orgfree.com
________

 

[Franc Zabkar]

On Tue, 24 Jul 2007 11:59:40 -0400, "MEB" <meb@not here@hotmail.com>
put finger to keyboard and composed:

>| I used Win98SE's format.com and Microsoft's updated fdisk.exe.
>|
>| FDISK EXE 64,460 05-18-00 8:35a FDISK.EXE
>| FORMAT COM 49,575 04-23-99 10:22p FORMAT.COM
>
> For reference, you chose this method instead of the manufacturer's tool(s)
>for what reason or were there none available or recommended?

Correction, I actually formatted the drive from within Explorer, not
by using format.com.

Sorry.

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.

 

[Franc Zabkar]

On Wed, 1 Aug 2007 15:51:14 -0400, "MEB" <meb@not here@hotmail.com>
put finger to keyboard and composed:

>"Franc Zabkar" <fzabkar@iinternode.on.net> wrote in message
>news:lke0b3dtbgcrrgop072bls7fp0kvksfifa@4ax.com...
>| On Sun, 29 Jul 2007 01:37:07 -0400, "MEB" <meb@not here@hotmail.com>
>| put finger to keyboard and composed:
>|
>| >| The SiS 7001 / 7002 controllers are on the motherboard. The SiS 7001
>| >| is the 1.1 controller, and the SiS 7002 is 2.0.
>| >|
>| >| I have been struggling to get the drive to connect at USB 2.0. No
>| >| matter what I do, Usbview always shows that the device is attached to
>| >| the SiS 7001. :-(
>| >
>| > Is it possible to disable that 7001 in Device Manager forcing 2.0 [or
>| >remove the 7001]?
>|
>| The problem is that I don't know whether the chipset is switching
>| between USB controllers on its own, or whether the driver is doing it.
>
> Did you try something like filemon or regmon by sysinternals to observe
>what was going on?
> Potential registry settings?

I've started a couple of related threads in other groups.

External 320GB HD / USB 2.0 / DOS / LFN / FAT32:
http://groups.google.com/group/comp.os.msdos.misc/msg/2050739e50b8a233?hl=en&

PCI subdevice and subvendor IDs:
http://groups.google.com/group/comp.sys.ibm.pc.hardware.chips/msg/164ced5a47b020ba?hl=en&

What I've found is that, even when using the USB 2.0 and USB mass
storage support in BIOS, DOS can see the USB HDD as drive D:, but the
transfer speeds are still only USB 1.1 (~0.7MB/s).

The same enclosure, HDD, and cables work fine on an XP box with Intel
chipset. The transfer rate in that case is about 30MB/s. This leads me
to believe that there is an incompatibility at the hardware level.

I notice that enabling USB support in BIOS causes the drive to appear
as both drive D: and drive N: in Windows Explorer, if the system is
booted with the drive attached. Otherwise, if the HDD is attached
after booting, only drive letter N: appears. Maybe this observation
will explain some mysterious ghosting issues that appear from time to
time in this group.

>| Anyway, I disabled both USB 1.1 controllers in DM and got a blue
>| screen error (openhci.sys) when I launched usbview.exe. I launched it
>| a second time and this time I saw that the SiS 7001 entries were still
>| present but there were no data associated with them. The USB 2.0
>| controller showed up as usual but the HD was not attached to it.
>
> So you were apparently using the 1.1 aspect, killed that, BSODed, restarted
>the computer, and the 2.0 didn't kick in?
> Did you do a *Find New Harware* to correct registry, VMM, and other
>aspects?

No, but I believe the problem may be a hardware one, as described
above.

<snip>

>| Incidentally there were no hangs while I was playing these files
>| through my DVD. Maybe it's a write issue.
>
> I would think its more the attempts in these two hard drive discussions, to
>use DOS... makes me question whether its a DOS test or Windows.... if it is
>DOS, then your in an entirely OLD world ... really, what do you expect,
>everything to run properly and fast??? There is a whole other set of
>limitations and constraints in that area.

I ran the system for several hours in real DOS mode using just the USB
support provided by the BIOS. There were no stability issues.

In the absence of responses from the other threads, I intend to try
Panasonic's DOS USB 2.0 EHCI drivers (which are documented to support
SiS chipsets), 4dos.com (for LFN support), and a Win98SE boot
diskette. If this doesn't work, then I'll be fairly certain that I'm
seeing a hardware incompatibility issue.

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.

 

[MEB]

"Franc Zabkar" <fzabkar@iinternode.on.net> wrote in message
news:32m7b3pbom8tm728id3f4j8a7j7oka5me9@4ax.com...
| On Tue, 24 Jul 2007 11:59:40 -0400, "MEB" <meb@not here@hotmail.com>
| put finger to keyboard and composed:
|
| >| I used Win98SE's format.com and Microsoft's updated fdisk.exe.
| >|
| >| FDISK EXE 64,460 05-18-00 8:35a FDISK.EXE
| >| FORMAT COM 49,575 04-23-99 10:22p FORMAT.COM
| >
| > For reference, you chose this method instead of the manufacturer's
tool(s)
| >for what reason or were there none available or recommended?
|
| Correction, I actually formatted the drive from within Explorer, not
| by using format.com.
|
| Sorry.
|
| - Franc Zabkar
| --
| Please remove one 'i' from my address when replying by email.

Thanks for the correction, got to keep this all correct for complete
analysis.. CRUNCH

--
MEB
http://peoplescounsel.orgfree.com
________

 

[MEB]

See the bottom, though refer to the entire post, seems there's some sort of
communication barrier here ...

"Franc Zabkar" <fzabkar@iinternode.on.net> wrote in message
news:33m7b3d85g49n3av1pmmdaqu1njcbva8m1@4ax.com...
| On Wed, 1 Aug 2007 15:51:14 -0400, "MEB" <meb@not here@hotmail.com>
| put finger to keyboard and composed:
|
| >"Franc Zabkar" <fzabkar@iinternode.on.net> wrote in message
| >news:lke0b3dtbgcrrgop072bls7fp0kvksfifa@4ax.com...
| >| On Sun, 29 Jul 2007 01:37:07 -0400, "MEB" <meb@not here@hotmail.com>
| >| put finger to keyboard and composed:
| >|
| >| >| The SiS 7001 / 7002 controllers are on the motherboard. The SiS 7001
| >| >| is the 1.1 controller, and the SiS 7002 is 2.0.
| >| >|
| >| >| I have been struggling to get the drive to connect at USB 2.0. No
| >| >| matter what I do, Usbview always shows that the device is attached
to
| >| >| the SiS 7001. :-(
| >| >
| >| > Is it possible to disable that 7001 in Device Manager forcing 2.0 [or
| >| >remove the 7001]?
| >|
| >| The problem is that I don't know whether the chipset is switching
| >| between USB controllers on its own, or whether the driver is doing it.
| >
| > Did you try something like filemon or regmon by sysinternals to observe
| >what was going on?
| > Potential registry settings?
|
| I've started a couple of related threads in other groups.
|
| External 320GB HD / USB 2.0 / DOS / LFN / FAT32:
|
http://groups.google.com/group/comp.os.msdos.misc/msg/2050739e50b8a233?hl=en&
|
| PCI subdevice and subvendor IDs:
|
http://groups.google.com/group/comp.sys.ibm.pc.hardware.chips/msg/164ced5a47b020ba?hl=en&
|
| What I've found is that, even when using the USB 2.0 and USB mass
| storage support in BIOS, DOS can see the USB HDD as drive D:, but the
| transfer speeds are still only USB 1.1 (~0.7MB/s).
|
| The same enclosure, HDD, and cables work fine on an XP box with Intel
| chipset. The transfer rate in that case is about 30MB/s. This leads me
| to believe that there is an incompatibility at the hardware level.
|
| I notice that enabling USB support in BIOS causes the drive to appear
| as both drive D: and drive N: in Windows Explorer, if the system is
| booted with the drive attached. Otherwise, if the HDD is attached
| after booting, only drive letter N: appears. Maybe this observation
| will explain some mysterious ghosting issues that appear from time to
| time in this group.
|
| >| Anyway, I disabled both USB 1.1 controllers in DM and got a blue
| >| screen error (openhci.sys) when I launched usbview.exe. I launched it
| >| a second time and this time I saw that the SiS 7001 entries were still
| >| present but there were no data associated with them. The USB 2.0
| >| controller showed up as usual but the HD was not attached to it.
| >
| > So you were apparently using the 1.1 aspect, killed that, BSODed,
restarted
| >the computer, and the 2.0 didn't kick in?
| > Did you do a *Find New Harware* to correct registry, VMM, and other
| >aspects?
|
| No, but I believe the problem may be a hardware one, as described
| above.
|
| <snip>
|
| >| Incidentally there were no hangs while I was playing these files
| >| through my DVD. Maybe it's a write issue.
| >
| > I would think its more the attempts in these two hard drive discussions,
to
| >use DOS... makes me question whether its a DOS test or Windows.... if it
is
| >DOS, then your in an entirely OLD world ... really, what do you expect,
| >everything to run properly and fast??? There is a whole other set of
| >limitations and constraints in that area.
|
| I ran the system for several hours in real DOS mode using just the USB
| support provided by the BIOS. There were no stability issues.
|
| In the absence of responses from the other threads, I intend to try
| Panasonic's DOS USB 2.0 EHCI drivers (which are documented to support
| SiS chipsets), 4dos.com (for LFN support), and a Win98SE boot
| diskette. If this doesn't work, then I'll be fairly certain that I'm
| seeing a hardware incompatibility issue.
|
| - Franc Zabkar
| --
| Please remove one 'i' from my address when replying by email.

Okay, so YOU state the barrier in DOS is the restriction to 1.1 USB, keep
that in mind.

Next you purport that there are no stability problems using pure DOS and
this 1.1 aspect.

HOWEVER< when using a DOS box in Windows [which has already loaded a virtual
mode driver for USB, perhaps the wrong one] you are having difficulties.

So what your saying is [correct me if I got this wrong] that you think you
should be able to use pure BIOS support IN A WINDOWS BOX without having any
residual or conflicting effect from:

1. Using a DOS box in this fashion.

2. Failing to run the batches/scripts without any special settings for
memory, Window aspects, and/or other WHILE attempting to use either the
DOS/BIOS support, or a driver which does not supply the support you need.

3. While using an Orangeware driver which does not support your device, OR
while using a Panasonic driver and an older generic support driver.

And this AFTER:

1. I suggested at least using Visual Basic so you could address virtual
mode, memory, Window aspects, AND of course proper driver/device usage?

2. showing that there was another, apparently better generic support driver
available which MAY provide what you need for full Windows usage,
potentially negating the need for another driver?

Did I get this correct?

--
MEB
http://peoplescounsel.orgfree.com
________

 

[Franc Zabkar]

On Sat, 4 Aug 2007 00:26:33 -0400, "MEB" <meb@not here@hotmail.com>
put finger to keyboard and composed:

> See the bottom, though refer to the entire post, seems there's some sort of
>communication barrier here ...
>
>"Franc Zabkar" <fzabkar@iinternode.on.net> wrote in message
>news:33m7b3d85g49n3av1pmmdaqu1njcbva8m1@4ax.com...
>| On Wed, 1 Aug 2007 15:51:14 -0400, "MEB" <meb@not here@hotmail.com>
>| put finger to keyboard and composed:
>|
>| >"Franc Zabkar" <fzabkar@iinternode.on.net> wrote in message
>| >news:lke0b3dtbgcrrgop072bls7fp0kvksfifa@4ax.com...
>| >| On Sun, 29 Jul 2007 01:37:07 -0400, "MEB" <meb@not here@hotmail.com>
>| >| put finger to keyboard and composed:
>| >|
>| >| >| The SiS 7001 / 7002 controllers are on the motherboard. The SiS 7001
>| >| >| is the 1.1 controller, and the SiS 7002 is 2.0.
>| >| >|
>| >| >| I have been struggling to get the drive to connect at USB 2.0. No
>| >| >| matter what I do, Usbview always shows that the device is attached
>to
>| >| >| the SiS 7001. :-(
>| >| >
>| >| > Is it possible to disable that 7001 in Device Manager forcing 2.0 [or
>| >| >remove the 7001]?
>| >|
>| >| The problem is that I don't know whether the chipset is switching
>| >| between USB controllers on its own, or whether the driver is doing it.
>| >
>| > Did you try something like filemon or regmon by sysinternals to observe
>| >what was going on?
>| > Potential registry settings?
>|
>| I've started a couple of related threads in other groups.
>|
>| External 320GB HD / USB 2.0 / DOS / LFN / FAT32:
>|
>http://groups.google.com/group/comp.os.msdos.misc/msg/2050739e50b8a233?hl=en&
>|
>| PCI subdevice and subvendor IDs:
>|
>http://groups.google.com/group/comp.sys.ibm.pc.hardware.chips/msg/164ced5a47b020ba?hl=en&
>|
>| What I've found is that, even when using the USB 2.0 and USB mass
>| storage support in BIOS, DOS can see the USB HDD as drive D:, but the
>| transfer speeds are still only USB 1.1 (~0.7MB/s).
>|
>| The same enclosure, HDD, and cables work fine on an XP box with Intel
>| chipset. The transfer rate in that case is about 30MB/s. This leads me
>| to believe that there is an incompatibility at the hardware level.
>|
>| I notice that enabling USB support in BIOS causes the drive to appear
>| as both drive D: and drive N: in Windows Explorer, if the system is
>| booted with the drive attached. Otherwise, if the HDD is attached
>| after booting, only drive letter N: appears. Maybe this observation
>| will explain some mysterious ghosting issues that appear from time to
>| time in this group.
>|
>| >| Anyway, I disabled both USB 1.1 controllers in DM and got a blue
>| >| screen error (openhci.sys) when I launched usbview.exe. I launched it
>| >| a second time and this time I saw that the SiS 7001 entries were still
>| >| present but there were no data associated with them. The USB 2.0
>| >| controller showed up as usual but the HD was not attached to it.
>| >
>| > So you were apparently using the 1.1 aspect, killed that, BSODed,
>restarted
>| >the computer, and the 2.0 didn't kick in?
>| > Did you do a *Find New Harware* to correct registry, VMM, and other
>| >aspects?
>|
>| No, but I believe the problem may be a hardware one, as described
>| above.
>|
>| <snip>
>|
>| >| Incidentally there were no hangs while I was playing these files
>| >| through my DVD. Maybe it's a write issue.
>| >
>| > I would think its more the attempts in these two hard drive discussions,
>to
>| >use DOS... makes me question whether its a DOS test or Windows.... if it
>is
>| >DOS, then your in an entirely OLD world ... really, what do you expect,
>| >everything to run properly and fast??? There is a whole other set of
>| >limitations and constraints in that area.
>|
>| I ran the system for several hours in real DOS mode using just the USB
>| support provided by the BIOS. There were no stability issues.
>|
>| In the absence of responses from the other threads, I intend to try
>| Panasonic's DOS USB 2.0 EHCI drivers (which are documented to support
>| SiS chipsets), 4dos.com (for LFN support), and a Win98SE boot
>| diskette. If this doesn't work, then I'll be fairly certain that I'm
>| seeing a hardware incompatibility issue.
>|
>| - Franc Zabkar
>| --
>| Please remove one 'i' from my address when replying by email.
>
> Okay, so YOU state the barrier in DOS is the restriction to 1.1 USB, keep
>that in mind.
>
> Next you purport that there are no stability problems using pure DOS and
>this 1.1 aspect.
>
>HOWEVER< when using a DOS box in Windows [which has already loaded a virtual
>mode driver for USB, perhaps the wrong one] you are having difficulties.
>
> So what your saying is [correct me if I got this wrong] that you think you
>should be able to use pure BIOS support IN A WINDOWS BOX without having any
>residual or conflicting effect from:
>
>1. Using a DOS box in this fashion.
>
>2. Failing to run the batches/scripts without any special settings for
>memory, Window aspects, and/or other WHILE attempting to use either the
>DOS/BIOS support, or a driver which does not supply the support you need.
>
>3. While using an Orangeware driver which does not support your device, OR
>while using a Panasonic driver and an older generic support driver.
>
> And this AFTER:
>
>1. I suggested at least using Visual Basic so you could address virtual
>mode, memory, Window aspects, AND of course proper driver/device usage?
>
>2. showing that there was another, apparently better generic support driver
>available which MAY provide what you need for full Windows usage,
>potentially negating the need for another driver?
>
> Did I get this correct?

Sorry for the confusion, but I think I'm getting closer to a solution.

Anyway, I *eventually* found a USB 2.0 driver (USBASPI.SYS, version
2.15) that works in DOS. It is available in this archive:

http://panasonic.co.jp/pcc/products/drive/other/driver/f2h_usb.exe

I tried several others, including an earlier version (2.06), but they
would either not see the SiS 700x controllers at all, or would only
detect and install the SiS 7001.

The HD driver that I used is DI1000DD.SYS which is available here:

http://www.stefan2000.com/darkehorse/PC/DOS/Drivers/USB/mhairu.zip

The AMI BIOS has native support for USB 1.1 but not USB 2.0.

To isolate the hanging issue, I used a 2GB flash drive and booted to
DOS using the above drivers. I then wrote about 1.9GB in 4 files at a
speed of 1.2 MB/s. Next I copied the files back to drive C: and
verified them. The read speed was 9.5 MB/s, and the system was stable
throughout.

I then tried to repeat this test with the 320GB HD. I was able to
reliably retrieve several gigabytes of data in more than 2000 files at
a speed of 20.6 MB/s. However, whenever I tried to write a large
amount of data to the external drive, either in a large number of
small files (~3MB), or in a small number of large files (~1GB), the
machine *always* hung. So it looks like the random hanging in Windows
is not an Explorer issue, but most likely a hardware one, especially
since it also happens on my socket 7 box. Both boxes use SiS 7001 OHCI
controllers.

My next step is to thrash this HD on my friend's XP/Intel PC by
filling it full of data. Depending on the outcome, I'll next hook up
the drive to my IDE port and run Seagate's diagnostics through it.

During my testing I discovered that the 2GB flash drive is detected by
the BIOS as an external floppy drive and subsequently by DOS as drive
B:. However, the way its file structure is seen by DOS depends on how
it is partitioned and/or formatted.

If I enable USB legacy mouse/keyboard/floppy support, and if the flash
drive has no partition, only a boot sector, then it is detected as a
floppy drive. In this case it takes drive letter B:, but only in real
DOS. A directory listing produces sensible output.

In Windows Explorer the flash drive is detected as a HD and is
assigned the next available drive letter, eg D:. However, drive B: is
not listed. If I now drop back into a Windows DOS box and attempt to
access drive B: (eg dir b, then the machine locks up and requires a
reboot.

Alternatively, if the flash drive is partitioned with Fdisk, then DOS
outputs gibberish in response to dir b:. I suspect this is because DOS
tries to interpret the MBR as a boot sector and then has trouble
finding the root directory.

FWIW, I found this thread which mentions the chip in my USB enclosure
(JMicron, JM20337):
http://forums.vr-zone.com/showthread.php?t=143880

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.

 

[MEB]

"Franc Zabkar" <fzabkar@iinternode.on.net> wrote in message
news9vnb3l1ucqasfm9ejr3575uvogffbbuin@4ax.com...
| On Sat, 4 Aug 2007 00:26:33 -0400, "MEB" <meb@not here@hotmail.com>
| put finger to keyboard and composed:
| >"Franc Zabkar" <fzabkar@iinternode.on.net> wrote in message
| >news:33m7b3d85g49n3av1pmmdaqu1njcbva8m1@4ax.com...
| >| On Wed, 1 Aug 2007 15:51:14 -0400, "MEB" <meb@not here@hotmail.com>
| >| put finger to keyboard and composed:
| >| >"Franc Zabkar" <fzabkar@iinternode.on.net> wrote in message
| >| >news:lke0b3dtbgcrrgop072bls7fp0kvksfifa@4ax.com...
| >| >| On Sun, 29 Jul 2007 01:37:07 -0400, "MEB" <meb@not here@hotmail.com>
| >| >| put finger to keyboard and composed:
| >| >|
| >| >| >| The SiS 7001 / 7002 controllers are on the motherboard. The SiS
7001
| >| >| >| is the 1.1 controller, and the SiS 7002 is 2.0.
| >| >| >|

[deleted material]

| >| What I've found is that, even when using the USB 2.0 and USB mass
| >| storage support in BIOS, DOS can see the USB HDD as drive D:, but the
| >| transfer speeds are still only USB 1.1 (~0.7MB/s).
| >|
| >| The same enclosure, HDD, and cables work fine on an XP box with Intel
| >| chipset. The transfer rate in that case is about 30MB/s. This leads me
| >| to believe that there is an incompatibility at the hardware level.
| >|
| >| I notice that enabling USB support in BIOS causes the drive to appear
| >| as both drive D: and drive N: in Windows Explorer, if the system is
| >| booted with the drive attached. Otherwise, if the HDD is attached
| >| after booting, only drive letter N: appears. Maybe this observation
| >| will explain some mysterious ghosting issues that appear from time to
| >| time in this group.
| >|
| >| >| Anyway, I disabled both USB 1.1 controllers in DM and got a blue
| >| >| screen error (openhci.sys) when I launched usbview.exe. I launched
it
| >| >| a second time and this time I saw that the SiS 7001 entries were
still
| >| >| present but there were no data associated with them. The USB 2.0
| >| >| controller showed up as usual but the HD was not attached to it.

| >| <snip>
| >|
| >| >| Incidentally there were no hangs while I was playing these files
| >| >| through my DVD. Maybe it's a write issue.
| >| >
| >| > I would think its more the attempts in these two hard drive
discussions,
| >to
| >| >use DOS... makes me question whether its a DOS test or Windows.... if
it
| >is
| >| >DOS, then your in an entirely OLD world ... really, what do you
expect,
| >| >everything to run properly and fast??? There is a whole other set of
| >| >limitations and constraints in that area.
| >|
| >| I ran the system for several hours in real DOS mode using just the USB
| >| support provided by the BIOS. There were no stability issues.
| >|
| >| In the absence of responses from the other threads, I intend to try
| >| Panasonic's DOS USB 2.0 EHCI drivers (which are documented to support
| >| SiS chipsets), 4dos.com (for LFN support), and a Win98SE boot
| >| diskette. If this doesn't work, then I'll be fairly certain that I'm
| >| seeing a hardware incompatibility issue.
| >|
| >| - Franc Zabkar
| >| --
| >| Please remove one 'i' from my address when replying by email.
| >

[deleted questions now clarified, at least for DOS aspects, WINDOWS issues
still in limbo]

|
| Sorry for the confusion, but I think I'm getting closer to a solution.

Ah, so you were concerned with DOS aspects.. boy you ARE a die hard DOS guy
....

|
| Anyway, I *eventually* found a USB 2.0 driver (USBASPI.SYS, version
| 2.15) that works in DOS. It is available in this archive:
|
| http://panasonic.co.jp/pcc/products/drive/other/driver/f2h_usb.exe

I had tried an older Panasonic driver, with various success... I will
download this one JUST IN CASE..

|
| I tried several others, including an earlier version (2.06), but they
| would either not see the SiS 700x controllers at all, or would only
| detect and install the SiS 7001.

Leaving just 1.1, gotcha..

|
| The HD driver that I used is DI1000DD.SYS which is available here:
|
| http://www.stefan2000.com/darkehorse/PC/DOS/Drivers/USB/mhairu.zip
|
| The AMI BIOS has native support for USB 1.1 but not USB 2.0.

So it was the *BIOS* that was limiting to 1.1, okay

|
| To isolate the hanging issue, I used a 2GB flash drive and booted to
| DOS using the above drivers. I then wrote about 1.9GB in 4 files at a
| speed of 1.2 MB/s. Next I copied the files back to drive C: and
| verified them. The read speed was 9.5 MB/s, and the system was stable
| throughout.

Respectable speed

|
| I then tried to repeat this test with the 320GB HD. I was able to
| reliably retrieve several gigabytes of data in more than 2000 files at
| a speed of 20.6 MB/s. However, whenever I tried to write a large
| amount of data to the external drive, either in a large number of
| small files (~3MB), or in a small number of large files (~1GB), the
| machine *always* hung. So it looks like the random hanging in Windows
| is not an Explorer issue, but most likely a hardware one, especially
| since it also happens on my socket 7 box. Both boxes use SiS 7001 OHCI
| controllers.

Soooo, it does look like the SIS 7001 MAY be the culprit

|
| My next step is to thrash this HD on my friend's XP/Intel PC by
| filling it full of data. Depending on the outcome, I'll next hook up
| the drive to my IDE port and run Seagate's diagnostics through it.

AH, not those old diag tools we discussed in other hard drive discussions
before, I hope, kinda pushing that doncha think..

Are you going to look at the disk [first few tracks] for XP residuals?

|
| During my testing I discovered that the 2GB flash drive is detected by
| the BIOS as an external floppy drive and subsequently by DOS as drive
| B:. However, the way its file structure is seen by DOS depends on how
| it is partitioned and/or formatted.
|
| If I enable USB legacy mouse/keyboard/floppy support, and if the flash
| drive has no partition, only a boot sector, then it is detected as a
| floppy drive. In this case it takes drive letter B:, but only in real
| DOS. A directory listing produces sensible output.

Uhm, okay, so was the Panasonic driver loaded?

|
| In Windows Explorer the flash drive is detected as a HD and is
| assigned the next available drive letter, eg D:. However, drive B: is
| not listed. If I now drop back into a Windows DOS box and attempt to
| access drive B: (eg dir b, then the machine locks up and requires a
| reboot.

Seems you've located another conflict/limitation between BIOS and OS. The
drive can not be BOTH a floppy AND a hard drive, with different assignments
[drive, memory, etc.]...

|
| Alternatively, if the flash drive is partitioned with FDisk, then DOS
| outputs gibberish in response to dir b:. I suspect this is because DOS
| tries to interpret the MBR as a boot sector and then has trouble
| finding the root directory.

Might it be a fat12 vs fat16 issue?

Most articles describe the format on flash drives as fat16 [1 and 2 gig
flash], so would it be that when a DOS box [is that what you used?] or pure
DOS/command prompt FDisk attempts to write fat12 data/mbr [BIOS routines
used - seen as floppy] which does not support 2 gig, using the Panasonic
driver you found, in the process? [Scratches head, huh, how the heck would
it do that,,, fdisk doesn't work on floppy drives,,, but .... between the
BIOS and the various drivers ???]
OR tries to write mbr preparing for fat32 which seems to mess up some flash
drives? {Obviously after formatting either would be noticeable.}

http://thestarman.pcministry.com/index.html
http://thestarman.pcministry.com/asm/mbr/index.html#MBR
http://thestarman.pcministry.com/asm/mbr/95BMEMBR.htm - fat32 MBR
http://thestarman.pcministry.com/asm/mbr/STDMBR.htm - standard/old MBR
http://thestarman.pcministry.com/asm/mbr/WIN98FDB.htm - 98 floppy FDBR
http://thestarman.pcministry.com/asm/mbr/DOS50FDB.htm - DOS 5.00 FDBR
http://thestarman.pcministry.com/asm/mbr/index.html#PT - partition tables

It would be interesting to see what it looked like with a disk/hex editor
after the attempts...

|
| FWIW, I found this thread which mentions the chip in my USB enclosure
| (JMicron, JM20337):
| http://forums.vr-zone.com/showthread.php?t=143880

Good, I'll look to see what's what....

|
| - Franc Zabkar
| --
| Please remove one 'i' from my address when replying by email.


--
MEB
http://peoplescounsel.orgfree.com
________

 

[Franc Zabkar]

On Fri, 10 Aug 2007 04:29:04 -0400, "MEB" <meb@not here@hotmail.com>
put finger to keyboard and composed:


>"Franc Zabkar" <fzabkar@iinternode.on.net> wrote in message
>news9vnb3l1ucqasfm9ejr3575uvogffbbuin@4ax.com...

> Ah, so you were concerned with DOS aspects.. boy you ARE a die hard DOS guy
>...

I really had no alternative. The Windows Orangeware drivers were
crippled, and I was experiencing hangs which appeared to be related to
Explorer. Dropping back to DOS reduced the number of variables.

The added benefit is that I now have a method of backing up a damaged
file system without opening up the box.

>|
>| Anyway, I *eventually* found a USB 2.0 driver (USBASPI.SYS, version
>| 2.15) that works in DOS. It is available in this archive:
>|
>| http://panasonic.co.jp/pcc/products/drive/other/driver/f2h_usb.exe
>
> I had tried an older Panasonic driver, with various success... I will
>download this one JUST IN CASE..

The 2.06 driver displays much more verbose connection information,
such as which device is connected to which controller. From that
standpoint it may be a better troubleshooting tool.

>| I tried several others, including an earlier version (2.06), but they
>| would either not see the SiS 700x controllers at all, or would only
>| detect and install the SiS 7001.
>
> Leaving just 1.1, gotcha..
>
>|
>| The HD driver that I used is DI1000DD.SYS which is available here:
>|
>| http://www.stefan2000.com/darkehorse/PC/DOS/Drivers/USB/mhairu.zip
>|
>| The AMI BIOS has native support for USB 1.1 but not USB 2.0.
>
> So it was the *BIOS* that was limiting to 1.1, okay

Yes, but you wouldn't expect this when you consider that there is an
entry in the BIOS setup which explicitly enables or disables USB 2.0
support. I now believe that this setting merely enables or disables
the SiS 7002 controller. It doesn't appear to enable or disable any
USB 2.0 *code*.

>| To isolate the hanging issue, I used a 2GB flash drive and booted to
>| DOS using the above drivers. I then wrote about 1.9GB in 4 files at a
>| speed of 1.2 MB/s. Next I copied the files back to drive C: and
>| verified them. The read speed was 9.5 MB/s, and the system was stable
>| throughout.
>
> Respectable speed
>
>|
>| I then tried to repeat this test with the 320GB HD. I was able to
>| reliably retrieve several gigabytes of data in more than 2000 files at
>| a speed of 20.6 MB/s. However, whenever I tried to write a large
>| amount of data to the external drive, either in a large number of
>| small files (~3MB), or in a small number of large files (~1GB), the
>| machine *always* hung. So it looks like the random hanging in Windows
>| is not an Explorer issue, but most likely a hardware one, especially
>| since it also happens on my socket 7 box. Both boxes use SiS 7001 OHCI
>| controllers.
>
> Soooo, it does look like the SIS 7001 MAY be the culprit

I now suspect that neither the SiS 7001 or 7002 controllers are
compatible with the JM20337 chip in the external box. For a while,
though, I thought that there may have been a gremlin in the BIOS.

I say this because version 2.06 of the Panasonic USBASPI driver would
produce this message (annotations added):

Device
| Function
Bus | |
| | |
Controller : 00-03-0 VID=1039h PID=7001h (1019h-1808h) OHCI
: MEM=CFFFD000h-CFFFDFFFh(4KBytes)

Controller : 00-03-1 VID=1039h PID=7001h (1019h-1808h) OHCI
: MEM=CFFFE000h-CFFFEFFFh(4KBytes)

ERROR : EHCI memory mapped I/O can not be assigned.

My suspicions were compounded when I actually dumped the ESCD table in
the flash BIOS chip (using Uniflash). I say this because there were no
references to an SiS 7001 device, only three instances of the SiS 7002
at bus 0, device 3, and functions 0,1,2. But this turned out to be a
red herring. To make matters worse, the Phoenix utility that I used
for my analysis
(http://www.users.on.net/~fzabkar/BIOSutil/Phoenix/nvram120.exe) had a
bug which parsed one of the devices (PCI bridge) incorrectly, and this
had me questioning the validity of the results until I confirmed them
with Uniflash.

>| My next step is to thrash this HD on my friend's XP/Intel PC by
>| filling it full of data. Depending on the outcome, I'll next hook up
>| the drive to my IDE port and run Seagate's diagnostics through it.
>
> AH, not those old diag tools we discussed in other hard drive discussions
>before, I hope, kinda pushing that doncha think..

Last night I wrote 300 1GB files to the drive from within a CMD window
on an Intel/XP box. The write speed was about 21MB/s and there were no
problems. A surface scan found no bad sectors.

> Are you going to look at the disk [first few tracks] for XP residuals?

See end of post.

>| During my testing I discovered that the 2GB flash drive is detected by
>| the BIOS as an external floppy drive and subsequently by DOS as drive
>| B:. However, the way its file structure is seen by DOS depends on how
>| it is partitioned and/or formatted.
>|
>| If I enable USB legacy mouse/keyboard/floppy support, and if the flash
>| drive has no partition, only a boot sector, then it is detected as a
>| floppy drive. In this case it takes drive letter B:, but only in real
>| DOS. A directory listing produces sensible output.
>
> Uhm, okay, so was the Panasonic driver loaded?

No, just plain DOS.

>| In Windows Explorer the flash drive is detected as a HD and is
>| assigned the next available drive letter, eg D:. However, drive B: is
>| not listed. If I now drop back into a Windows DOS box and attempt to
>| access drive B: (eg dir b, then the machine locks up and requires a
>| reboot.
>
> Seems you've located another conflict/limitation between BIOS and OS. The
>drive can not be BOTH a floppy AND a hard drive, with different assignments
>[drive, memory, etc.]...

>| Alternatively, if the flash drive is partitioned with FDisk, then DOS
>| outputs gibberish in response to dir b:. I suspect this is because DOS
>| tries to interpret the MBR as a boot sector and then has trouble
>| finding the root directory.
>
> Might it be a fat12 vs fat16 issue?

The flash drive was originally formatted to look like a 2GB FAT16
floppy diskette. By that I mean that there was no partition, only a
boot sector.

boot sector | FAT #1 | FAT #2 | Root | File
FAT 16 | | | directory | area


The flash drive has now been partitioned and formatted as a FAT32 HD.

Sector 0 | Track 0 | FAT32 | | | |
MBR | | Boot | FAT#1 | FAT#2 | Root | File
Partition | 63 sectors | Parameter | | | dir | area
table | | Block | | | |

In both cases the media descriptor byte is/was F8 (= hard disc)
whereas a real floppy diskette has a media descriptor of F0 (= unknown
media type) and a FAT12 file system.

I think what is happening is this. DOS thinks that the flash drive is
a floppy (because the BIOS says so), which causes it to misinterpret
the MBR code as a boot sector. DOS then uses the "boot sector" data to
locate what it thinks is the root directory, but instead ends up in
some strange part of the disk.

> Most articles describe the format on flash drives as fat16 [1 and 2 gig
>flash], so would it be that when a DOS box [is that what you used?] or pure
>DOS/command prompt FDisk attempts to write fat12 data/mbr [BIOS routines
>used - seen as floppy] which does not support 2 gig, using the Panasonic
>driver you found, in the process? [Scratches head, huh, how the heck would
>it do that,,, fdisk doesn't work on floppy drives,,, but .... between the
>BIOS and the various drivers ???]

See above.

> OR tries to write mbr preparing for fat32 which seems to mess up some flash
>drives? {Obviously after formatting either would be noticeable.}
>
>http://thestarman.pcministry.com/index.html
>http://thestarman.pcministry.com/asm/mbr/index.html#MBR
>http://thestarman.pcministry.com/asm/mbr/95BMEMBR.htm - fat32 MBR
>http://thestarman.pcministry.com/asm/mbr/STDMBR.htm - standard/old MBR
>http://thestarman.pcministry.com/asm/mbr/WIN98FDB.htm - 98 floppy FDBR
>http://thestarman.pcministry.com/asm/mbr/DOS50FDB.htm - DOS 5.00 FDBR
>http://thestarman.pcministry.com/asm/mbr/index.html#PT - partition tables
>
> It would be interesting to see what it looked like with a disk/hex editor
>after the attempts...

See http://www.users.on.net/~fzabkar/USB_HD/

2GB flash drive boot sector (FAT16):
http://www.users.on.net/~fzabkar/USB_HD/2GB_F_BOO_FAT16.BIN

2GB flash drive MBR text dump (FAT32):
http://www.users.on.net/~fzabkar/USB_HD/2GB_F_MBR.txt

2GB flash drive MBR binary image (FAT32):
http://www.users.on.net/~fzabkar/USB_HD/2GB_F_MBR_FAT32.bin

2GB flash drive track 0 binary image (FAT32):
http://www.users.on.net/~fzabkar/USB_HD/2GB_F_TRK0.bin

The above image contains the remnants of the original FAT16 FAT
tables. I used "fdisk /actok" to turn off integrity checking,
otherwise the sectors would normally be filled with zeroes (I think).

13GB hard disc track 0 (Paragon Boot Manager):
http://www.users.on.net/~fzabkar/USB_HD/13GB_HD_TRK0.bin

6GB hard disc track 0 (EZ-Drive HD overlay):
http://www.users.on.net/~fzabkar/USB_HD/6GB_HD_TRK0.bin

320GB hard disc MBR and track 0:
http://www.users.on.net/~fzabkar/USB_HD/320GB_MBR.BIN
http://www.users.on.net/~fzabkar/USB_HD/320GB_MBR.txt
http://www.users.on.net/~fzabkar/USB_HD/320GB_TRK0.bin

The above images were taken after the 320GB drive had been attached to
the XP machine. The root directory contains a Recycled folder and a
System Volume Information folder. And yes, it *is* annoying.

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.