[glocK_94]

PROBLEMCHYLD, on Nov 1 2009, 11:44 PM, said:

Is the NTFS for Win98 Winternals V2 any good?
I really don't expect to write anything to a NTFS drive but you never know.
The read-only may be good.

I've been using version 2 for a while now and it works pretty well. You can copy/read files perfectly. You can even empty the recycle bin, no error yet.
Unlike version 1.x, it accepts XP files as dependancies. However, I've only been able to run the included checkdisk with NT files.

[wsxedcrfv]

Quote

NEWS FLASH:
Microsoft introduced FAT64 (exFAT) (Extended File Allocation Table) file system for Windows XP/2003/Vista/2008/7 similar to FAT32/FAT32X, but without
the boot sector, cluster size, directory entries or file size limitations that plagued FAT32/FAT32X, designed for SSD + USB Flash drives and compatible with hard drives:

Cluster size limitation?
What exactly is that?

Quote

- largest hard/SSD/Flash disk/partition size 16 EB (ExaBytes) (theoretical limit);
- largest file size 16 EB (ExaBytes) (theoretical limit);
- largest cluster size 32 MB;

16 EB is hyperbole (even if true). I still don't trust any drives over 500 gb for reliability at this point.
And explain what exactly is the attraction of large cluster sizes - beyond 32 kb?
Why is the fallacy still being put forward that FAT32 volumes are limited to 2 to 4 million clusters? There is no such FAT32 limitation. I've run win-98 on volumes with 40 and even 120 million clusters (a 500 gb drive formatted as a single FAT32 partition with 120 million 4kb cluster-size).

Quote

- > 1000 files per directory;

Only 1,000? Doesn't fat32 allow for up to 64k files per directory - depending on the number of characters in the file name?

Quote

- Transaction-Safe FAT File System (TFAT) support;

What exactly is that?

Quote

Disadvantages:
- no file system journaling support.

That's not a disadvantage. With journaling, data already written to open files is usually wiped out back to the last checkpoint. Any time my NT4 server goes down, I can count on that day's IIS log file to be wiped out. Thanks for the journalling. And thanks for the extra wear and tear on the drive itself.

[MDGx]

wsxedcrfv, on Nov 21 2009, 06:55 PM, said:

Cluster size limitation?
What exactly is that?

...

I meant cluster size [smallest 4 KB -> largest 32 KB]:
http://technet.micro...y/cc768180.aspx

I've being collecting this info for years, and posted all here:
http://www.mdgx.com/secrets.htm#FAT32

Quote

Why is the fallacy still being put forward that FAT32 volumes are limited to 2 to 4 million clusters? There is no such FAT32 limitation. I've run win-98 on volumes with 40 and even 120 million clusters (a 500 gb drive formatted as a single FAT32 partition with 120 million 4kb cluster-size).

Source:
http://www.ntfs.com/ntfs_vs_fat.htm
You may want to take this up with the people @ NTFS.com.

Quote

- Transaction-Safe FAT File System (TFAT) support;

Source:
http://en.wikipedia.org/wiki/ExFAT
Please e-mail me if there is some eroneous info @ this Wikipedia page, so I can correct it asap.

Please e-mail me with any corrections, so I can update the web page.

Thanks for your time + concern.

Edit:
Updated my page with latest Wikipedia info:
http://en.wikipedia.org/wiki/ExFAT

[dencorso]

wsxedcrfv, on Nov 21 2009, 11:55 PM, said:

Why is the fallacy still being put forward that FAT32 volumes are limited to 2 to 4 million clusters? There is no such FAT32 limitation. I've run win-98 on volumes with 40 and even 120 million clusters (a 500 gb drive formatted as a single FAT32 partition with 120 million 4kb cluster-size).

True. FAT32 is very flexible and has wide limits. But going above 26 million clusters limits its usability because most of the available programs to give it maintenance weren't written to support so many clusters, as discussed elsewhere:

dencorso, on Oct 22 2009, 10:45 AM, said:

...and you'd surely have read this post and the ones it links to, so you wouldn't be surprised, as ScanDskW and Defrag (both from Win ME, provided one also uses DiskMaint.dll from Win ME) are known to work OK with up to slightly above 850 GB partitions (= 26,389,392 clusters of 32 kiB ).

[MDGx]

jaclaz, on Nov 18 2009, 08:39 AM, said:

MDGx, on Nov 18 2009, 04:09 PM, said:

But some1 with driver level programming skills who can afford to spend a lot of time developing + testing this can probably make a native DOS driver, or even adapt the XP one [EXFAT.SYS], or at least make a VXD driver compatible with 9x OSes. [?]

As often happens, OFF-TOPIC, but not much :
How much money do you think these guys are making nowadays from selling this?:
http://www.techsoftp...m/vxd/index.htm

at US$ 429 apiece?

http://www.techsoftp...m/vxd/order.htm

Quote

Single user license for VxDWriter 2.0 costs US$429.00

Maybe if they would decide to "let go" that software or provide a "complimentary copy" a C programmer willing to take the challenge could even be found.

jaclaz

Well, if the VXD model is not a viable solution [those guys sell their developer package waaaay too expensive, something I had no idea about because I am not a programmer], how about the SYS driver format [WDM, introduced in WinME + 2000]?
The complete 98 DDK is still available [and free]...
http://www.mdgx.com/add.htm#DDK
respectively here:
http://www.mdgx.com/spx/98DDK.RAR
BTW: WinXP + 2003 DDK packages include all that is needed for Win98/ME programming, so one needs XP or 2003 to work on 98 SE/ME drivers, in case 98DDK [run from within Win98/ME] is not an option.

About the native DOS driver model...
I know 2 DOS low-level/driver programmers who might be able [and hopefully willing] to do this, I just haven't had time to write them yet, and I was actually waiting for some positive feedback from you guys @ MSFN.
1. Jack Ellis [makes free native DOS drivers for the FreeDOS.org project]:
http://johnson.tmfc....dos/driver.html
2. Bret Johnson [makes free generic USB drivers for native DOS]:
http://www.bretjohnson.us/

[All I'm saying is that FAT64 should be implemented in 9x OSes if possible *and* feasible.
And I know it's possible, we just need to find some1 who agrees to do this for free.

Edit:
According to this Wikipedia page:
http://en.wikipedia.org/wiki/ExFAT
one needs a Microsoft license to distribute FAT64 software/drivers/etc.
I have not checked to see if this license is free or not.
If any1 knows more on this subject, please post here.
Thanks.

HTH

Edit #2:
Unfortunately MS charges fees for distribution, therefore not feasible nor affordable to develop independent exFAT driver.

[rloew]

dencorso, on Nov 22 2009, 11:21 AM, said:

wsxedcrfv, on Nov 21 2009, 11:55 PM, said:

Why is the fallacy still being put forward that FAT32 volumes are limited to 2 to 4 million clusters? There is no such FAT32 limitation. I've run win-98 on volumes with 40 and even 120 million clusters (a 500 gb drive formatted as a single FAT32 partition with 120 million 4kb cluster-size).

True. FAT32 is very flexible and has wide limits. But going above 26 million clusters limits its usability because most of the available programs to give it maintenance weren't written to support so many clusters, as discussed elsewhere:

dencorso, on Oct 22 2009, 10:45 AM, said:

...and you'd surely have read this post and the ones it links to, so you wouldn't be surprised, as ScanDskW and Defrag (both from Win ME, provided one also uses DiskMaint.dll from Win ME) are known to work OK with up to slightly above 850 GB partitions (= 26,389,392 clusters of 32 kiB ).

You can double these limits, for Windows SCANDISK only, by using 64KiB Clusters, but you cannot use such a partition as a boot drive or with some older 16-Bit Programs including DOS SCANDISK

128KiB Clusters will raise the limit to the full 2TiB using Patched Filesystem Drivers..

[wsxedcrfv]

wsxedcrfv, on Nov 21 2009, 11:55 PM, said:

Why is the fallacy still being put forward that FAT32 volumes are limited to 2 to 4 million clusters? There is no such FAT32 limitation. I've run win-98 on volumes with 40 and even 120 million clusters (a 500 gb drive formatted as a single FAT32 partition with 120 million 4kb cluster-size).

dencorso, on Nov 22 2009, 09:21 AM, said:

True. FAT32 is very flexible and has wide limits. But going above 26 million clusters limits its usability because most of the available programs to give it maintenance weren't written to support so many clusters, as discussed elsewhere:

Most people don't talk about these issues with regard to win-9x. The idea that FAT32 has cluster-size or cluster-count limitations with regard to drive maintenance tools don't really apply to the typical use-case where the OS is XP or higher. I've installed and run XP on FAT32 volumes where I manually pre-formatted the drive (250 gb) with third-party tools such that the volume was 250 gb, 4kb cluster size. XP had no problems being installed on and running from such a volume, nor maintaining it.

[wsxedcrfv]

MDGx, on Nov 22 2009, 09:24 AM, said:

All I'm saying is that FAT64 should be implemented in 9x OSes if possible *and* feasible.

I don't understand why, given my experience with large drives and the use of non-standard (but perfectly valid) cluster size. What can you do with win-9x and FAT64 that you couldn't do FAT32? I would NEVER use a 64kb cluster size anyways for a volume (especially c:) and I rarely use 32 kb as it is.

What I'd like to see is someone hack format.com to remove the /z: handicap.

The /z: parameter is supposed to let you specify a cluster size (in increments of 512 bytes). But format checks the values and won't let you enter a non-standard parameter anyways, which makes the /z switch completely useless.

[jaclaz]

MDGx, on Nov 22 2009, 04:24 PM, said:

According to this Wikipedia page:
http://en.wikipedia.org/wiki/ExFAT
one needs a Microsoft license to distribute FAT64 software/drivers/etc.
I have not checked to see if this license is free or not.
If any1 knows more on this subject, please post here.

The "source" for that Wikipedia is this page (that doesn't work in Opera) :
http://www.microsoft.com/iplicensing/produ...nsing%20Program

compare it with the "known" FAT one:
http://www.microsoft.com/iplicensing/produ...20File%20System

Try searching for "Any" in category "Filesystem":

Quote

Showing 1 results for your search criteria.

FAT File System
.....

The good guys at MS have "queer" ways....

I will cite this from the benefits of FAT:

Quote

Title
FAT File System

Benefits

• The FAT File System provides a way to time stamp when a file is created or changed, as well as to identify the size of the file.
  
• This system provides a mechanism to store other attributes of a file, such as whether a file is read-only, whether the file should be hidden in a directory display, or whether a file should be archived during the next disk backup.
  
• The FAT File System is ideal for removable flash media used in consumer electronic devices, such as digital cameras, media players, flash drives, and so on.

Definitely CP/M filesystem appears a bit dated when we see the first two "benefits" :whistling:

Remember to take your time reading "Resources" and expecially "FAQ's" ....

Isn't it unusual that you have a FAQ section but apparently nowhere to submit a question?



jaclaz

[BenoitRen]

"rloew" said:

There are very few people creating quality free products without the other considerations I listed, so #1 still applies.

You must have missed the entire free and open-source community.

[PROBLEMCHYLD]

Which files should I use for NTFS for Win98, NT4, Win2000, WinXP SP1/2/3

[submix8c]

Use the XP ones. I have used both the XP and Server2k3 ones. Read the instructions carefully or it won't work. I have also "started" the "service" at Windows Startup. Can't remember exactly how I had it all set up (used the previous version).

[MDGx]

jaclaz, on Nov 22 2009, 11:55 AM, said:

The "source" for that Wikipedia is this page (that doesn't work in Opera) :
http://www.microsoft.com/iplicensing/produ...nsing%20Program

BTW, this page doesn't function properly with Firefox 3.5.5 either [links in tabs don't work]. Seems it requires MS IE. Another "brilliant" implementation of MS proprietary "standards"?

Thanks for pointing it out.
Unfortunately MS fee [they don't post the actual amount of money, but I'm sure it's a hefty one] are the show stopper.
And looks like MS will make a lot of money from this license, because soon SDXC memory cards/sticks will need exFAT support:
http://en.wikipedia....T#Disadvantages

[jaclaz]

MDGx, on Nov 23 2009, 07:29 PM, said:

Unfortunately MS fee [they don't post the actual amount of money, but I'm sure it's a hefty one] are the show stopper.

Well, not really.

Just like the known FAT issues, it is still to be proved AT LEAST:
a. that the technology is licensable
b. that writing a driver for interchange with other systems is infringing any patent, copyright or any other Law, and it is NOT covered by the "fair use" provisions

Microsoft licensing (and eventual lawsuits) are NOT aimed to the developers of a driver, but rather on large corporations making use of the filesystem on their hardware devices.

AFAIK, Microsoft patents on FAT have been confirmed in the U.S. and rejected in Europe.

jaclaz

This post has been edited by jaclaz: 23 November 2009 - 02:38 PM

[MDGx]

jaclaz, on Nov 23 2009, 01:37 PM, said:

Just like the known FAT issues, it is still to be proved AT LEAST:
a. that the technology is licensable
b. that writing a driver for interchange with other systems is infringing any patent, copyright or any other Law, and it is NOT covered by the "fair use" provisions

Microsoft licensing (and eventual lawsuits) are NOT aimed to the developers of a driver, but rather on large corporations making use of the filesystem on their hardware devices.

AFAIK, Microsoft patents on FAT have been confirmed in the U.S. and rejected in Europe.

jaclaz

If that means we have [a small] chance some1 would like to volunteer to develop exFAT drivers, that would be awesome!
But if that requires to pay for licensing, and implicitly charge for the drivers, thanks, but no thanks.

Because a boatload of new devices will be released in the near future, I'm hoping:
a. Either MS will relax the licensing issue, and *if* that happens, maybe a Linux developer can create a nix driver. In that case, I'm assuming it should be fairly easy to port such *nix driver to native DOS environment or even Win9x.
b. Better, maybe some1 can develop their own exFAT driver [outside US] without having to go thru the MS license/patent process, and therefore that driver should be distributed for free [GPL?].
So... any volunteers?
Just my 2 ¢ ...

[rant]
Patent on FAT? Sure, why not? Let's also patent the air we breathe.
[/rant]

[PROBLEMCHYLD]

submix8c, on Nov 22 2009, 04:41 PM, said:

Use the XP ones. I have used both the XP and Server2k3 ones. Read the instructions carefully or it won't work. I have also "started" the "service" at Windows Startup. Can't remember exactly how I had it all set up (used the previous version).

Can you specify from which service pack please. WinXP SP1 SP2 SP3.Thanks

[wsxedcrfv]

MDGx, on Nov 26 2009, 11:46 PM, said:

If that means we have [a small] chance some1 would like to volunteer to develop exFAT drivers, that would be awesome! ;)

Please explain the practical benefits of exFAT over FAT32 with regard to Win-98.

Will Win-98's (or win-ME's) native drive maintenance tools function on exFAT-formatted volumes?

[jaclaz]

wsxedcrfv, on Nov 28 2009, 02:33 PM, said:

Please explain the practical benefits of exFAT over FAT32 with regard to Win-98.

As I see it the only advantage would be that of having files bigger than 4 Gb:
http://www.ntfs.com/ntfs_vs_fat.htm

With all due respect for the people asking for NTFS support, the only practical use of NTFS on Windows 9x is IMHO for easily exchange of files in NT/2k/Xp/2003/Vista/2008/7 multi-boot systems or as a quick recovery system, as it seems to me that all the "advanced" features of NTFS (permissions, quotas, etc.) are pretty much useless in a 9x system which is not intneded/used as "multi-user" or server.

wsxedcrfv, on Nov 28 2009, 02:33 PM, said:

Will Win-98's (or win-ME's) native drive maintenance tools function on exFAT-formatted volumes?

They definitely won't, but one could write new ones as well.

jaclaz

[TmEE]

Regarding Paragon NTFS for Win98, it is now free, though it requires registration for downloading. I've been using it for quite a while and have had no issues, but I've kept away from 4GB+ files aswell...

If I had more time and looked into windows programming, and knew C/C++, I'd be more than häppy to help with some file system driver. ASM + DOS works for me though, but my x86 is not half as good as my Z80 and 680x0 are

[jaclaz]

TmEE, on 20 September 2010 - 01:22 AM, said:

Regarding Paragon NTFS for Win98, it is now free, though it requires registration for downloading.

Really?
Should someone post it under "News"?
http://www.msfn.org/...ic=110814&st=20



jaclaz